From cb14ca0e561d32a8c570bd15e7cdf1eca81a3061 Mon Sep 17 00:00:00 2001 From: mhjensen Date: Thu, 26 Jul 2018 19:11:08 +0200 Subject: [PATCH] missing update --- doc/pub/BM/html/BM-bs.html | 1639 ++++++ doc/pub/BM/html/BM-reveal.html | 1681 ++++++ doc/pub/BM/html/BM-solarized.html | 1527 +++++ doc/pub/BM/html/BM.html | 1532 +++++ doc/pub/BM/html/figures/Hudson_Bay_data.pdf | Bin 0 -> 15045 bytes doc/pub/BM/html/figures/Hudson_Bay_data.png | Bin 0 -> 53586 bytes doc/pub/BM/html/figures/MLrbm.pdf | Bin 0 -> 11007 bytes doc/pub/BM/html/figures/MLrbm.png | Bin 0 -> 11333 bytes doc/pub/BM/html/figures/RBM.pdf | 1623 +++++ doc/pub/BM/html/figures/RBM.png | Bin 0 -> 38001 bytes doc/pub/BM/html/figures/RMBenergy.dat | 100 + doc/pub/BM/html/figures/RMBenergy.dat~ | 100 + doc/pub/BM/html/figures/dnn.png | Bin 0 -> 274267 bytes doc/pub/BM/html/figures/figN2.pdf | Bin 0 -> 22919 bytes doc/pub/BM/html/figures/figN2.png | Bin 0 -> 20041 bytes doc/pub/BM/html/figures/figN6.pdf | Bin 0 -> 23104 bytes doc/pub/BM/html/figures/figN6.png | Bin 0 -> 20344 bytes doc/pub/BM/html/figures/perceptron.png | Bin 0 -> 61405 bytes doc/pub/BM/html/figures/plot.py | 14 + doc/pub/BM/html/figures/plot.py~ | 14 + doc/pub/BM/html/figures/plotEnergies.py | 61 + doc/pub/BM/html/figures/plotEnergies.py~ | 62 + .../BM/html/figures/w1.0_D2_N2_Energies.txt | 177 + .../BM/html/figures/w1.0_D2_N6_Energies.txt | 164 + doc/pub/BM/html/reveal.js/.gitignore | 13 + doc/pub/BM/html/reveal.js/.travis.yml | 7 + doc/pub/BM/html/reveal.js/CONTRIBUTING.md | 23 + doc/pub/BM/html/reveal.js/Gruntfile.js | 140 + doc/pub/BM/html/reveal.js/LICENSE | 19 + doc/pub/BM/html/reveal.js/README.md | 1246 ++++ doc/pub/BM/html/reveal.js/bower.json | 27 + .../html/reveal.js/css/images/cbc_footer.png | Bin 0 -> 10008 bytes .../html/reveal.js/css/images/cbc_symbol.png | Bin 0 -> 2946 bytes .../reveal.js/css/images/simula_footer.png | Bin 0 -> 2513 bytes .../html/reveal.js/css/images/simula_logo.png | Bin 0 -> 2138 bytes .../reveal.js/css/images/simula_symbol.png | Bin 0 -> 2138 bytes .../html/reveal.js/css/images/uio_footer.png | Bin 0 -> 18189 bytes .../html/reveal.js/css/images/uio_symbol.png | Bin 0 -> 11352 bytes doc/pub/BM/html/reveal.js/css/print/paper.css | 203 + doc/pub/BM/html/reveal.js/css/print/pdf.css | 178 + doc/pub/BM/html/reveal.js/css/reveal.css | 1886 ++++++ doc/pub/BM/html/reveal.js/css/reveal.scss | 1717 ++++++ doc/pub/BM/html/reveal.js/css/theme/README.md | 21 + doc/pub/BM/html/reveal.js/css/theme/beige.css | 154 + .../html/reveal.js/css/theme/beigesmall.css | 155 + doc/pub/BM/html/reveal.js/css/theme/black.css | 273 + doc/pub/BM/html/reveal.js/css/theme/blood.css | 180 + doc/pub/BM/html/reveal.js/css/theme/cbc.css | 144 + .../BM/html/reveal.js/css/theme/darkgray.css | 153 + .../BM/html/reveal.js/css/theme/default.css | 153 + .../BM/html/reveal.js/css/theme/league.css | 279 + doc/pub/BM/html/reveal.js/css/theme/moon.css | 153 + doc/pub/BM/html/reveal.js/css/theme/night.css | 141 + doc/pub/BM/html/reveal.js/css/theme/serif.css | 143 + .../BM/html/reveal.js/css/theme/simple.css | 144 + .../BM/html/reveal.js/css/theme/simula.css | 144 + doc/pub/BM/html/reveal.js/css/theme/sky.css | 150 + .../BM/html/reveal.js/css/theme/solarized.css | 153 + .../reveal.js/css/theme/source/beige.scss | 50 + .../css/theme/source/beigesmall.scss | 51 + .../reveal.js/css/theme/source/black.scss | 49 + .../reveal.js/css/theme/source/blood.scss | 91 + .../html/reveal.js/css/theme/source/cbc.scss | 39 + .../reveal.js/css/theme/source/darkgray.scss | 42 + .../reveal.js/css/theme/source/default.scss | 42 + .../reveal.js/css/theme/source/league.scss | 34 + .../html/reveal.js/css/theme/source/moon.scss | 68 + .../reveal.js/css/theme/source/night.scss | 35 + .../reveal.js/css/theme/source/serif.scss | 35 + .../reveal.js/css/theme/source/simple.scss | 38 + .../reveal.js/css/theme/source/simula.scss | 39 + .../html/reveal.js/css/theme/source/sky.scss | 46 + .../reveal.js/css/theme/source/solarized.scss | 74 + .../reveal.js/css/theme/source/white.scss | 49 + .../reveal.js/css/theme/template/mixins.scss | 29 + .../css/theme/template/settings.scss | 34 + .../reveal.js/css/theme/template/theme.scss | 171 + doc/pub/BM/html/reveal.js/css/theme/white.css | 273 + doc/pub/BM/html/reveal.js/demo.html | 410 ++ doc/pub/BM/html/reveal.js/index.html | 49 + doc/pub/BM/html/reveal.js/js/reveal.js | 5241 +++++++++++++++++ doc/pub/BM/html/reveal.js/lib/css/zenburn.css | 80 + .../reveal.js/lib/font/league-gothic/LICENSE | 2 + .../lib/font/league-gothic/league-gothic.css | 10 + .../lib/font/league-gothic/league-gothic.eot | Bin 0 -> 25696 bytes .../lib/font/league-gothic/league-gothic.ttf | Bin 0 -> 64256 bytes .../lib/font/league-gothic/league-gothic.woff | Bin 0 -> 30764 bytes .../lib/font/source-sans-pro/LICENSE | 45 + .../source-sans-pro-italic.eot | Bin 0 -> 75720 bytes .../source-sans-pro-italic.ttf | Bin 0 -> 238084 bytes .../source-sans-pro-italic.woff | Bin 0 -> 98556 bytes .../source-sans-pro-regular.eot | Bin 0 -> 88070 bytes .../source-sans-pro-regular.ttf | Bin 0 -> 288008 bytes .../source-sans-pro-regular.woff | Bin 0 -> 114324 bytes .../source-sans-pro-semibold.eot | Bin 0 -> 89897 bytes .../source-sans-pro-semibold.ttf | Bin 0 -> 284640 bytes .../source-sans-pro-semibold.woff | Bin 0 -> 115648 bytes .../source-sans-pro-semibolditalic.eot | Bin 0 -> 75706 bytes .../source-sans-pro-semibolditalic.ttf | Bin 0 -> 240944 bytes .../source-sans-pro-semibolditalic.woff | Bin 0 -> 98816 bytes .../font/source-sans-pro/source-sans-pro.css | 39 + doc/pub/BM/html/reveal.js/lib/js/classList.js | 2 + doc/pub/BM/html/reveal.js/lib/js/head.min.js | 9 + doc/pub/BM/html/reveal.js/lib/js/html5shiv.js | 7 + doc/pub/BM/html/reveal.js/package.json | 43 + .../reveal.js/plugin/highlight/highlight.js | 77 + doc/pub/BM/html/reveal.js/plugin/leap/leap.js | 159 + .../reveal.js/plugin/markdown/example.html | 136 + .../html/reveal.js/plugin/markdown/example.md | 36 + .../reveal.js/plugin/markdown/markdown.js | 412 ++ .../html/reveal.js/plugin/markdown/marked.js | 6 + doc/pub/BM/html/reveal.js/plugin/math/math.js | 67 + .../html/reveal.js/plugin/multiplex/client.js | 13 + .../html/reveal.js/plugin/multiplex/index.js | 64 + .../html/reveal.js/plugin/multiplex/master.js | 34 + .../reveal.js/plugin/multiplex/package.json | 19 + .../reveal.js/plugin/notes-server/client.js | 65 + .../reveal.js/plugin/notes-server/index.js | 69 + .../reveal.js/plugin/notes-server/notes.html | 585 ++ .../BM/html/reveal.js/plugin/notes/notes.html | 759 +++ .../BM/html/reveal.js/plugin/notes/notes.js | 155 + .../reveal.js/plugin/print-pdf/print-pdf.js | 69 + .../html/reveal.js/plugin/remotes/remotes.js | 39 + .../BM/html/reveal.js/plugin/search/search.js | 206 + .../BM/html/reveal.js/plugin/zoom-js/zoom.js | 272 + .../reveal.js/test/examples/assets/image1.png | Bin 0 -> 21991 bytes .../reveal.js/test/examples/assets/image2.png | Bin 0 -> 10237 bytes .../reveal.js/test/examples/barebones.html | 41 + .../test/examples/embedded-media.html | 49 + .../BM/html/reveal.js/test/examples/math.html | 185 + .../test/examples/slide-backgrounds.html | 144 + .../test/examples/slide-transitions.html | 101 + .../BM/html/reveal.js/test/qunit-1.12.0.css | 244 + .../BM/html/reveal.js/test/qunit-1.12.0.js | 2212 +++++++ doc/pub/BM/html/reveal.js/test/simple.md | 12 + .../test-markdown-element-attributes.html | 134 + .../test/test-markdown-element-attributes.js | 46 + .../test/test-markdown-external.html | 36 + .../reveal.js/test/test-markdown-external.js | 24 + .../reveal.js/test/test-markdown-options.html | 41 + .../reveal.js/test/test-markdown-options.js | 26 + .../test/test-markdown-slide-attributes.html | 128 + .../test/test-markdown-slide-attributes.js | 47 + .../BM/html/reveal.js/test/test-markdown.html | 52 + .../BM/html/reveal.js/test/test-markdown.js | 15 + doc/pub/BM/html/reveal.js/test/test-pdf.html | 83 + doc/pub/BM/html/reveal.js/test/test-pdf.js | 15 + doc/pub/BM/html/reveal.js/test/test.html | 86 + doc/pub/BM/html/reveal.js/test/test.js | 597 ++ doc/pub/BM/html/src/Hudson_Bay.csv | 22 + doc/pub/BM/html/src/Hudson_Bay.py | 43 + doc/pub/BM/html/src/Hudson_Bay.py~ | 43 + doc/pub/BM/html/src/Hudson_Bay_sim.pdf | Bin 0 -> 17919 bytes doc/pub/BM/html/src/Hudson_Bay_sim.png | Bin 0 -> 67454 bytes doc/pub/BM/ipynb/BM.ipynb | 1826 ++++++ doc/pub/BM/ipynb/figures/Hudson_Bay_data.pdf | Bin 0 -> 15045 bytes doc/pub/BM/ipynb/figures/Hudson_Bay_data.png | Bin 0 -> 53586 bytes doc/pub/BM/ipynb/figures/MLrbm.pdf | Bin 0 -> 11007 bytes doc/pub/BM/ipynb/figures/MLrbm.png | Bin 0 -> 11333 bytes doc/pub/BM/ipynb/figures/RBM.pdf | 1623 +++++ doc/pub/BM/ipynb/figures/RBM.png | Bin 0 -> 38001 bytes doc/pub/BM/ipynb/figures/RMBenergy.dat | 100 + doc/pub/BM/ipynb/figures/RMBenergy.dat~ | 100 + doc/pub/BM/ipynb/figures/dnn.png | Bin 0 -> 274267 bytes doc/pub/BM/ipynb/figures/figN2.pdf | Bin 0 -> 22919 bytes doc/pub/BM/ipynb/figures/figN2.png | Bin 0 -> 20041 bytes doc/pub/BM/ipynb/figures/figN6.pdf | Bin 0 -> 23104 bytes doc/pub/BM/ipynb/figures/figN6.png | Bin 0 -> 20344 bytes doc/pub/BM/ipynb/figures/perceptron.png | Bin 0 -> 61405 bytes doc/pub/BM/ipynb/figures/plot.py | 14 + doc/pub/BM/ipynb/figures/plot.py~ | 14 + doc/pub/BM/ipynb/figures/plotEnergies.py | 61 + doc/pub/BM/ipynb/figures/plotEnergies.py~ | 62 + .../BM/ipynb/figures/w1.0_D2_N2_Energies.txt | 177 + .../BM/ipynb/figures/w1.0_D2_N6_Energies.txt | 164 + doc/pub/BM/ipynb/ipynb-BM-src.tar.gz | Bin 0 -> 429509 bytes doc/pub/BM/ipynb/src/Hudson_Bay.csv | 22 + doc/pub/BM/ipynb/src/Hudson_Bay.py | 43 + doc/pub/BM/ipynb/src/Hudson_Bay.py~ | 43 + doc/pub/BM/ipynb/src/Hudson_Bay_sim.pdf | Bin 0 -> 17919 bytes doc/pub/BM/ipynb/src/Hudson_Bay_sim.png | Bin 0 -> 67454 bytes doc/pub/BM/pdf/BM-beamer-handouts2x3.pdf | Bin 0 -> 1176339 bytes doc/pub/BM/pdf/BM-beamer.pdf | Bin 0 -> 1158940 bytes doc/pub/BM/pdf/BM-minted.pdf | Bin 0 -> 1161785 bytes 184 files changed, 35836 insertions(+) create mode 100644 doc/pub/BM/html/BM-bs.html create mode 100644 doc/pub/BM/html/BM-reveal.html create mode 100644 doc/pub/BM/html/BM-solarized.html create mode 100644 doc/pub/BM/html/BM.html create mode 100644 doc/pub/BM/html/figures/Hudson_Bay_data.pdf create mode 100644 doc/pub/BM/html/figures/Hudson_Bay_data.png create mode 100644 doc/pub/BM/html/figures/MLrbm.pdf create mode 100644 doc/pub/BM/html/figures/MLrbm.png create mode 100644 doc/pub/BM/html/figures/RBM.pdf create mode 100644 doc/pub/BM/html/figures/RBM.png create mode 100644 doc/pub/BM/html/figures/RMBenergy.dat create mode 100644 doc/pub/BM/html/figures/RMBenergy.dat~ create mode 100644 doc/pub/BM/html/figures/dnn.png create mode 100644 doc/pub/BM/html/figures/figN2.pdf create mode 100644 doc/pub/BM/html/figures/figN2.png create mode 100644 doc/pub/BM/html/figures/figN6.pdf create mode 100644 doc/pub/BM/html/figures/figN6.png create mode 100644 doc/pub/BM/html/figures/perceptron.png create mode 100644 doc/pub/BM/html/figures/plot.py create mode 100644 doc/pub/BM/html/figures/plot.py~ create mode 100644 doc/pub/BM/html/figures/plotEnergies.py create mode 100644 doc/pub/BM/html/figures/plotEnergies.py~ create mode 100644 doc/pub/BM/html/figures/w1.0_D2_N2_Energies.txt create mode 100644 doc/pub/BM/html/figures/w1.0_D2_N6_Energies.txt create mode 100644 doc/pub/BM/html/reveal.js/.gitignore create mode 100644 doc/pub/BM/html/reveal.js/.travis.yml create mode 100644 doc/pub/BM/html/reveal.js/CONTRIBUTING.md create mode 100644 doc/pub/BM/html/reveal.js/Gruntfile.js create mode 100644 doc/pub/BM/html/reveal.js/LICENSE create mode 100644 doc/pub/BM/html/reveal.js/README.md create mode 100644 doc/pub/BM/html/reveal.js/bower.json create mode 100644 doc/pub/BM/html/reveal.js/css/images/cbc_footer.png create mode 100644 doc/pub/BM/html/reveal.js/css/images/cbc_symbol.png create mode 100644 doc/pub/BM/html/reveal.js/css/images/simula_footer.png create mode 100644 doc/pub/BM/html/reveal.js/css/images/simula_logo.png create mode 100644 doc/pub/BM/html/reveal.js/css/images/simula_symbol.png create mode 100644 doc/pub/BM/html/reveal.js/css/images/uio_footer.png create mode 100644 doc/pub/BM/html/reveal.js/css/images/uio_symbol.png create mode 100644 doc/pub/BM/html/reveal.js/css/print/paper.css create mode 100644 doc/pub/BM/html/reveal.js/css/print/pdf.css create mode 100644 doc/pub/BM/html/reveal.js/css/reveal.css create mode 100644 doc/pub/BM/html/reveal.js/css/reveal.scss create mode 100644 doc/pub/BM/html/reveal.js/css/theme/README.md create mode 100644 doc/pub/BM/html/reveal.js/css/theme/beige.css create mode 100644 doc/pub/BM/html/reveal.js/css/theme/beigesmall.css create mode 100644 doc/pub/BM/html/reveal.js/css/theme/black.css create mode 100644 doc/pub/BM/html/reveal.js/css/theme/blood.css create mode 100644 doc/pub/BM/html/reveal.js/css/theme/cbc.css create mode 100644 doc/pub/BM/html/reveal.js/css/theme/darkgray.css create mode 100644 doc/pub/BM/html/reveal.js/css/theme/default.css create mode 100644 doc/pub/BM/html/reveal.js/css/theme/league.css create mode 100644 doc/pub/BM/html/reveal.js/css/theme/moon.css create mode 100644 doc/pub/BM/html/reveal.js/css/theme/night.css create mode 100644 doc/pub/BM/html/reveal.js/css/theme/serif.css create mode 100644 doc/pub/BM/html/reveal.js/css/theme/simple.css create mode 100644 doc/pub/BM/html/reveal.js/css/theme/simula.css create mode 100644 doc/pub/BM/html/reveal.js/css/theme/sky.css create mode 100644 doc/pub/BM/html/reveal.js/css/theme/solarized.css create mode 100644 doc/pub/BM/html/reveal.js/css/theme/source/beige.scss create mode 100644 doc/pub/BM/html/reveal.js/css/theme/source/beigesmall.scss create mode 100644 doc/pub/BM/html/reveal.js/css/theme/source/black.scss create mode 100644 doc/pub/BM/html/reveal.js/css/theme/source/blood.scss create mode 100644 doc/pub/BM/html/reveal.js/css/theme/source/cbc.scss create mode 100644 doc/pub/BM/html/reveal.js/css/theme/source/darkgray.scss create mode 100644 doc/pub/BM/html/reveal.js/css/theme/source/default.scss create mode 100644 doc/pub/BM/html/reveal.js/css/theme/source/league.scss create mode 100644 doc/pub/BM/html/reveal.js/css/theme/source/moon.scss create mode 100644 doc/pub/BM/html/reveal.js/css/theme/source/night.scss create mode 100644 doc/pub/BM/html/reveal.js/css/theme/source/serif.scss create mode 100644 doc/pub/BM/html/reveal.js/css/theme/source/simple.scss create mode 100644 doc/pub/BM/html/reveal.js/css/theme/source/simula.scss create mode 100644 doc/pub/BM/html/reveal.js/css/theme/source/sky.scss create mode 100644 doc/pub/BM/html/reveal.js/css/theme/source/solarized.scss create mode 100644 doc/pub/BM/html/reveal.js/css/theme/source/white.scss create mode 100644 doc/pub/BM/html/reveal.js/css/theme/template/mixins.scss create mode 100644 doc/pub/BM/html/reveal.js/css/theme/template/settings.scss create mode 100644 doc/pub/BM/html/reveal.js/css/theme/template/theme.scss create mode 100644 doc/pub/BM/html/reveal.js/css/theme/white.css create mode 100644 doc/pub/BM/html/reveal.js/demo.html create mode 100644 doc/pub/BM/html/reveal.js/index.html create mode 100644 doc/pub/BM/html/reveal.js/js/reveal.js create mode 100644 doc/pub/BM/html/reveal.js/lib/css/zenburn.css create mode 100644 doc/pub/BM/html/reveal.js/lib/font/league-gothic/LICENSE create mode 100644 doc/pub/BM/html/reveal.js/lib/font/league-gothic/league-gothic.css create mode 100644 doc/pub/BM/html/reveal.js/lib/font/league-gothic/league-gothic.eot create mode 100644 doc/pub/BM/html/reveal.js/lib/font/league-gothic/league-gothic.ttf create mode 100644 doc/pub/BM/html/reveal.js/lib/font/league-gothic/league-gothic.woff create mode 100644 doc/pub/BM/html/reveal.js/lib/font/source-sans-pro/LICENSE create mode 100644 doc/pub/BM/html/reveal.js/lib/font/source-sans-pro/source-sans-pro-italic.eot create mode 100644 doc/pub/BM/html/reveal.js/lib/font/source-sans-pro/source-sans-pro-italic.ttf create mode 100644 doc/pub/BM/html/reveal.js/lib/font/source-sans-pro/source-sans-pro-italic.woff create mode 100644 doc/pub/BM/html/reveal.js/lib/font/source-sans-pro/source-sans-pro-regular.eot create mode 100644 doc/pub/BM/html/reveal.js/lib/font/source-sans-pro/source-sans-pro-regular.ttf create mode 100644 doc/pub/BM/html/reveal.js/lib/font/source-sans-pro/source-sans-pro-regular.woff create mode 100644 doc/pub/BM/html/reveal.js/lib/font/source-sans-pro/source-sans-pro-semibold.eot create mode 100644 doc/pub/BM/html/reveal.js/lib/font/source-sans-pro/source-sans-pro-semibold.ttf create mode 100644 doc/pub/BM/html/reveal.js/lib/font/source-sans-pro/source-sans-pro-semibold.woff create mode 100644 doc/pub/BM/html/reveal.js/lib/font/source-sans-pro/source-sans-pro-semibolditalic.eot create mode 100644 doc/pub/BM/html/reveal.js/lib/font/source-sans-pro/source-sans-pro-semibolditalic.ttf create mode 100644 doc/pub/BM/html/reveal.js/lib/font/source-sans-pro/source-sans-pro-semibolditalic.woff create mode 100644 doc/pub/BM/html/reveal.js/lib/font/source-sans-pro/source-sans-pro.css create mode 100644 doc/pub/BM/html/reveal.js/lib/js/classList.js create mode 100644 doc/pub/BM/html/reveal.js/lib/js/head.min.js create mode 100644 doc/pub/BM/html/reveal.js/lib/js/html5shiv.js create mode 100644 doc/pub/BM/html/reveal.js/package.json create mode 100644 doc/pub/BM/html/reveal.js/plugin/highlight/highlight.js create mode 100644 doc/pub/BM/html/reveal.js/plugin/leap/leap.js create mode 100644 doc/pub/BM/html/reveal.js/plugin/markdown/example.html create mode 100644 doc/pub/BM/html/reveal.js/plugin/markdown/example.md create mode 100644 doc/pub/BM/html/reveal.js/plugin/markdown/markdown.js create mode 100644 doc/pub/BM/html/reveal.js/plugin/markdown/marked.js create mode 100644 doc/pub/BM/html/reveal.js/plugin/math/math.js create mode 100644 doc/pub/BM/html/reveal.js/plugin/multiplex/client.js create mode 100644 doc/pub/BM/html/reveal.js/plugin/multiplex/index.js create mode 100644 doc/pub/BM/html/reveal.js/plugin/multiplex/master.js create mode 100644 doc/pub/BM/html/reveal.js/plugin/multiplex/package.json create mode 100644 doc/pub/BM/html/reveal.js/plugin/notes-server/client.js create mode 100644 doc/pub/BM/html/reveal.js/plugin/notes-server/index.js create mode 100644 doc/pub/BM/html/reveal.js/plugin/notes-server/notes.html create mode 100644 doc/pub/BM/html/reveal.js/plugin/notes/notes.html create mode 100644 doc/pub/BM/html/reveal.js/plugin/notes/notes.js create mode 100644 doc/pub/BM/html/reveal.js/plugin/print-pdf/print-pdf.js create mode 100644 doc/pub/BM/html/reveal.js/plugin/remotes/remotes.js create mode 100644 doc/pub/BM/html/reveal.js/plugin/search/search.js create mode 100644 doc/pub/BM/html/reveal.js/plugin/zoom-js/zoom.js create mode 100644 doc/pub/BM/html/reveal.js/test/examples/assets/image1.png create mode 100644 doc/pub/BM/html/reveal.js/test/examples/assets/image2.png create mode 100644 doc/pub/BM/html/reveal.js/test/examples/barebones.html create mode 100644 doc/pub/BM/html/reveal.js/test/examples/embedded-media.html create mode 100644 doc/pub/BM/html/reveal.js/test/examples/math.html create mode 100644 doc/pub/BM/html/reveal.js/test/examples/slide-backgrounds.html create mode 100644 doc/pub/BM/html/reveal.js/test/examples/slide-transitions.html create mode 100644 doc/pub/BM/html/reveal.js/test/qunit-1.12.0.css create mode 100644 doc/pub/BM/html/reveal.js/test/qunit-1.12.0.js create mode 100644 doc/pub/BM/html/reveal.js/test/simple.md create mode 100644 doc/pub/BM/html/reveal.js/test/test-markdown-element-attributes.html create mode 100644 doc/pub/BM/html/reveal.js/test/test-markdown-element-attributes.js create mode 100644 doc/pub/BM/html/reveal.js/test/test-markdown-external.html create mode 100644 doc/pub/BM/html/reveal.js/test/test-markdown-external.js create mode 100644 doc/pub/BM/html/reveal.js/test/test-markdown-options.html create mode 100644 doc/pub/BM/html/reveal.js/test/test-markdown-options.js create mode 100644 doc/pub/BM/html/reveal.js/test/test-markdown-slide-attributes.html create mode 100644 doc/pub/BM/html/reveal.js/test/test-markdown-slide-attributes.js create mode 100644 doc/pub/BM/html/reveal.js/test/test-markdown.html create mode 100644 doc/pub/BM/html/reveal.js/test/test-markdown.js create mode 100644 doc/pub/BM/html/reveal.js/test/test-pdf.html create mode 100644 doc/pub/BM/html/reveal.js/test/test-pdf.js create mode 100644 doc/pub/BM/html/reveal.js/test/test.html create mode 100644 doc/pub/BM/html/reveal.js/test/test.js create mode 100644 doc/pub/BM/html/src/Hudson_Bay.csv create mode 100644 doc/pub/BM/html/src/Hudson_Bay.py create mode 100644 doc/pub/BM/html/src/Hudson_Bay.py~ create mode 100644 doc/pub/BM/html/src/Hudson_Bay_sim.pdf create mode 100644 doc/pub/BM/html/src/Hudson_Bay_sim.png create mode 100644 doc/pub/BM/ipynb/BM.ipynb create mode 100644 doc/pub/BM/ipynb/figures/Hudson_Bay_data.pdf create mode 100644 doc/pub/BM/ipynb/figures/Hudson_Bay_data.png create mode 100644 doc/pub/BM/ipynb/figures/MLrbm.pdf create mode 100644 doc/pub/BM/ipynb/figures/MLrbm.png create mode 100644 doc/pub/BM/ipynb/figures/RBM.pdf create mode 100644 doc/pub/BM/ipynb/figures/RBM.png create mode 100644 doc/pub/BM/ipynb/figures/RMBenergy.dat create mode 100644 doc/pub/BM/ipynb/figures/RMBenergy.dat~ create mode 100644 doc/pub/BM/ipynb/figures/dnn.png create mode 100644 doc/pub/BM/ipynb/figures/figN2.pdf create mode 100644 doc/pub/BM/ipynb/figures/figN2.png create mode 100644 doc/pub/BM/ipynb/figures/figN6.pdf create mode 100644 doc/pub/BM/ipynb/figures/figN6.png create mode 100644 doc/pub/BM/ipynb/figures/perceptron.png create mode 100644 doc/pub/BM/ipynb/figures/plot.py create mode 100644 doc/pub/BM/ipynb/figures/plot.py~ create mode 100644 doc/pub/BM/ipynb/figures/plotEnergies.py create mode 100644 doc/pub/BM/ipynb/figures/plotEnergies.py~ create mode 100644 doc/pub/BM/ipynb/figures/w1.0_D2_N2_Energies.txt create mode 100644 doc/pub/BM/ipynb/figures/w1.0_D2_N6_Energies.txt create mode 100644 doc/pub/BM/ipynb/ipynb-BM-src.tar.gz create mode 100644 doc/pub/BM/ipynb/src/Hudson_Bay.csv create mode 100644 doc/pub/BM/ipynb/src/Hudson_Bay.py create mode 100644 doc/pub/BM/ipynb/src/Hudson_Bay.py~ create mode 100644 doc/pub/BM/ipynb/src/Hudson_Bay_sim.pdf create mode 100644 doc/pub/BM/ipynb/src/Hudson_Bay_sim.png create mode 100644 doc/pub/BM/pdf/BM-beamer-handouts2x3.pdf create mode 100644 doc/pub/BM/pdf/BM-beamer.pdf create mode 100644 doc/pub/BM/pdf/BM-minted.pdf diff --git a/doc/pub/BM/html/BM-bs.html b/doc/pub/BM/html/BM-bs.html new file mode 100644 index 000000000..a2a6f79f8 --- /dev/null +++ b/doc/pub/BM/html/BM-bs.html @@ -0,0 +1,1639 @@ + + + + + + + +Solving quantum mechanical problems with Machine Learning and Boltzmann machines + + + + + + + + + + + + + + + + + + + + + + + +
+
+ + Solving quantum mechanical problems with Machine Learning and Boltzmann machines +
+ +
+ +
+
+ + +
+ +

 

 

 

+ + + + + +
+

Solving quantum mechanical problems with Machine Learning and Boltzmann machines

+ +

+ + +

+Morten Hjorth-Jensen +
+ +

+ + +

Department of Physics and Astronomy and National Superconducting Cyclotron Laboratory, Michigan State University and Department of Physics, University of Oslo, Norway
+
+

+

Jul 26, 2018

+
+

+ +

+ + + +

What is this talk about?

+
+
+

+The main aim is to give you a short and pedestrian introduction to how we can use Machine Learning methods +to solve quantum mechanical many-body problems. And why this could be of interest. + +

+The hope is that after this talk you have gotten the basic ideas to get you started. Peeping into https://github.com/mhjensenseminars/MachineLearningTalk, you'll find a Jupyter notebook, slides, codes etc that will allow you to reproduce the simulations discussed here, and perhaps run your own very first calculations. + +

+Furthermore, since we will also talk about Monte Carlo simulations, the exciting idea of Monte Carlo optimization and Machine Learning can be used to mesh strategies for reducing the curse of dimensionality. + +

+

+
+ + +

+ + +

Why?

+ +

+How can we avoid the dimensionality curse? Many possibilities + +

    +
  1. smarter basis functions
    2. +
  2. resummation of specific correlations
    3. +
  3. stochastic sampling of high-lying states (stochastic FCI, CC and soon SRG)
    4. +
  4. many more
    5. +
+ +Machine Learning and Quantum Computing hold also great promise in tackling the +ever increasing dimensionalities. Here we will focus on Machine Learning. + +

+ + +

Overview

+
+
+

+ +

    +
  • Short intro to Machine Learning
    • +
  • Variational Monte Carlo (Markov Chain Monte Carlo, \( \mathrm{MC}^2 \)) and two-electron quantum dots, solving quantum mechanical problems in a stochastic way. It will serve as our motivation for switching to Machine Learning.
    • +
  • From Variational Monte Carlo to Boltzmann Machines and Machine Learning
    • +
+
+
+ + +

+ + +

What are the Machine Learning calculations here based on?

+ +

+This work is inspired by the idea of representing the wave function with +a restricted Boltzmann machine (RBM), presented recently by G. Carleo and M. Troyer, Science 355, Issue 6325, pp. 602-606 (2017). They +named such a wave function/network a neural network quantum state (NQS). In their article they apply it to the quantum mechanical +spin lattice systems of the Ising model and Heisenberg model, with +encouraging results. + +

+Thanks to Vilde Flugsrud and Alfred Alocias Mariadason for many discussions and interpretations of the above paper. + +

+ + +

A new world

+
+
+

+Machine learning (ML) is an extremely rich field, in spite of its young age. The +increases we have seen during the last three decades in computational +capabilities have been followed by developments of methods and +techniques for analyzing and handling large date sets, relying heavily +on statistics, computer science and mathematics. The field is rather +new and developing rapidly. + +

+Popular software packages written in Python for ML are + +

+ +These are all freely available at their respective GitHub sites. They +encompass communities of developers in the thousands or more. And the number +of code developers and contributors keeps increasing. +
+
+ + +

+ + +

Lots of room for creativity

+
+
+

+Not all the +algorithms and methods can be given a rigorous mathematical +justification, opening up thereby for experimenting +and trial and error and thereby exciting new developments. +

+
+ + +

+

+
+

+A solid command of linear algebra, multivariate theory, +probability theory, statistical data analysis, +understanding errors and Monte Carlo methods is important in order to understand many of the +various algorithms and methods. +

+
+ + +

+

+
+

+A familiarity with ML is almost becoming a prerequisite for many of the most exciting employment +opportunities. Add quantum computing and there you are. +

+
+ + +

+ + +

Some members of the ML family

+ +

+

+
+

+ +

    +
  1. Linear regression and its variants, in essence polynomial regression
    2. +
  2. Decision tree algorithms, from simpler to more complex ones like random forests
    3. +
  3. Nearest neighbors models
    4. +
  4. Bayesian statistics and regression
    5. +
  5. Support vector machines and finally various variants of
    6. +
  6. Artifical neural networks and deep learning
    7. +
  7. ....
    8. +
+
+
+ + +

+ + +

What are the basic ingredients?

+
+
+

+Almost every problem in ML and data science starts with the same ingredients: + +

    +
  • The dataset \( \mathbf{x} \) (could be some observable quantity of the system we are studying)
    • +
  • A model which is a function of a set of parameters \( \mathbf{\alpha} \) that relates to the dataset, say a likelihood function \( p(\mathbf{x}\vert \mathbf{\alpha}) \) or just a simple model \( f(\mathbf{\alpha}) \)
    • +
  • A so-called cost function \( \mathcal{C} (\mathbf{x}, f(\mathbf{\alpha})) \) which allows us to decide how well our model represents the dataset.
    • +
+ +We seek to minimize the function \( \mathcal{C} (\mathbf{x}, f(\mathbf{\alpha})) \) by finding the parameter values which minimize \( \mathcal{C} \). This leads to various minimization algorithms. +
+
+ + +

+ + +

What is Machine Learning?

+
+
+

+Machine learning is the science of giving computers the ability to +learn without being explicitly programmed. The idea is that there +exist generic algorithms which can be used to find patterns in a broad +class of data sets without having to write code specifically for each +problem. The algorithm will build its own logic based on the data. + +

+Machine learning is a subfield of computer science, and is closely +related to computational statistics. It evolved from the study of +pattern recognition in artificial intelligence (AI) research, and has +made contributions to AI tasks like computer vision, natural language +processing and speech recognition. It has also, especially in later +years, found applications in a wide variety of other areas, including +bioinformatics, economy, physics, finance and marketing. + +

+You will notice however that many of the basic ideas discussed do come from Physics! +

+
+ + +

+ + +

Types of Machine Learning

+ +

+

+
+

+The approaches to machine learning are many, but are often split into two main categories. +In supervised learning we know the answer to a problem, +and let the computer deduce the logic behind it. On the other hand, unsupervised learning +is a method for finding patterns and relationship in data sets without any prior knowledge of the system. +Some authours also operate with a third category, namely reinforcement learning. This is a paradigm +of learning inspired by behavioural psychology, where learning is achieved by trial-and-error, +solely from rewards and punishment. + +

+Another way to categorize machine learning tasks is to consider the desired output of a system. +Some of the most common tasks are: + +

    +
  • Classification: Outputs are divided into two or more classes. The goal is to produce a model that assigns inputs into one of these classes. An example is to identify digits based on pictures of hand-written ones. Classification is typically supervised learning.
    • +
  • Regression: Finding a functional relationship between an input data set and a reference data set. The goal is to construct a function that maps input data to continuous output values.
    • +
  • Clustering: Data are divided into groups with certain common traits, without knowing the different groups beforehand. It is thus a form of unsupervised learning.
    • +
+
+
+ + +

+ + +

References

+
+
+

+ +

+
+
+ + +

+ + +

Another interesting article

+ +

+Here we will use so-called reduced Boltzmann Machines to simulate quantum many-body problems. For Monte Carlo aficionados, there is a very close similarity with what are called shadow wave functions, see the work of Pederiva and Kalos and collaborators, Phys Rev. E 90, 053304 (2014). + +

+ + +

Just for fun illustration of scikit-learn: Hudson bay data

+
+
+

+Most mammalian predators rely on a variety of prey, which complicates mathematical modeling; however, a few predators have become highly specialized and seek almost exclusively a single prey species. An example of this simplified predator-prey interaction is seen in Canadian northern forests, where the populations of the lynx and the snowshoe hare are intertwined in a life and death struggle. + +

+One reason that this particular system has been so extensively studied is that the Hudson Bay company kept careful records of all furs from the early 1800s into the 1900s. The records for the furs collected by the Hudson Bay company showed distinct oscillations (approximately 12 year periods), suggesting that these species caused almost periodic fluctuations of each other's populations. The table here shows data from 1900 to 1920. + +

+

+
+ + +

+ + +

Hares and lynx in Hudson bay from 1900 to 1920

+ +

+



+ +

+ + +

Linear regression in Python

+
+
+

+

+ + +

import numpy as np
+import matplotlib.pyplot as plt
+from IPython.display import display
+import sklearn
+from sklearn.linear_model import LinearRegression
+from sklearn.tree import DecisionTreeRegressor
+
+
+data = np.loadtxt('src/Hudson_Bay.csv', delimiter=',', skiprows=1)
+x = data[:,0]
+y = data[:,1]
+line = np.linspace(1900,1920,1000,endpoint=False).reshape(-1,1)
+reg = DecisionTreeRegressor(min_samples_split=3).fit(x.reshape(-1,1),y.reshape(-1,1))
+plt.plot(line, reg.predict(line), label="decision tree")
+regline = LinearRegression().fit(x.reshape(-1,1),y.reshape(-1,1))
+plt.plot(line, regline.predict(line), label= "Linear Regression")
+plt.plot(x, y, label= "Linear Regression")
+plt.show()
+
+

+

+
+ + +

+ + +

Artificial neurons

+
+
+

+ +

+The field of artificial neural networks has a long history of +development, and is closely connected with the advancement of computer +science and computers in general. A model of artificial neurons was +first developed by McCulloch and Pitts in 1943 to study signal +processing in the brain and has later been refined by others. The +general idea is to mimic neural networks in the human brain, which is +composed of billions of neurons that communicate with each other by +sending electrical signals. Each neuron accumulates its incoming +signals, which must exceed an activation threshold to yield an +output. If the threshold is not overcome, the neuron remains inactive, +i.e. has zero output. + +

+This behaviour has inspired a simple mathematical model for an artificial neuron. + +$$ + y = f\left(\sum_{i=1}^n w_ix_i\right) = f(u) +$$ + +Here, the output \( y \) of the neuron is the value of its activation function, which have as input +a weighted sum of signals \( x_i, \dots ,x_n \) received by \( n \) other neurons. +

+
+ + +

+ + +

A simple perceptron model

+ +

+



+ +

+ + +

Neural network types

+
+
+

+An artificial neural network (NN), is a computational model that consists of layers of connected neurons, or nodes. +It is supposed to mimic a biological nervous system by letting each neuron interact with other neurons +by sending signals in the form of mathematical functions between layers. +A wide variety of different NNs have +been developed, but most of them consist of an input layer, an output layer and eventual layers in-between, called +hidden layers. All layers can contain an arbitrary number of nodes, and each connection between two nodes +is associated with a weight variable. + +

+



+
+
+ + +

+ + +

The system: two electrons in a harmonic oscillator trap in two dimensions

+ +

+The Hamiltonian of the quantum dot is given by +$$ \hat{H} = \hat{H}_0 + \hat{V}, +$$ + +where \( \hat{H}_0 \) is the many-body HO Hamiltonian, and \( \hat{V} \) is the +inter-electron Coulomb interactions. In dimensionless units, +$$ \hat{V}= \sum_{i < j}^N \frac{1}{r_{ij}}, +$$ + +with \( r_{ij}=\sqrt{\mathbf{r}_i^2 - \mathbf{r}_j^2} \). + +

+This leads to the separable Hamiltonian, with the relative motion part given by (\( r_{ij}=r \)) +$$ +\hat{H}_r=-\nabla^2_r + \frac{1}{4}\omega^2r^2+ \frac{1}{r}, +$$ + +plus a standard Harmonic Oscillator problem for the center-of-mass motion. +This system has analytical solutions in two and three dimensions (M. Taut 1993 and 1994). + +

+ + +

Quantum Monte Carlo Motivation

+
+
+

+Given a hamiltonian \( H \) and a trial wave function \( \Psi_T \), the variational principle states that the expectation value of \( \langle H \rangle \), defined through +$$ + \langle E \rangle = + \frac{\int d\boldsymbol{R}\Psi^{\ast}_T(\boldsymbol{R})H(\boldsymbol{R})\Psi_T(\boldsymbol{R})} + {\int d\boldsymbol{R}\Psi^{\ast}_T(\boldsymbol{R})\Psi_T(\boldsymbol{R})}, +$$ + +is an upper bound to the ground state energy \( E_0 \) of the hamiltonian \( H \), that is +$$ + E_0 \le \langle H \rangle . +$$ + +In general, the integrals involved in the calculation of various expectation values are multi-dimensional ones. Traditional integration methods such as the Gauss-Legendre will not be adequate for say the computation of the energy of a many-body system. +

+
+ + +

+ + +

Quantum Monte Carlo Motivation

+
+
+

+Choose a trial wave function +\( \psi_T(\boldsymbol{R}) \). +$$ + P(\boldsymbol{R},\boldsymbol{\alpha})= \frac{\left|\psi_T(\boldsymbol{R},\boldsymbol{\alpha})\right|^2}{\int \left|\psi_T(\boldsymbol{R},\boldsymbol{\alpha})\right|^2d\boldsymbol{R}}. +$$ + +This is our model, or likelihood/probability distribution function (PDF). It depends on some variational parameters \( \boldsymbol{\alpha} \). +The approximation to the expectation value of the Hamiltonian is now +$$ + \langle E[\boldsymbol{\alpha}] \rangle = + \frac{\int d\boldsymbol{R}\Psi^{\ast}_T(\boldsymbol{R},\boldsymbol{\alpha})H(\boldsymbol{R})\Psi_T(\boldsymbol{R},\boldsymbol{\alpha})} + {\int d\boldsymbol{R}\Psi^{\ast}_T(\boldsymbol{R},\boldsymbol{\alpha})\Psi_T(\boldsymbol{R},\boldsymbol{\alpha})}. +$$ +

+
+ + +

+ + +

Quantum Monte Carlo Motivation

+
+
+

+$$ + E_L(\boldsymbol{R},\boldsymbol{\alpha})=\frac{1}{\psi_T(\boldsymbol{R},\boldsymbol{\alpha})}H\psi_T(\boldsymbol{R},\boldsymbol{\alpha}), +$$ + +called the local energy, which, together with our trial PDF yields +$$ + E[\boldsymbol{\alpha}]=\int P(\boldsymbol{R})E_L(\boldsymbol{R},\boldsymbol{\alpha}) d\boldsymbol{R}\approx \frac{1}{N}\sum_{i=1}^NE_L(\boldsymbol{R_i},\boldsymbol{\alpha}) +$$ + +with \( N \) being the number of Monte Carlo samples. +

+
+ + +

+ + +

Quantum Monte Carlo

+
+
+

+The Algorithm for performing a variational Monte Carlo calculations runs thus as this + +

    +
  • Initialisation: Fix the number of Monte Carlo steps. Choose an initial \( \boldsymbol{R} \) and variational parameters \( \alpha \) and calculate \( \left|\psi_T(\boldsymbol{R},\boldsymbol{\alpha})\right|^2 \).
    • +
  • Initialise the energy and the variance and start the Monte Carlo calculation by looping over trials.
    • + +
      +
    • Calculate a trial position \( \boldsymbol{R}_p=\boldsymbol{R}+r*step \) where \( r \) is a random variable \( r \in [0,1] \).
      • +
    • Metropolis algorithm to accept or reject this move \( w = P(\boldsymbol{R}_p,\boldsymbol{\alpha})/P(\boldsymbol{R},\boldsymbol{\alpha}) \).
      • +
    • If the step is accepted, then we set \( \boldsymbol{R}=\boldsymbol{R}_p \).
      • +
    • Update averages
      • +
    + +
  • Finish and compute final averages.
    • +
+ +Observe that the jumping in space is governed by the variable step. This is often called brute-force sampling. +Need importance sampling to get more relevant sampling. +
+
+ + +

+ + +

The trial wave function

+
+
+

+We want to perform a Variational Monte Carlo calculation of the ground state of two electrons in a quantum dot well with different oscillator energies, assuming total spin \( S=0 \). +Our trial wave function has the following form +$$ +\begin{equation} + \psi_{T}(\boldsymbol{r}_1,\boldsymbol{r}_2) = + C\exp{\left(-\alpha_1\omega(r_1^2+r_2^2)/2\right)} + \exp{\left(\frac{r_{12}}{(1+\alpha_2 r_{12})}\right)}, +\label{eq:trial} +\end{equation} +$$ + +where the $\alpha$s represent our variational parameters, two in this case. + +

+Why does the trial function look like this? How did we get there? This will be our main motivation for switching to +Machine Learning. + +

+

+
+ + +

+ + +

The correlation part of the wave function

+ +

+To find an ansatz for the correlated part of the wave function, it is useful to rewrite the two-particle +local energy in terms of the relative and center-of-mass motion. +Let us denote the distance between the two electrons as +\( r_{12} \). We omit the center-of-mass motion since we are only interested in the case when +\( r_{12} \rightarrow 0 \). The contribution from the center-of-mass (CoM) variable \( {\bf R}_{\mathrm{CoM}} \) +gives only a finite contribution. +We focus only on the terms that are relevant for \( r_{12} \) and for three dimensions. The relevant local energy becomes then +$$ +\lim_{r_{12} \rightarrow 0}E_L(R)= + \frac{1}{{\cal R}_T(r_{12})}\left(2\frac{d^2}{dr_{ij}^2}+\frac{4}{r_{ij}}\frac{d}{dr_{ij}}+ +\frac{2}{r_{ij}}-\frac{l(l+1)}{r_{ij}^2}+2E +\right){\cal R}_T(r_{12}) = 0. +$$ + +Set \( l=0 \) and we have the so-called cusp condition +$$ +\frac{d {\cal R}_T(r_{12})}{dr_{12}} = -\frac{1}{2(l+1)} +{\cal R}_T(r_{12})\qquad r_{12}\to 0 +$$ + +

+ + +

Resulting ansatz

+The above results in +$$ +{\cal R}_T \propto \exp{(r_{ij}/2)}, +$$ + +for anti-parallel spins and +$$ +{\cal R}_T \propto \exp{(r_{ij}/4)}, +$$ + +for anti-parallel spins. +This is the so-called cusp condition for the relative motion, resulting in a minimal requirement +for the correlation part of the wave fuction. +For general systems containing more than say two electrons, we have this +condition for each electron pair \( ij \). + +

+ + +

The VMC code

+ +

+ + +

# Importing various packages
+from math import exp, sqrt
+from random import random, seed
+import numpy as np
+import matplotlib.pyplot as plt
+from mpl_toolkits.mplot3d import Axes3D
+from matplotlib import cm
+from matplotlib.ticker import LinearLocator, FormatStrFormatter
+import sys
+
+#Trial wave function for quantum dots in two dims
+def WaveFunction(r,alpha,beta):
+    r1 = r[0,0]**2 + r[0,1]**2
+    r2 = r[1,0]**2 + r[1,1]**2
+    r12 = sqrt((r[0,0]-r[1,0])**2 + (r[0,1]-r[1,1])**2)
+    deno = r12/(1+beta*r12)
+    return exp(-0.5*alpha*(r1+r2)+deno)
+
+#Local energy  for quantum dots in two dims, using analytical local energy
+def LocalEnergy(r,alpha,beta):
+    
+    r1 = (r[0,0]**2 + r[0,1]**2)
+    r2 = (r[1,0]**2 + r[1,1]**2)
+    r12 = sqrt((r[0,0]-r[1,0])**2 + (r[0,1]-r[1,1])**2)
+    deno = 1.0/(1+beta*r12)
+    deno2 = deno*deno
+    return 0.5*(1-alpha*alpha)*(r1 + r2) +2.0*alpha + 1.0/r12+deno2*(alpha*r12-deno2+2*beta*deno-1.0/r12)
+
+# The Monte Carlo sampling with the Metropolis algo
+def MonteCarloSampling():
+
+    NumberMCcycles= 100000
+    StepSize = 1.0
+    # positions
+    PositionOld = np.zeros((NumberParticles,Dimension), np.double)
+    PositionNew = np.zeros((NumberParticles,Dimension), np.double)
+    # seed for rng generator
+    seed()
+    # start variational parameter
+    alpha = 0.9
+    for ia in range(MaxVariations):
+        alpha += .025
+        AlphaValues[ia] = alpha
+        beta = 0.2 
+        for jb in range(MaxVariations):
+            beta += .01
+            BetaValues[jb] = beta
+            energy = energy2 = 0.0
+            DeltaE = 0.0
+            #Initial position
+            for i in range(NumberParticles):
+                for j in range(Dimension):
+                    PositionOld[i,j] = StepSize * (random() - .5)
+            wfold = WaveFunction(PositionOld,alpha,beta)
+
+            #Loop over MC MCcycles
+            for MCcycle in range(NumberMCcycles):
+                #Trial position
+                for i in range(NumberParticles):
+                    for j in range(Dimension):
+                        PositionNew[i,j] = PositionOld[i,j] + StepSize * (random() - .5)
+                wfnew = WaveFunction(PositionNew,alpha,beta)
+
+                #Metropolis test to see whether we accept the move
+                if random() < wfnew**2 / wfold**2:
+                   PositionOld = PositionNew.copy()
+                   wfold = wfnew
+                   DeltaE = LocalEnergy(PositionOld,alpha,beta)
+                energy += DeltaE
+                energy2 += DeltaE**2
+
+            #We calculate mean, variance and error ...
+            energy /= NumberMCcycles
+            energy2 /= NumberMCcycles
+            variance = energy2 - energy**2
+            error = sqrt(variance/NumberMCcycles)
+            Energies[ia,jb] = energy    
+    return Energies, AlphaValues, BetaValues
+
+
+#Here starts the main program with variable declarations
+NumberParticles = 2
+Dimension = 2
+MaxVariations = 10
+Energies = np.zeros((MaxVariations,MaxVariations))
+AlphaValues = np.zeros(MaxVariations)
+BetaValues = np.zeros(MaxVariations)
+(Energies, AlphaValues, BetaValues) = MonteCarloSampling()
+
+# Prepare for plots
+fig = plt.figure()
+ax = fig.gca(projection='3d')
+# Plot the surface.
+X, Y = np.meshgrid(AlphaValues, BetaValues)
+surf = ax.plot_surface(X, Y, Energies,cmap=cm.coolwarm,linewidth=0, antialiased=False)
+# Customize the z axis.
+zmin = np.matrix(Energies).min()
+zmax = np.matrix(Energies).max()
+ax.set_zlim(zmin, zmax)
+ax.set_xlabel(r'$\alpha$')
+ax.set_ylabel(r'$\beta$')
+ax.set_zlabel(r'$\langle E \rangle$')
+ax.zaxis.set_major_locator(LinearLocator(10))
+ax.zaxis.set_major_formatter(FormatStrFormatter('%.02f'))
+# Add a color bar which maps values to colors.
+fig.colorbar(surf, shrink=0.5, aspect=5)
+plt.show()
+
+

+ + +

Technical aspect, improvements and how to define the cost function

+
+
+

+ +

+The above procedure is not the smartest one. Looping over all variational parameters becomes expensive. +Also, we don't use importance sampling and optimizations of the standard deviation (blocking, bootstrap, jackknife). +Such codes are included in the above Github address. + +

+We can also be smarter and use minimization methods to find the optimal variational parameters with fewer Monte Carlo cycles and then +fire up our heavy artillery. + +

+One way to achieve this is to minimize the energy as function of the variational parameters. + +

+

+
+ + +

+ + +

Energy derivatives

+
+
+

+To find the derivatives of the local energy expectation value as function of the variational parameters, we can use the chain rule and the hermiticity of the Hamiltonian. + +

+Let us define +$$ +\bar{E}_{\alpha_i}=\frac{d\langle E_L\rangle}{d\alpha_i}. +$$ + +as the derivative of the energy with respect to the variational parameter \( \alpha_i \) +We define also the derivative of the trial function (skipping the subindex \( T \)) as +$$ +\bar{\Psi}_{i}=\frac{d\Psi}{d\alpha_i}. +$$ +

+
+ + +

+ + +

Derivatives of the local energy

+
+
+

+The elements of the gradient of the local energy are then (using the chain rule and the hermiticity of the Hamiltonian) +$$ +\bar{E}_{i}= 2\left( \langle \frac{\bar{\Psi}_{i}}{\Psi}E_L\rangle -\langle \frac{\bar{\Psi}_{i}}{\Psi}\rangle\langle E_L \rangle\right). +$$ + +From a computational point of view it means that you need to compute the expectation values of +$$ +\langle \frac{\bar{\Psi}_{i}}{\Psi}E_L\rangle, +$$ + +and +$$ +\langle \frac{\bar{\Psi}_{i}}{\Psi}\rangle\langle E_L\rangle +$$ + +These integrals are evaluted using MC intergration (with all its possible error sources). +We can then use methods like stochastic gradient or other minimization methods to find the optimal variational parameters (I don't discuss this topic here, but these methods are very important in ML). +

+
+ + +

+ + +

How do we define our cost function?

+
+
+

+We have a model, our likelihood function. + +

+How should we define the cost function? + +

+

+
+ + +

+ + +

Meet the variance and its derivatives

+
+
+

+Suppose the trial function (our model) is the exact wave function. The action of the hamiltionan +on the wave function +$$ + H\Psi = \mathrm{constant}\times \Psi, +$$ + +The integral which defines various +expectation values involving moments of the hamiltonian becomes then +$$ + \langle E^n \rangle = \langle H^n \rangle = + \frac{\int d\boldsymbol{R}\Psi^{\ast}(\boldsymbol{R})H^n(\boldsymbol{R})\Psi(\boldsymbol{R})} + {\int d\boldsymbol{R}\Psi^{\ast}(\boldsymbol{R})\Psi(\boldsymbol{R})}= +\mathrm{constant}\times\frac{\int d\boldsymbol{R}\Psi^{\ast}(\boldsymbol{R})\Psi(\boldsymbol{R})} + {\int d\boldsymbol{R}\Psi^{\ast}(\boldsymbol{R})\Psi(\boldsymbol{R})}=\mathrm{constant}. +$$ + +This gives an important information: If I want the variance, the exact wave function leads to zero variance! +The variance is defined as +$$ +\sigma_E = \langle E^2\rangle - \langle E\rangle^2. +$$ + +Variation is then performed by minimizing both the energy and the variance. + +

+

+
+ + +

+ + +

The variance defines the cost function

+
+
+

+We can then take the derivatives of +$$ +\sigma_E = \langle E^2\rangle - \langle E\rangle^2, +$$ + +with respect to the variational parameters. The derivatives of the variance can then be used to defined the +so-called Hessian matrix, which in turn allows us to use minimization methods like Newton's method or +standard gradient methods. + +

+This leads to however a more complicated expression, with obvious errors when evaluating integrals by Monte Carlo integration. Less used, see however Filippi and Umrigar. The expression becomes complicated +$$ +\bar{E}_{ij} = 2\left[ \langle (\frac{\bar{\Psi}_{ij}}{\Psi}+\frac{\bar{\Psi}_{j}}{\Psi}\frac{\bar{\Psi}_{i}}{\Psi})(E_L-\langle E\rangle)\rangle -\langle \frac{\bar{\Psi}_{i}}{\Psi}\rangle\bar{E}_j-\langle \frac{\bar{\Psi}_{j}}{\Psi}\rangle\bar{E}_i\right] +\langle \frac{\bar{\Psi}_{i}}{\Psi}E_L{_j}\rangle +\langle \frac{\bar{\Psi}_{j}}{\Psi}E_L{_i}\rangle -\langle \frac{\bar{\Psi}_{i}}{\Psi}\rangle\langle E_L{_j}\rangle \langle \frac{\bar{\Psi}_{j}}{\Psi}\rangle\langle E_L{_i}\rangle. +$$ + +

+Evaluating the cost function means having to evaluate the above second derivative of the energy. + +

+

+
+ + +

+ + +

Boltzmann Machines

+ +

+Why use a generative model rather than the more well known discriminative deep neural networks (DNN)? + +

+ + + +

Some similarities and differences from DNNs

+ +
    +
  1. Both use gradient-descent based learning procedures for minimizing cost functions
    2. +
  2. Energy based models don't use backpropagation and automatic differentiation for computing gradients, instead turning to Markov Chain Monte Carlo methods.
    3. +
  3. DNNs often have several hidden layers. A restricted Boltzmann machine has only one hidden layer, however several RBMs can be stacked to make up Deep Belief Networks, of which they constitute the building blocks.
    4. +
+ +History: The RBM was developed by amongst others Geoffrey Hinton, called by some the "Godfather of Deep Learning", working with the University of Toronto and Google. + +

+ + +

The structure of the RBM network

+ +

+



+ +

+ + +

The network

+ +

+The network layers: + +

    +
  1. A function \( \mathbf{x} \) that represents the visible layer, a vector of \( M \) elements (nodes). This layer represents both what the RBM might be given as training input, and what we want it to be able to reconstruct. This might for example be the pixels of an image, the spin values of the Ising model, or coefficients representing speech.
    2. +
  2. The function \( \mathbf{h} \) represents the hidden, or latent, layer. A vector of \( N \) elements (nodes). Also called "feature detectors".
    3. +
+ + + +

Goals

+ +

+The goal of the hidden layer is to increase the model's expressive power. We encode complex interactions between visible variables by introducing additional, hidden variables that interact with visible degrees of freedom in a simple manner, yet still reproduce the complex correlations between visible degrees in the data once marginalized over (integrated out). + +

+Examples of this trick being employed in physics: + +

    +
  1. The Hubbard-Stratonovich transformation
    2. +
  2. The introduction of ghost fields in gauge theory
    3. +
  3. Shadow wave functions in Quantum Monte Carlo simulations
    4. +
+ +The network parameters, to be optimized/learned: + +
    +
  1. \( \mathbf{a} \) represents the visible bias, a vector of same length as \( \mathbf{x} \).
    2. +
  2. \( \mathbf{b} \) represents the hidden bias, a vector of same lenght as \( \mathbf{h} \).
    3. +
  3. \( W \) represents the interaction weights, a matrix of size \( M\times N \).
    4. +
+ + + +

Joint distribution and the Energy function

+The restricted Boltzmann machine is described by a Bolztmann distribution +$$ +\begin{align} + P_{rbm}(\mathbf{x},\mathbf{h}) = \frac{1}{Z} e^{-\frac{1}{T_0}E(\mathbf{x},\mathbf{h})}, +\label{_auto1} +\end{align} +$$ + +where \( Z \) is the normalization constant or partition function, defined as +$$ +\begin{align} + Z = \int \int e^{-\frac{1}{T_0}E(\mathbf{x},\mathbf{h})} d\mathbf{x} d\mathbf{h}. +\label{_auto2} +\end{align} +$$ + +It is common to ignore \( T_0 \) by setting it to one. + +

+ + +

Network Elements

+ +

+The function \( E(\mathbf{x},\mathbf{h}) \) gives the energy of a +configuration (pair of vectors) \( (\mathbf{x}, \mathbf{h}) \). The lower +the energy of a configuration, the higher the probability of it. This +function also depends on the parameters \( \mathbf{a} \), \( \mathbf{b} \) and +\( W \). Thus, when we adjust them during the learning procedure, we are +adjusting the energy function to best fit our problem. + +

+ + +

Defining different types of RBMs

+There are different variants of RBMs, and the differences lie in the types of visible and hidden units we choose as well as in the implementation of the energy function \( E(\mathbf{x},\mathbf{h}) \). + +

+

+
+

+ +

+RBMs were first developed using binary units in both the visible and hidden layer. The corresponding energy function is defined as follows: +$$ +\begin{align} + E(\mathbf{x}, \mathbf{h}) = - \sum_i^M x_i a_i- \sum_j^N b_j h_j - \sum_{i,j}^{M,N} x_i w_{ij} h_j, +\label{_auto3} +\end{align} +$$ + +where the binary values taken on by the nodes are most commonly 0 and 1. +

+
+ +
+
+

+ +

+Another varient is the RBM where the visible units are Gaussian while the hidden units remain binary: +$$ +\begin{align} + E(\mathbf{x}, \mathbf{h}) = \sum_i^M \frac{(x_i - a_i)^2}{2\sigma_i^2} - \sum_j^N b_j h_j - \sum_{i,j}^{M,N} \frac{x_i w_{ij} h_j}{\sigma_i^2}. +\label{_auto4} +\end{align} +$$ +

+
+ + +

+ + +

More about RBMs

+ +
    +
  1. Useful when we model continuous data (i.e., we wish \( \mathbf{x} \) to be continuous)
    2. +
  2. Requires a smaller learning rate, since there's no upper bound to the value a component might take in the reconstruction
    3. +
+ +Other types of units include: + +
    +
  1. Softmax and multinomial units
    2. +
  2. Gaussian visible and hidden units
    3. +
  3. Binomial units
    4. +
  4. Rectified linear units
    5. +
+ + + +

Sampling: Metropolis sampling

+In order to sample from the RBM probability distribution it is common to use Markov Chain Monte Carlo (MCMC) algorithms such as Metropolis-Hastings or Gibbs sampling. + +

+Metropolis sampling starts by suggesting a new configuration \( \boldsymbol{x}^{k+1} \). In the brute force method this is done by some random change of the visible units. The new configuration is then accepted with the acceptance probability +$$ +\begin{align} + A(\boldsymbol{x}^k \rightarrow \boldsymbol{x}^{k+1}) = \text{min} (1, \frac{P(\boldsymbol{x}^{k+1})}{P(\boldsymbol{x}^k)}), +\label{_auto5} +\end{align} +$$ + +where we need the marginalized probability +$$ +\begin{align} + P(\boldsymbol{x}) &= \sum_\mathbf{h} P_{rbm}(\mathbf{x}, \mathbf{h}) +\label{_auto6}\\ + &= \frac{1}{Z}\sum_\mathbf{h} e^{-E(\mathbf{x}, \mathbf{h})}. +\label{_auto7} +\end{align} +$$ + +

+ + +

Sampling: Gibbs sampling

+ +

+In this method we sample from the joint probability \( P_{rbm} (\mathbf{x}, \mathbf{h}) \) by way of a two step sampling process. We alternately update the visible and hidden units. +New samples are generated according to the conditional probabilities \( P(x_i|\mathbf{h}) \) and \( P(h_j|\mathbf{x}) \) respectively and accepted with the probability of \( 1 \). While the the visible nodes are dependent on the hidden nodes and vice versa, the nodes are independent of other nodes within the same layer. This is due to there being no intra layer interactions in the restricted Boltzmann machine. + +

+The conditional probabilities are often referred to as the activitation functions in the neural networks context due to their role in determining the node outputs. For the binary-binary RBM they are +$$ +\begin{align} + P(h_j = 1 | \boldsymbol{x}) &= \frac{1}{1 + e^{-b_j - \sum_i x_i w_{ij}}} +\label{_auto8}\\ + P(x_i = 1 | \boldsymbol{h}) &= \frac{1}{1 + e^{-a_j - \sum_j h_j w_{ij}}}, +\label{_auto9} +\end{align} +$$ + +where we recognize the logistic sigmoid function \( \sigma (x) = 1/(1+exp(-x)) \). + +

+ + +

Gaussian RBM

+For the Gaussian-Binary RBM the conditional probabilities are +$$ +\begin{align} + P(x_i|\mathbf{h}) &= \mathcal{N}(x_i; a_i+ \sum_j h_j w_{ij}, \sigma^2) +\label{_auto10}\\ + P(h_j=1|\mathbf{x}) &= \frac{1}{1+e^{-b_j-\frac{1}{\sigma^2} \sum_i x_i w_{ij}}}, +\label{_auto11} +\end{align} +$$ + +while the visible units now follow a normal distribution, we see the hidden units again follow the logistic sigmoid function. + +

+ + +

Cost function

+ +

+When working with a training dataset, the most common training approach is maximizing the log-likelihood of the training data. The log likelihood characterizes the log-probability of generating the observed data using our generative model. Using this method our cost function is chosen as the negative log-likelihood. The learning then consists of trying to find parameters that maximize the probability of the dataset, and is known as Maximum Likelihood Estimation (MLE). +Denoting the parameters as \( \boldsymbol{\theta} = a_1,...,a_M,b_1,...,b_N,w_{11},...,w_{MN} \), the log-likelihood is given by +$$ +\begin{align} + \mathcal{L}(\{ \theta_i \}) &= \langle \text{log} P_\theta(\boldsymbol{x}) \rangle_{data} +\label{_auto12}\\ + &= - \langle E(\boldsymbol{x}; \{ \theta_i\}) \rangle_{data} - \text{log} Z(\{ \theta_i\}), +\label{_auto13} +\end{align} +$$ + +where we used that the normalization constant does not depend on the data, \( \langle \text{log} Z(\{ \theta_i\}) \rangle = \text{log} Z(\{ \theta_i\}) \) +Our cost function is the negative log-likelihood, \( \mathcal{C}(\{ \theta_i \}) = - \mathcal{L}(\{ \theta_i \}) \) + +

+ + +

Optimization / Training

+The training procedure of choice often is Stochastic Gradient Descent (SGD). It consists of a series of iterations where we update the parameters according to the equation +$$ +\begin{align} + \boldsymbol{\theta}_{k+1} = \boldsymbol{\theta}_k - \eta \nabla \mathcal{C} (\boldsymbol{\theta}_k) +\label{_auto14} +\end{align} +$$ + +at each \( k \)-th iteration. There are a range of variants of the algorithm which aim at making the learning rate \( \eta \) more adaptive so the method might be more efficient while remaining stable. + +

+We now need the gradient of the cost function in order to minimize it. We find that +$$ +\begin{align} + \frac{\partial \mathcal{C}(\{ \theta_i\})}{\partial \theta_i} + &= \langle \frac{\partial E(\boldsymbol{x}; \theta_i)}{\partial \theta_i} \rangle_{data} + + \frac{\partial \text{log} Z(\{ \theta_i\})}{\partial \theta_i} +\label{_auto15}\\ + &= \langle O_i(\boldsymbol{x}) \rangle_{data} - \langle O_i(\boldsymbol{x}) \rangle_{model}, +\label{_auto16} +\end{align} +$$ + +where in order to simplify notation we defined the "operator" +$$ +\begin{align} + O_i(\boldsymbol{x}) = \frac{\partial E(\boldsymbol{x}; \theta_i)}{\partial \theta_i}, +\label{_auto17} +\end{align} +$$ + +and used the statistical mechanics relationship between expectation values and the log-partition function: +$$ +\begin{align} + \langle O_i(\boldsymbol{x}) \rangle_{model} = \text{Tr} P_\theta(\boldsymbol{x})O_i(\boldsymbol{x}) = - \frac{\partial \text{log} Z(\{ \theta_i\})}{\partial \theta_i}. +\label{_auto18} +\end{align} +$$ + +

+ + +

More on RBMs

+ +

+The data-dependent term in the gradient is known as the positive phase of the gradient, while the model-dependent term is known as the negative phase of the gradient. The aim of the training is to lower the energy of configurations that are near observed data points (increasing their probability), and raising the energy of configurations that are far from observed data points (decreasing their probability). + +

+The gradient of the negative log-likelihood cost function of a Binary-Binary RBM is then +$$ +\begin{align} + \frac{\partial \mathcal{C} (w_{ij}, a_i, b_j)}{\partial w_{ij}} =& \langle x_i h_j \rangle_{data} - \langle x_i h_j \rangle_{model} +\label{_auto19}\\ + \frac{\partial \mathcal{C} (w_{ij}, a_i, b_j)}{\partial a_{ij}} =& \langle x_i \rangle_{data} - \langle x_i \rangle_{model} +\label{_auto20}\\ + \frac{\partial \mathcal{C} (w_{ij}, a_i, b_j)}{\partial b_{ij}} =& \langle h_i \rangle_{data} - \langle h_i \rangle_{model}. +\label{_auto21}\\ +\label{_auto22} +\end{align} +$$ + +To get the expecation values with respect to the data, we set the visible units to each of the observed samples in the training data, then update the hidden units according to the conditional probability found before. We then average over all samples in the training data to calculate expectation values with respect to the data. + +

+ + +

Which sampling to use

+ +

+To get the expectation values with respect to the model, we use Gibbs sampling. We can either initialize the \( \boldsymbol{x} \) randomly or with a training sample. While we ideally want a large number of Gibbs iterations \( n\rightarrow n \), one might decide to truncate it earlier for efficiency. Doing this while having intialized \( \boldsymbol{x} \) with a training data vector is referred to as contrastive divergence (CD), because one is then closer to approximating the gradient of this function than the negative log-likelihood. The contrastive divergence function is the difference between two Kullback-Leibler divergences (also called relative entropy), which measure how one probability distribution diverges from a second, expected probability distribution (in this case the estimated one from the ground truth one). + +

+ + +

RBMs for the quantum many body problem

+ +

+The idea of applying RBMs to quantum many body problems was presented by G. Carleo and M. Troyer, working with ETH Zurich and Microsoft Research. + +

+Some of their motivation included + +

+ + + +

Choose the right RBM

+ +

+Carleo and Troyer applied the RBM to the quantum mechanical spin lattice systems of the Ising model and Heisenberg model, with encouraging results. Our goal is to test the method on systems of moving particles. For the spin lattice systems it was natural to use a binary-binary RBM, with the nodes taking values of 1 and -1. For moving particles, on the other hand, we want the visible nodes to be continuous, representing position coordinates. Thus, we start by choosing a Gaussian-binary RBM, where the visible nodes are continuous and hidden nodes take on values of 0 or 1. If eventually we would like the hidden nodes to be continuous as well the rectified linear units seem like the most relevant choice. + +

+ + +

Representing the wave function

+The wavefunction should be a probability amplitude depending on \( \boldsymbol{x} \). The RBM model is given by the joint distribution of \( \boldsymbol{x} \) and \( \boldsymbol{h} \) +$$ +\begin{align} + F_{rbm}(\mathbf{x},\mathbf{h}) = \frac{1}{Z} e^{-\frac{1}{T_0}E(\mathbf{x},\mathbf{h})}. +\label{_auto23} +\end{align} +$$ + +To find the marginal distribution of \( \boldsymbol{x} \) we set: +$$ +\begin{align} + F_{rbm}(\mathbf{x}) &= \sum_\mathbf{h} F_{rbm}(\mathbf{x}, \mathbf{h}) +\label{_auto24}\\ + &= \frac{1}{Z}\sum_\mathbf{h} e^{-E(\mathbf{x}, \mathbf{h})}. +\label{_auto25} +\end{align} +$$ + +Now this is what we use to represent the wave function, calling it a neural-network quantum state (NQS) +$$ +\begin{align} + \Psi (\mathbf{X}) &= F_{rbm}(\mathbf{x}) +\label{_auto26}\\ + &= \frac{1}{Z}\sum_{\boldsymbol{h}} e^{-E(\mathbf{x}, \mathbf{h})} +\label{_auto27}\\ + &= \frac{1}{Z} \sum_{\{h_j\}} e^{-\sum_i^M \frac{(x_i - a_i)^2}{2\sigma^2} + \sum_j^N b_j h_j + \sum_{i,j}^{M,N} \frac{x_i w_{ij} h_j}{\sigma^2}} +\label{_auto28}\\ + &= \frac{1}{Z} e^{-\sum_i^M \frac{(x_i - a_i)^2}{2\sigma^2}} \prod_j^N (1 + e^{b_j + \sum_i^M \frac{x_i w_{ij}}{\sigma^2}}). +\label{_auto29}\\ +\label{_auto30} +\end{align} +$$ + +

+ + +

Choose the cost function

+Now we don't necessarily have training data (unless we generate it by using some other method). However, what we do have is the variational principle which allows us to obtain the ground state wave function by minimizing the expectation value of the energy of a trial wavefunction (corresponding to the untrained NQS). Similarly to the traditional variational Monte Carlo method then, it is the local energy we wish to minimize. The gradient to use for the stochastic gradient descent procedure is +$$ +\begin{align} + G_i = \frac{\partial \langle E_L \rangle}{\partial \theta_i} + = 2(\langle E_L \frac{1}{\Psi}\frac{\partial \Psi}{\partial \theta_i} \rangle - \langle E_L \rangle \langle \frac{1}{\Psi}\frac{\partial \Psi}{\partial \theta_i} \rangle ), +\label{_auto31} +\end{align} +$$ + +where the local energy is given by +$$ +\begin{align} + E_L = \frac{1}{\Psi} \hat{\mathbf{H}} \Psi. +\label{_auto32} +\end{align} +$$ + +

+ + +

Running the codes

+
+
+

+You can find the codes for the simple two-electron case at the Github repository https://github.com/mhjensenseminars/MachineLearningTalk/tree/master/doc/Programs/MLcpp/src. Python codes to come, only c++ as of now. + +

+The trial wave function is based on the product of a Slater determinant with Gaussian orbitals, a simple Jastrow factor \( \exp{(r_{ij})} \) and the reduced Boltzmann machines. + +

+The Broyden-Fletcher-Goldfarb-Shanno algorithm was used to perform the minimization. We used \( 14 \) hidden nodes in the calculations below. + +

+

+
+ + +

+ + +

Energy as function of iterations, \( N=2 \) electrons

+
+
+

+



+
+
+ + +

+ + +

Energy as function of iterations, \( N=6 \) electrons

+
+
+

+



+
+
+ + +

+ + +

Conclusions and where do we stand

+
+
+

+ +

    +
  • A simple extension of the work of G. Carleo and M. Troyer, Science 355, Issue 6325, pp. 602-606 (2017) gives excellent results for two-electron systems as well as good agreement with standard VMC calculations for \( N=6 \) and \( N=12 \) electrons.
    • +
  • Minimization problem can be tricky.
    • +
  • Anti-symmetry dealt with multiplying the trail wave function with an optimized Slater determinant.
    • +
  • To come: Analysis of wave function from ML and compare with diffusion and Variational Monte Carlo calculations as well as the analytical results of Taut for the two-electron case.
    • +
  • Extend to more fermions. How do we deal with the antisymmetry of the multi-fermion wave function? + +
      +
    1. Here we used standard Hartree-Fock theory to define an optimal Slater determinant. Takes care of the antisymmetry. What about constructing an anti-symmetrized network function?
      2. +
    2. Use thereafter ML to determine the correlated part of the wafe function (including a standard Jastrow factor).
      3. +
    3. Test this for multi-fermion systems and compare with other many-body methods.
      4. +
    + +
  • Can we use ML to find out which correlations are relevant and thereby diminish the dimensionality problem in say CC or SRG theories?
    • +
+
+
+ + +

+ + + +

+ + + + + + + +
+ © 1999-2018, Morten Hjorth-Jensen. Released under CC Attribution-NonCommercial 4.0 license +
+ + + + + + diff --git a/doc/pub/BM/html/BM-reveal.html b/doc/pub/BM/html/BM-reveal.html new file mode 100644 index 000000000..d84d03de9 --- /dev/null +++ b/doc/pub/BM/html/BM-reveal.html @@ -0,0 +1,1681 @@ +\ + + + + + + +Solving quantum mechanical problems with Machine Learning and Boltzmann machines + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
+ + + +
+ + + + + + + + + + + + + + +
+ + + + +

Solving quantum mechanical problems with Machine Learning and Boltzmann machines

+ +

+ + +

+Morten Hjorth-Jensen +
+ +

+ + +

Department of Physics and Astronomy and National Superconducting Cyclotron Laboratory, Michigan State University and Department of Physics, University of Oslo, Norway
+
+

 
+

Jul 26, 2018

+
+

+ +

+ © 1999-2018, Morten Hjorth-Jensen. Released under CC Attribution-NonCommercial 4.0 license +
+
+ + +
+

What is this talk about?

+
+ +

+The main aim is to give you a short and pedestrian introduction to how we can use Machine Learning methods +to solve quantum mechanical many-body problems. And why this could be of interest. + +

+The hope is that after this talk you have gotten the basic ideas to get you started. Peeping into https://github.com/mhjensenseminars/MachineLearningTalk, you'll find a Jupyter notebook, slides, codes etc that will allow you to reproduce the simulations discussed here, and perhaps run your own very first calculations. + +

+Furthermore, since we will also talk about Monte Carlo simulations, the exciting idea of Monte Carlo optimization and Machine Learning can be used to mesh strategies for reducing the curse of dimensionality. + + +

+
+ + +
+

Why?

+ +

+How can we avoid the dimensionality curse? Many possibilities + +

    +

  1. smarter basis functions
    2. +

  2. resummation of specific correlations
    3. +

  3. stochastic sampling of high-lying states (stochastic FCI, CC and soon SRG)
    4. +

  4. many more
    5. +
+

+ +Machine Learning and Quantum Computing hold also great promise in tackling the +ever increasing dimensionalities. Here we will focus on Machine Learning. +

+ + +
+

Overview

+
+ +
    +

  • Short intro to Machine Learning
    • +

  • Variational Monte Carlo (Markov Chain Monte Carlo, \( \mathrm{MC}^2 \)) and two-electron quantum dots, solving quantum mechanical problems in a stochastic way. It will serve as our motivation for switching to Machine Learning.
    • +

  • From Variational Monte Carlo to Boltzmann Machines and Machine Learning
    • +
+
+
+ + +
+

What are the Machine Learning calculations here based on?

+ +

+This work is inspired by the idea of representing the wave function with +a restricted Boltzmann machine (RBM), presented recently by G. Carleo and M. Troyer, Science 355, Issue 6325, pp. 602-606 (2017). They +named such a wave function/network a neural network quantum state (NQS). In their article they apply it to the quantum mechanical +spin lattice systems of the Ising model and Heisenberg model, with +encouraging results. + +

+Thanks to Vilde Flugsrud and Alfred Alocias Mariadason for many discussions and interpretations of the above paper. +

+ + +
+

A new world

+
+ +

+Machine learning (ML) is an extremely rich field, in spite of its young age. The +increases we have seen during the last three decades in computational +capabilities have been followed by developments of methods and +techniques for analyzing and handling large date sets, relying heavily +on statistics, computer science and mathematics. The field is rather +new and developing rapidly. + +

+Popular software packages written in Python for ML are + +

+

+ +These are all freely available at their respective GitHub sites. They +encompass communities of developers in the thousands or more. And the number +of code developers and contributors keeps increasing. +

+
+ + +
+

Lots of room for creativity

+
+ +

+Not all the +algorithms and methods can be given a rigorous mathematical +justification, opening up thereby for experimenting +and trial and error and thereby exciting new developments. +

+ +

+

+ +

+A solid command of linear algebra, multivariate theory, +probability theory, statistical data analysis, +understanding errors and Monte Carlo methods is important in order to understand many of the +various algorithms and methods. +

+ +

+

+Job market, a personal statement. +

+A familiarity with ML is almost becoming a prerequisite for many of the most exciting employment +opportunities. Add quantum computing and there you are. +

+
+ + +
+

Some members of the ML family

+ +

+

+ +
    +

  1. Linear regression and its variants, in essence polynomial regression
    2. +

  2. Decision tree algorithms, from simpler to more complex ones like random forests
    3. +

  3. Nearest neighbors models
    4. +

  4. Bayesian statistics and regression
    5. +

  5. Support vector machines and finally various variants of
    6. +

  6. Artifical neural networks and deep learning
    7. +

  7. ....
    8. +
+
+
+ + +
+

What are the basic ingredients?

+
+ +

+Almost every problem in ML and data science starts with the same ingredients: + +

    +

  • The dataset \( \mathbf{x} \) (could be some observable quantity of the system we are studying)
    • +

  • A model which is a function of a set of parameters \( \mathbf{\alpha} \) that relates to the dataset, say a likelihood function \( p(\mathbf{x}\vert \mathbf{\alpha}) \) or just a simple model \( f(\mathbf{\alpha}) \)
    • +

  • A so-called cost function \( \mathcal{C} (\mathbf{x}, f(\mathbf{\alpha})) \) which allows us to decide how well our model represents the dataset.
    • +
+

+ +We seek to minimize the function \( \mathcal{C} (\mathbf{x}, f(\mathbf{\alpha})) \) by finding the parameter values which minimize \( \mathcal{C} \). This leads to various minimization algorithms. +

+
+ + +
+

What is Machine Learning?

+
+ +

+Machine learning is the science of giving computers the ability to +learn without being explicitly programmed. The idea is that there +exist generic algorithms which can be used to find patterns in a broad +class of data sets without having to write code specifically for each +problem. The algorithm will build its own logic based on the data. + +

+Machine learning is a subfield of computer science, and is closely +related to computational statistics. It evolved from the study of +pattern recognition in artificial intelligence (AI) research, and has +made contributions to AI tasks like computer vision, natural language +processing and speech recognition. It has also, especially in later +years, found applications in a wide variety of other areas, including +bioinformatics, economy, physics, finance and marketing. + +

+You will notice however that many of the basic ideas discussed do come from Physics! +

+
+ + +
+

Types of Machine Learning

+ +

+

+ +

+The approaches to machine learning are many, but are often split into two main categories. +In supervised learning we know the answer to a problem, +and let the computer deduce the logic behind it. On the other hand, unsupervised learning +is a method for finding patterns and relationship in data sets without any prior knowledge of the system. +Some authours also operate with a third category, namely reinforcement learning. This is a paradigm +of learning inspired by behavioural psychology, where learning is achieved by trial-and-error, +solely from rewards and punishment. + +

+Another way to categorize machine learning tasks is to consider the desired output of a system. +Some of the most common tasks are: + +

    + +

  • Classification: Outputs are divided into two or more classes. The goal is to produce a model that assigns inputs into one of these classes. An example is to identify digits based on pictures of hand-written ones. Classification is typically supervised learning.
    • + +

  • Regression: Finding a functional relationship between an input data set and a reference data set. The goal is to construct a function that maps input data to continuous output values.
    • + +

  • Clustering: Data are divided into groups with certain common traits, without knowing the different groups beforehand. It is thus a form of unsupervised learning.
    • +
+
+
+ + +
+

References

+
+ + +
+
+ + +
+

Another interesting article

+ +

+Here we will use so-called reduced Boltzmann Machines to simulate quantum many-body problems. For Monte Carlo aficionados, there is a very close similarity with what are called shadow wave functions, see the work of Pederiva and Kalos and collaborators, Phys Rev. E 90, 053304 (2014). +

+ + +
+

Just for fun illustration of scikit-learn: Hudson bay data

+
+ +

+Most mammalian predators rely on a variety of prey, which complicates mathematical modeling; however, a few predators have become highly specialized and seek almost exclusively a single prey species. An example of this simplified predator-prey interaction is seen in Canadian northern forests, where the populations of the lynx and the snowshoe hare are intertwined in a life and death struggle. + +

+One reason that this particular system has been so extensively studied is that the Hudson Bay company kept careful records of all furs from the early 1800s into the 1900s. The records for the furs collected by the Hudson Bay company showed distinct oscillations (approximately 12 year periods), suggesting that these species caused almost periodic fluctuations of each other's populations. The table here shows data from 1900 to 1920. + + +

+
+ + +
+

Hares and lynx in Hudson bay from 1900 to 1920

+ +

+



+
+ + +
+

Linear regression in Python

+
+ +

+ + +

import numpy as np
+import matplotlib.pyplot as plt
+from IPython.display import display
+import sklearn
+from sklearn.linear_model import LinearRegression
+from sklearn.tree import DecisionTreeRegressor
+
+
+data = np.loadtxt('src/Hudson_Bay.csv', delimiter=',', skiprows=1)
+x = data[:,0]
+y = data[:,1]
+line = np.linspace(1900,1920,1000,endpoint=False).reshape(-1,1)
+reg = DecisionTreeRegressor(min_samples_split=3).fit(x.reshape(-1,1),y.reshape(-1,1))
+plt.plot(line, reg.predict(line), label="decision tree")
+regline = LinearRegression().fit(x.reshape(-1,1),y.reshape(-1,1))
+plt.plot(line, regline.predict(line), label= "Linear Regression")
+plt.plot(x, y, label= "Linear Regression")
+plt.show()
+
+ +
+
+ + +
+

Artificial neurons

+
+ +

+The field of artificial neural networks has a long history of +development, and is closely connected with the advancement of computer +science and computers in general. A model of artificial neurons was +first developed by McCulloch and Pitts in 1943 to study signal +processing in the brain and has later been refined by others. The +general idea is to mimic neural networks in the human brain, which is +composed of billions of neurons that communicate with each other by +sending electrical signals. Each neuron accumulates its incoming +signals, which must exceed an activation threshold to yield an +output. If the threshold is not overcome, the neuron remains inactive, +i.e. has zero output. + +

+This behaviour has inspired a simple mathematical model for an artificial neuron. + +

 
+$$ + y = f\left(\sum_{i=1}^n w_ix_i\right) = f(u) +$$ +

 
+ +Here, the output \( y \) of the neuron is the value of its activation function, which have as input +a weighted sum of signals \( x_i, \dots ,x_n \) received by \( n \) other neurons. +

+
+ + +
+

A simple perceptron model

+ +

+



+
+ + +
+

Neural network types

+
+ +

+An artificial neural network (NN), is a computational model that consists of layers of connected neurons, or nodes. +It is supposed to mimic a biological nervous system by letting each neuron interact with other neurons +by sending signals in the form of mathematical functions between layers. +A wide variety of different NNs have +been developed, but most of them consist of an input layer, an output layer and eventual layers in-between, called +hidden layers. All layers can contain an arbitrary number of nodes, and each connection between two nodes +is associated with a weight variable. + +

+



+
+
+ + +
+

The system: two electrons in a harmonic oscillator trap in two dimensions

+ +

+The Hamiltonian of the quantum dot is given by +

 
+$$ \hat{H} = \hat{H}_0 + \hat{V}, +$$ +

 
+ +where \( \hat{H}_0 \) is the many-body HO Hamiltonian, and \( \hat{V} \) is the +inter-electron Coulomb interactions. In dimensionless units, +

 
+$$ \hat{V}= \sum_{i < j}^N \frac{1}{r_{ij}}, +$$ +

 
+ +with \( r_{ij}=\sqrt{\mathbf{r}_i^2 - \mathbf{r}_j^2} \). + +

+This leads to the separable Hamiltonian, with the relative motion part given by (\( r_{ij}=r \)) +

 
+$$ +\hat{H}_r=-\nabla^2_r + \frac{1}{4}\omega^2r^2+ \frac{1}{r}, +$$ +

 
+ +plus a standard Harmonic Oscillator problem for the center-of-mass motion. +This system has analytical solutions in two and three dimensions (M. Taut 1993 and 1994). +

+ + +
+

Quantum Monte Carlo Motivation

+
+ +

+Given a hamiltonian \( H \) and a trial wave function \( \Psi_T \), the variational principle states that the expectation value of \( \langle H \rangle \), defined through +

 
+$$ + \langle E \rangle = + \frac{\int d\boldsymbol{R}\Psi^{\ast}_T(\boldsymbol{R})H(\boldsymbol{R})\Psi_T(\boldsymbol{R})} + {\int d\boldsymbol{R}\Psi^{\ast}_T(\boldsymbol{R})\Psi_T(\boldsymbol{R})}, +$$ +

 
+ +is an upper bound to the ground state energy \( E_0 \) of the hamiltonian \( H \), that is +

 
+$$ + E_0 \le \langle H \rangle . +$$ +

 
+ +In general, the integrals involved in the calculation of various expectation values are multi-dimensional ones. Traditional integration methods such as the Gauss-Legendre will not be adequate for say the computation of the energy of a many-body system. +

+
+ + +
+

Quantum Monte Carlo Motivation

+
+Basic steps. +

+Choose a trial wave function +\( \psi_T(\boldsymbol{R}) \). +

 
+$$ + P(\boldsymbol{R},\boldsymbol{\alpha})= \frac{\left|\psi_T(\boldsymbol{R},\boldsymbol{\alpha})\right|^2}{\int \left|\psi_T(\boldsymbol{R},\boldsymbol{\alpha})\right|^2d\boldsymbol{R}}. +$$ +

 
+ +This is our model, or likelihood/probability distribution function (PDF). It depends on some variational parameters \( \boldsymbol{\alpha} \). +The approximation to the expectation value of the Hamiltonian is now +

 
+$$ + \langle E[\boldsymbol{\alpha}] \rangle = + \frac{\int d\boldsymbol{R}\Psi^{\ast}_T(\boldsymbol{R},\boldsymbol{\alpha})H(\boldsymbol{R})\Psi_T(\boldsymbol{R},\boldsymbol{\alpha})} + {\int d\boldsymbol{R}\Psi^{\ast}_T(\boldsymbol{R},\boldsymbol{\alpha})\Psi_T(\boldsymbol{R},\boldsymbol{\alpha})}. +$$ +

 
+

+
+ + +
+

Quantum Monte Carlo Motivation

+
+Define a new quantity. +

 
+$$ + E_L(\boldsymbol{R},\boldsymbol{\alpha})=\frac{1}{\psi_T(\boldsymbol{R},\boldsymbol{\alpha})}H\psi_T(\boldsymbol{R},\boldsymbol{\alpha}), +$$ +

 
+ +called the local energy, which, together with our trial PDF yields +

 
+$$ + E[\boldsymbol{\alpha}]=\int P(\boldsymbol{R})E_L(\boldsymbol{R},\boldsymbol{\alpha}) d\boldsymbol{R}\approx \frac{1}{N}\sum_{i=1}^NE_L(\boldsymbol{R_i},\boldsymbol{\alpha}) +$$ +

 
+ +with \( N \) being the number of Monte Carlo samples. +

+
+ + +
+

Quantum Monte Carlo

+
+ +

+The Algorithm for performing a variational Monte Carlo calculations runs thus as this + +

    + +

  • Initialisation: Fix the number of Monte Carlo steps. Choose an initial \( \boldsymbol{R} \) and variational parameters \( \alpha \) and calculate \( \left|\psi_T(\boldsymbol{R},\boldsymbol{\alpha})\right|^2 \).
    • + +

  • Initialise the energy and the variance and start the Monte Carlo calculation by looping over trials.
    • + +
      + +

    • Calculate a trial position \( \boldsymbol{R}_p=\boldsymbol{R}+r*step \) where \( r \) is a random variable \( r \in [0,1] \).
      • + +

    • Metropolis algorithm to accept or reject this move \( w = P(\boldsymbol{R}_p,\boldsymbol{\alpha})/P(\boldsymbol{R},\boldsymbol{\alpha}) \).
      • + +

    • If the step is accepted, then we set \( \boldsymbol{R}=\boldsymbol{R}_p \).
      • + +

    • Update averages
      • +
    +

  • Finish and compute final averages.
    • +
+

+ +Observe that the jumping in space is governed by the variable step. This is often called brute-force sampling. +Need importance sampling to get more relevant sampling. +

+
+ + +
+

The trial wave function

+
+ +

+We want to perform a Variational Monte Carlo calculation of the ground state of two electrons in a quantum dot well with different oscillator energies, assuming total spin \( S=0 \). +Our trial wave function has the following form +

 
+$$ +\begin{equation} + \psi_{T}(\boldsymbol{r}_1,\boldsymbol{r}_2) = + C\exp{\left(-\alpha_1\omega(r_1^2+r_2^2)/2\right)} + \exp{\left(\frac{r_{12}}{(1+\alpha_2 r_{12})}\right)}, +\tag{1} +\end{equation} +$$ +

 
+ +where the $\alpha$s represent our variational parameters, two in this case. + +

+Why does the trial function look like this? How did we get there? This will be our main motivation for switching to +Machine Learning. + + +

+
+ + +
+

The correlation part of the wave function

+ +

+To find an ansatz for the correlated part of the wave function, it is useful to rewrite the two-particle +local energy in terms of the relative and center-of-mass motion. +Let us denote the distance between the two electrons as +\( r_{12} \). We omit the center-of-mass motion since we are only interested in the case when +\( r_{12} \rightarrow 0 \). The contribution from the center-of-mass (CoM) variable \( {\bf R}_{\mathrm{CoM}} \) +gives only a finite contribution. +We focus only on the terms that are relevant for \( r_{12} \) and for three dimensions. The relevant local energy becomes then +

 
+$$ +\lim_{r_{12} \rightarrow 0}E_L(R)= + \frac{1}{{\cal R}_T(r_{12})}\left(2\frac{d^2}{dr_{ij}^2}+\frac{4}{r_{ij}}\frac{d}{dr_{ij}}+ +\frac{2}{r_{ij}}-\frac{l(l+1)}{r_{ij}^2}+2E +\right){\cal R}_T(r_{12}) = 0. +$$ +

 
+ +Set \( l=0 \) and we have the so-called cusp condition +

 
+$$ +\frac{d {\cal R}_T(r_{12})}{dr_{12}} = -\frac{1}{2(l+1)} +{\cal R}_T(r_{12})\qquad r_{12}\to 0 +$$ +

 
+

+ + +
+

Resulting ansatz

+The above results in +

 
+$$ +{\cal R}_T \propto \exp{(r_{ij}/2)}, +$$ +

 
+ +for anti-parallel spins and +

 
+$$ +{\cal R}_T \propto \exp{(r_{ij}/4)}, +$$ +

 
+ +for anti-parallel spins. +This is the so-called cusp condition for the relative motion, resulting in a minimal requirement +for the correlation part of the wave fuction. +For general systems containing more than say two electrons, we have this +condition for each electron pair \( ij \). +

+ + +
+

The VMC code

+ +

+ + +

# Importing various packages
+from math import exp, sqrt
+from random import random, seed
+import numpy as np
+import matplotlib.pyplot as plt
+from mpl_toolkits.mplot3d import Axes3D
+from matplotlib import cm
+from matplotlib.ticker import LinearLocator, FormatStrFormatter
+import sys
+
+#Trial wave function for quantum dots in two dims
+def WaveFunction(r,alpha,beta):
+    r1 = r[0,0]**2 + r[0,1]**2
+    r2 = r[1,0]**2 + r[1,1]**2
+    r12 = sqrt((r[0,0]-r[1,0])**2 + (r[0,1]-r[1,1])**2)
+    deno = r12/(1+beta*r12)
+    return exp(-0.5*alpha*(r1+r2)+deno)
+
+#Local energy  for quantum dots in two dims, using analytical local energy
+def LocalEnergy(r,alpha,beta):
+    
+    r1 = (r[0,0]**2 + r[0,1]**2)
+    r2 = (r[1,0]**2 + r[1,1]**2)
+    r12 = sqrt((r[0,0]-r[1,0])**2 + (r[0,1]-r[1,1])**2)
+    deno = 1.0/(1+beta*r12)
+    deno2 = deno*deno
+    return 0.5*(1-alpha*alpha)*(r1 + r2) +2.0*alpha + 1.0/r12+deno2*(alpha*r12-deno2+2*beta*deno-1.0/r12)
+
+# The Monte Carlo sampling with the Metropolis algo
+def MonteCarloSampling():
+
+    NumberMCcycles= 100000
+    StepSize = 1.0
+    # positions
+    PositionOld = np.zeros((NumberParticles,Dimension), np.double)
+    PositionNew = np.zeros((NumberParticles,Dimension), np.double)
+    # seed for rng generator
+    seed()
+    # start variational parameter
+    alpha = 0.9
+    for ia in range(MaxVariations):
+        alpha += .025
+        AlphaValues[ia] = alpha
+        beta = 0.2 
+        for jb in range(MaxVariations):
+            beta += .01
+            BetaValues[jb] = beta
+            energy = energy2 = 0.0
+            DeltaE = 0.0
+            #Initial position
+            for i in range(NumberParticles):
+                for j in range(Dimension):
+                    PositionOld[i,j] = StepSize * (random() - .5)
+            wfold = WaveFunction(PositionOld,alpha,beta)
+
+            #Loop over MC MCcycles
+            for MCcycle in range(NumberMCcycles):
+                #Trial position
+                for i in range(NumberParticles):
+                    for j in range(Dimension):
+                        PositionNew[i,j] = PositionOld[i,j] + StepSize * (random() - .5)
+                wfnew = WaveFunction(PositionNew,alpha,beta)
+
+                #Metropolis test to see whether we accept the move
+                if random() < wfnew**2 / wfold**2:
+                   PositionOld = PositionNew.copy()
+                   wfold = wfnew
+                   DeltaE = LocalEnergy(PositionOld,alpha,beta)
+                energy += DeltaE
+                energy2 += DeltaE**2
+
+            #We calculate mean, variance and error ...
+            energy /= NumberMCcycles
+            energy2 /= NumberMCcycles
+            variance = energy2 - energy**2
+            error = sqrt(variance/NumberMCcycles)
+            Energies[ia,jb] = energy    
+    return Energies, AlphaValues, BetaValues
+
+
+#Here starts the main program with variable declarations
+NumberParticles = 2
+Dimension = 2
+MaxVariations = 10
+Energies = np.zeros((MaxVariations,MaxVariations))
+AlphaValues = np.zeros(MaxVariations)
+BetaValues = np.zeros(MaxVariations)
+(Energies, AlphaValues, BetaValues) = MonteCarloSampling()
+
+# Prepare for plots
+fig = plt.figure()
+ax = fig.gca(projection='3d')
+# Plot the surface.
+X, Y = np.meshgrid(AlphaValues, BetaValues)
+surf = ax.plot_surface(X, Y, Energies,cmap=cm.coolwarm,linewidth=0, antialiased=False)
+# Customize the z axis.
+zmin = np.matrix(Energies).min()
+zmax = np.matrix(Energies).max()
+ax.set_zlim(zmin, zmax)
+ax.set_xlabel(r'$\alpha$')
+ax.set_ylabel(r'$\beta$')
+ax.set_zlabel(r'$\langle E \rangle$')
+ax.zaxis.set_major_locator(LinearLocator(10))
+ax.zaxis.set_major_formatter(FormatStrFormatter('%.02f'))
+# Add a color bar which maps values to colors.
+fig.colorbar(surf, shrink=0.5, aspect=5)
+plt.show()
+
+
+ + +
+

Technical aspect, improvements and how to define the cost function

+
+ +

+The above procedure is not the smartest one. Looping over all variational parameters becomes expensive. +Also, we don't use importance sampling and optimizations of the standard deviation (blocking, bootstrap, jackknife). +Such codes are included in the above Github address. + +

+We can also be smarter and use minimization methods to find the optimal variational parameters with fewer Monte Carlo cycles and then +fire up our heavy artillery. + +

+One way to achieve this is to minimize the energy as function of the variational parameters. + + +

+
+ + +
+

Energy derivatives

+
+ +

+To find the derivatives of the local energy expectation value as function of the variational parameters, we can use the chain rule and the hermiticity of the Hamiltonian. + +

+Let us define +

 
+$$ +\bar{E}_{\alpha_i}=\frac{d\langle E_L\rangle}{d\alpha_i}. +$$ +

 
+ +as the derivative of the energy with respect to the variational parameter \( \alpha_i \) +We define also the derivative of the trial function (skipping the subindex \( T \)) as +

 
+$$ +\bar{\Psi}_{i}=\frac{d\Psi}{d\alpha_i}. +$$ +

 
+

+
+ + +
+

Derivatives of the local energy

+
+ +

+The elements of the gradient of the local energy are then (using the chain rule and the hermiticity of the Hamiltonian) +

 
+$$ +\bar{E}_{i}= 2\left( \langle \frac{\bar{\Psi}_{i}}{\Psi}E_L\rangle -\langle \frac{\bar{\Psi}_{i}}{\Psi}\rangle\langle E_L \rangle\right). +$$ +

 
+ +From a computational point of view it means that you need to compute the expectation values of +

 
+$$ +\langle \frac{\bar{\Psi}_{i}}{\Psi}E_L\rangle, +$$ +

 
+ +and +

 
+$$ +\langle \frac{\bar{\Psi}_{i}}{\Psi}\rangle\langle E_L\rangle +$$ +

 
+ +These integrals are evaluted using MC intergration (with all its possible error sources). +We can then use methods like stochastic gradient or other minimization methods to find the optimal variational parameters (I don't discuss this topic here, but these methods are very important in ML). +

+
+ + +
+

How do we define our cost function?

+
+ +

+We have a model, our likelihood function. + +

+How should we define the cost function? + + +

+
+ + +
+

Meet the variance and its derivatives

+
+Why the variance? +

+Suppose the trial function (our model) is the exact wave function. The action of the hamiltionan +on the wave function +

 
+$$ + H\Psi = \mathrm{constant}\times \Psi, +$$ +

 
+ +The integral which defines various +expectation values involving moments of the hamiltonian becomes then +

 
+$$ + \langle E^n \rangle = \langle H^n \rangle = + \frac{\int d\boldsymbol{R}\Psi^{\ast}(\boldsymbol{R})H^n(\boldsymbol{R})\Psi(\boldsymbol{R})} + {\int d\boldsymbol{R}\Psi^{\ast}(\boldsymbol{R})\Psi(\boldsymbol{R})}= +\mathrm{constant}\times\frac{\int d\boldsymbol{R}\Psi^{\ast}(\boldsymbol{R})\Psi(\boldsymbol{R})} + {\int d\boldsymbol{R}\Psi^{\ast}(\boldsymbol{R})\Psi(\boldsymbol{R})}=\mathrm{constant}. +$$ +

 
+ +This gives an important information: If I want the variance, the exact wave function leads to zero variance! +The variance is defined as +

 
+$$ +\sigma_E = \langle E^2\rangle - \langle E\rangle^2. +$$ +

 
+ +Variation is then performed by minimizing both the energy and the variance. + + +

+
+ + +
+

The variance defines the cost function

+
+ +

+We can then take the derivatives of +

 
+$$ +\sigma_E = \langle E^2\rangle - \langle E\rangle^2, +$$ +

 
+ +with respect to the variational parameters. The derivatives of the variance can then be used to defined the +so-called Hessian matrix, which in turn allows us to use minimization methods like Newton's method or +standard gradient methods. + +

+This leads to however a more complicated expression, with obvious errors when evaluating integrals by Monte Carlo integration. Less used, see however Filippi and Umrigar. The expression becomes complicated +

 
+$$ +\bar{E}_{ij} = 2\left[ \langle (\frac{\bar{\Psi}_{ij}}{\Psi}+\frac{\bar{\Psi}_{j}}{\Psi}\frac{\bar{\Psi}_{i}}{\Psi})(E_L-\langle E\rangle)\rangle -\langle \frac{\bar{\Psi}_{i}}{\Psi}\rangle\bar{E}_j-\langle \frac{\bar{\Psi}_{j}}{\Psi}\rangle\bar{E}_i\right] +\langle \frac{\bar{\Psi}_{i}}{\Psi}E_L{_j}\rangle +\langle \frac{\bar{\Psi}_{j}}{\Psi}E_L{_i}\rangle -\langle \frac{\bar{\Psi}_{i}}{\Psi}\rangle\langle E_L{_j}\rangle \langle \frac{\bar{\Psi}_{j}}{\Psi}\rangle\langle E_L{_i}\rangle. +$$ +

 
+ +

+Evaluating the cost function means having to evaluate the above second derivative of the energy. + + +

+
+ + +
+

Boltzmann Machines

+ +

+Why use a generative model rather than the more well known discriminative deep neural networks (DNN)? + +

    +

  • Discriminitave methods have several limitations: They are mainly supervised learning methods, thus requiring labeled data. And there are tasks they cannot accomplish, like drawing new examples from an unknown probability distribution.
    • +

  • A generative model can learn to represent and sample from a probability distribution. The core idea is to learn a parametric model of the probability distribution from which the training data was drawn. As an example + +
      +

    1. A model for images could learn to draw new examples of cats and dogs, given a training dataset of images of cats and dogs.
      2. +

    2. Generate a sample of an ordered or disordered Ising model phase, having been given samples of such phases.
      3. +

    3. Model the trial wave function for Monte Carlo calculations
      4. +
    +

    +

+
+ + +
+

Some similarities and differences from DNNs

+ +
    +

  1. Both use gradient-descent based learning procedures for minimizing cost functions
    2. +

  2. Energy based models don't use backpropagation and automatic differentiation for computing gradients, instead turning to Markov Chain Monte Carlo methods.
    3. +

  3. DNNs often have several hidden layers. A restricted Boltzmann machine has only one hidden layer, however several RBMs can be stacked to make up Deep Belief Networks, of which they constitute the building blocks.
    4. +
+

+ +History: The RBM was developed by amongst others Geoffrey Hinton, called by some the "Godfather of Deep Learning", working with the University of Toronto and Google. +

+ + +
+

The structure of the RBM network

+ +

+



+
+ + +
+

The network

+ +

+The network layers: + +

    +

  1. A function \( \mathbf{x} \) that represents the visible layer, a vector of \( M \) elements (nodes). This layer represents both what the RBM might be given as training input, and what we want it to be able to reconstruct. This might for example be the pixels of an image, the spin values of the Ising model, or coefficients representing speech.
    2. +

  2. The function \( \mathbf{h} \) represents the hidden, or latent, layer. A vector of \( N \) elements (nodes). Also called "feature detectors".
    3. +
+
+ + +
+

Goals

+ +

+The goal of the hidden layer is to increase the model's expressive power. We encode complex interactions between visible variables by introducing additional, hidden variables that interact with visible degrees of freedom in a simple manner, yet still reproduce the complex correlations between visible degrees in the data once marginalized over (integrated out). + +

+Examples of this trick being employed in physics: + +

    +

  1. The Hubbard-Stratonovich transformation
    2. +

  2. The introduction of ghost fields in gauge theory
    3. +

  3. Shadow wave functions in Quantum Monte Carlo simulations
    4. +
+

+ +The network parameters, to be optimized/learned: + +

    +

  1. \( \mathbf{a} \) represents the visible bias, a vector of same length as \( \mathbf{x} \).
    2. +

  2. \( \mathbf{b} \) represents the hidden bias, a vector of same lenght as \( \mathbf{h} \).
    3. +

  3. \( W \) represents the interaction weights, a matrix of size \( M\times N \).
    4. +
+
+ + +
+

Joint distribution and the Energy function

+The restricted Boltzmann machine is described by a Bolztmann distribution +

 
+$$ +\begin{align} + P_{rbm}(\mathbf{x},\mathbf{h}) = \frac{1}{Z} e^{-\frac{1}{T_0}E(\mathbf{x},\mathbf{h})}, +\tag{2} +\end{align} +$$ +

 
+ +where \( Z \) is the normalization constant or partition function, defined as +

 
+$$ +\begin{align} + Z = \int \int e^{-\frac{1}{T_0}E(\mathbf{x},\mathbf{h})} d\mathbf{x} d\mathbf{h}. +\tag{3} +\end{align} +$$ +

 
+ +It is common to ignore \( T_0 \) by setting it to one. +

+ + +
+

Network Elements

+ +

+The function \( E(\mathbf{x},\mathbf{h}) \) gives the energy of a +configuration (pair of vectors) \( (\mathbf{x}, \mathbf{h}) \). The lower +the energy of a configuration, the higher the probability of it. This +function also depends on the parameters \( \mathbf{a} \), \( \mathbf{b} \) and +\( W \). Thus, when we adjust them during the learning procedure, we are +adjusting the energy function to best fit our problem. +

+ + +
+

Defining different types of RBMs

+There are different variants of RBMs, and the differences lie in the types of visible and hidden units we choose as well as in the implementation of the energy function \( E(\mathbf{x},\mathbf{h}) \). + +

+

+Binary-Binary RBM: +

+RBMs were first developed using binary units in both the visible and hidden layer. The corresponding energy function is defined as follows: +

 
+$$ +\begin{align} + E(\mathbf{x}, \mathbf{h}) = - \sum_i^M x_i a_i- \sum_j^N b_j h_j - \sum_{i,j}^{M,N} x_i w_{ij} h_j, +\tag{4} +\end{align} +$$ +

 
+ +where the binary values taken on by the nodes are most commonly 0 and 1. +

+ +
+Gaussian-Binary RBM: +

+Another varient is the RBM where the visible units are Gaussian while the hidden units remain binary: +

 
+$$ +\begin{align} + E(\mathbf{x}, \mathbf{h}) = \sum_i^M \frac{(x_i - a_i)^2}{2\sigma_i^2} - \sum_j^N b_j h_j - \sum_{i,j}^{M,N} \frac{x_i w_{ij} h_j}{\sigma_i^2}. +\tag{5} +\end{align} +$$ +

 
+

+
+ + +
+

More about RBMs

+ +
    +

  1. Useful when we model continuous data (i.e., we wish \( \mathbf{x} \) to be continuous)
    2. +

  2. Requires a smaller learning rate, since there's no upper bound to the value a component might take in the reconstruction
    3. +
+

+ +Other types of units include: + +

    +

  1. Softmax and multinomial units
    2. +

  2. Gaussian visible and hidden units
    3. +

  3. Binomial units
    4. +

  4. Rectified linear units
    5. +
+
+ + +
+

Sampling: Metropolis sampling

+In order to sample from the RBM probability distribution it is common to use Markov Chain Monte Carlo (MCMC) algorithms such as Metropolis-Hastings or Gibbs sampling. + +

+Metropolis sampling starts by suggesting a new configuration \( \boldsymbol{x}^{k+1} \). In the brute force method this is done by some random change of the visible units. The new configuration is then accepted with the acceptance probability +

 
+$$ +\begin{align} + A(\boldsymbol{x}^k \rightarrow \boldsymbol{x}^{k+1}) = \text{min} (1, \frac{P(\boldsymbol{x}^{k+1})}{P(\boldsymbol{x}^k)}), +\tag{6} +\end{align} +$$ +

 
+ +where we need the marginalized probability +

 
+$$ +\begin{align} + P(\boldsymbol{x}) &= \sum_\mathbf{h} P_{rbm}(\mathbf{x}, \mathbf{h}) +\tag{7}\\ + &= \frac{1}{Z}\sum_\mathbf{h} e^{-E(\mathbf{x}, \mathbf{h})}. +\tag{8} +\end{align} +$$ +

 
+

+ + +
+

Sampling: Gibbs sampling

+ +

+In this method we sample from the joint probability \( P_{rbm} (\mathbf{x}, \mathbf{h}) \) by way of a two step sampling process. We alternately update the visible and hidden units. +New samples are generated according to the conditional probabilities \( P(x_i|\mathbf{h}) \) and \( P(h_j|\mathbf{x}) \) respectively and accepted with the probability of \( 1 \). While the the visible nodes are dependent on the hidden nodes and vice versa, the nodes are independent of other nodes within the same layer. This is due to there being no intra layer interactions in the restricted Boltzmann machine. + +

+The conditional probabilities are often referred to as the activitation functions in the neural networks context due to their role in determining the node outputs. For the binary-binary RBM they are +

 
+$$ +\begin{align} + P(h_j = 1 | \boldsymbol{x}) &= \frac{1}{1 + e^{-b_j - \sum_i x_i w_{ij}}} +\tag{9}\\ + P(x_i = 1 | \boldsymbol{h}) &= \frac{1}{1 + e^{-a_j - \sum_j h_j w_{ij}}}, +\tag{10} +\end{align} +$$ +

 
+ +where we recognize the logistic sigmoid function \( \sigma (x) = 1/(1+exp(-x)) \). +

+ + +
+

Gaussian RBM

+For the Gaussian-Binary RBM the conditional probabilities are +

 
+$$ +\begin{align} + P(x_i|\mathbf{h}) &= \mathcal{N}(x_i; a_i+ \sum_j h_j w_{ij}, \sigma^2) +\tag{11}\\ + P(h_j=1|\mathbf{x}) &= \frac{1}{1+e^{-b_j-\frac{1}{\sigma^2} \sum_i x_i w_{ij}}}, +\tag{12} +\end{align} +$$ +

 
+ +while the visible units now follow a normal distribution, we see the hidden units again follow the logistic sigmoid function. +

+ + +
+

Cost function

+ +

+When working with a training dataset, the most common training approach is maximizing the log-likelihood of the training data. The log likelihood characterizes the log-probability of generating the observed data using our generative model. Using this method our cost function is chosen as the negative log-likelihood. The learning then consists of trying to find parameters that maximize the probability of the dataset, and is known as Maximum Likelihood Estimation (MLE). +Denoting the parameters as \( \boldsymbol{\theta} = a_1,...,a_M,b_1,...,b_N,w_{11},...,w_{MN} \), the log-likelihood is given by +

 
+$$ +\begin{align} + \mathcal{L}(\{ \theta_i \}) &= \langle \text{log} P_\theta(\boldsymbol{x}) \rangle_{data} +\tag{13}\\ + &= - \langle E(\boldsymbol{x}; \{ \theta_i\}) \rangle_{data} - \text{log} Z(\{ \theta_i\}), +\tag{14} +\end{align} +$$ +

 
+ +where we used that the normalization constant does not depend on the data, \( \langle \text{log} Z(\{ \theta_i\}) \rangle = \text{log} Z(\{ \theta_i\}) \) +Our cost function is the negative log-likelihood, \( \mathcal{C}(\{ \theta_i \}) = - \mathcal{L}(\{ \theta_i \}) \) +

+ + +
+

Optimization / Training

+The training procedure of choice often is Stochastic Gradient Descent (SGD). It consists of a series of iterations where we update the parameters according to the equation +

 
+$$ +\begin{align} + \boldsymbol{\theta}_{k+1} = \boldsymbol{\theta}_k - \eta \nabla \mathcal{C} (\boldsymbol{\theta}_k) +\tag{15} +\end{align} +$$ +

 
+ +at each \( k \)-th iteration. There are a range of variants of the algorithm which aim at making the learning rate \( \eta \) more adaptive so the method might be more efficient while remaining stable. + +

+We now need the gradient of the cost function in order to minimize it. We find that +

 
+$$ +\begin{align} + \frac{\partial \mathcal{C}(\{ \theta_i\})}{\partial \theta_i} + &= \langle \frac{\partial E(\boldsymbol{x}; \theta_i)}{\partial \theta_i} \rangle_{data} + + \frac{\partial \text{log} Z(\{ \theta_i\})}{\partial \theta_i} +\tag{16}\\ + &= \langle O_i(\boldsymbol{x}) \rangle_{data} - \langle O_i(\boldsymbol{x}) \rangle_{model}, +\tag{17} +\end{align} +$$ +

 
+ +where in order to simplify notation we defined the "operator" +

 
+$$ +\begin{align} + O_i(\boldsymbol{x}) = \frac{\partial E(\boldsymbol{x}; \theta_i)}{\partial \theta_i}, +\tag{18} +\end{align} +$$ +

 
+ +and used the statistical mechanics relationship between expectation values and the log-partition function: +

 
+$$ +\begin{align} + \langle O_i(\boldsymbol{x}) \rangle_{model} = \text{Tr} P_\theta(\boldsymbol{x})O_i(\boldsymbol{x}) = - \frac{\partial \text{log} Z(\{ \theta_i\})}{\partial \theta_i}. +\tag{19} +\end{align} +$$ +

 
+

+ + +
+

More on RBMs

+ +

+The data-dependent term in the gradient is known as the positive phase of the gradient, while the model-dependent term is known as the negative phase of the gradient. The aim of the training is to lower the energy of configurations that are near observed data points (increasing their probability), and raising the energy of configurations that are far from observed data points (decreasing their probability). + +

+The gradient of the negative log-likelihood cost function of a Binary-Binary RBM is then +

 
+$$ +\begin{align} + \frac{\partial \mathcal{C} (w_{ij}, a_i, b_j)}{\partial w_{ij}} =& \langle x_i h_j \rangle_{data} - \langle x_i h_j \rangle_{model} +\tag{20}\\ + \frac{\partial \mathcal{C} (w_{ij}, a_i, b_j)}{\partial a_{ij}} =& \langle x_i \rangle_{data} - \langle x_i \rangle_{model} +\tag{21}\\ + \frac{\partial \mathcal{C} (w_{ij}, a_i, b_j)}{\partial b_{ij}} =& \langle h_i \rangle_{data} - \langle h_i \rangle_{model}. +\tag{22}\\ +\tag{23} +\end{align} +$$ +

 
+ +To get the expecation values with respect to the data, we set the visible units to each of the observed samples in the training data, then update the hidden units according to the conditional probability found before. We then average over all samples in the training data to calculate expectation values with respect to the data. +

+ + +
+

Which sampling to use

+ +

+To get the expectation values with respect to the model, we use Gibbs sampling. We can either initialize the \( \boldsymbol{x} \) randomly or with a training sample. While we ideally want a large number of Gibbs iterations \( n\rightarrow n \), one might decide to truncate it earlier for efficiency. Doing this while having intialized \( \boldsymbol{x} \) with a training data vector is referred to as contrastive divergence (CD), because one is then closer to approximating the gradient of this function than the negative log-likelihood. The contrastive divergence function is the difference between two Kullback-Leibler divergences (also called relative entropy), which measure how one probability distribution diverges from a second, expected probability distribution (in this case the estimated one from the ground truth one). +

+ + +
+

RBMs for the quantum many body problem

+ +

+The idea of applying RBMs to quantum many body problems was presented by G. Carleo and M. Troyer, working with ETH Zurich and Microsoft Research. + +

+Some of their motivation included + +

    +

  • "The wave function \( \Psi \) is a monolithic mathematical quantity that contains all the information on a quantum state, be it a single particle or a complex molecule. In principle, an exponential amount of information is needed to fully encode a generic many-body quantum state."
    • +

  • There are still interesting open problems, including fundamental questions ranging from the dynamical properties of high-dimensional systems to the exact ground-state properties of strongly interacting fermions.
    • +

  • The difficulty lies in finding a general strategy to reduce the exponential complexity of the full many-body wave function down to its most essential features. That is + +
      +

    1. \( \rightarrow \) Dimensional reduction
      2. +

    2. \( \rightarrow \) Feature extraction
      3. +
    +

  • Among the most successful techniques to attack these challenges, artifical neural networks play a prominent role.
    • +

  • Want to understand whether an artifical neural network may adapt to describe a quantum system.
    • +
+
+ + +
+

Choose the right RBM

+ +

+Carleo and Troyer applied the RBM to the quantum mechanical spin lattice systems of the Ising model and Heisenberg model, with encouraging results. Our goal is to test the method on systems of moving particles. For the spin lattice systems it was natural to use a binary-binary RBM, with the nodes taking values of 1 and -1. For moving particles, on the other hand, we want the visible nodes to be continuous, representing position coordinates. Thus, we start by choosing a Gaussian-binary RBM, where the visible nodes are continuous and hidden nodes take on values of 0 or 1. If eventually we would like the hidden nodes to be continuous as well the rectified linear units seem like the most relevant choice. +

+ + +
+

Representing the wave function

+The wavefunction should be a probability amplitude depending on \( \boldsymbol{x} \). The RBM model is given by the joint distribution of \( \boldsymbol{x} \) and \( \boldsymbol{h} \) +

 
+$$ +\begin{align} + F_{rbm}(\mathbf{x},\mathbf{h}) = \frac{1}{Z} e^{-\frac{1}{T_0}E(\mathbf{x},\mathbf{h})}. +\tag{24} +\end{align} +$$ +

 
+ +To find the marginal distribution of \( \boldsymbol{x} \) we set: +

 
+$$ +\begin{align} + F_{rbm}(\mathbf{x}) &= \sum_\mathbf{h} F_{rbm}(\mathbf{x}, \mathbf{h}) +\tag{25}\\ + &= \frac{1}{Z}\sum_\mathbf{h} e^{-E(\mathbf{x}, \mathbf{h})}. +\tag{26} +\end{align} +$$ +

 
+ +Now this is what we use to represent the wave function, calling it a neural-network quantum state (NQS) +

 
+$$ +\begin{align} + \Psi (\mathbf{X}) &= F_{rbm}(\mathbf{x}) +\tag{27}\\ + &= \frac{1}{Z}\sum_{\boldsymbol{h}} e^{-E(\mathbf{x}, \mathbf{h})} +\tag{28}\\ + &= \frac{1}{Z} \sum_{\{h_j\}} e^{-\sum_i^M \frac{(x_i - a_i)^2}{2\sigma^2} + \sum_j^N b_j h_j + \sum_{i,j}^{M,N} \frac{x_i w_{ij} h_j}{\sigma^2}} +\tag{29}\\ + &= \frac{1}{Z} e^{-\sum_i^M \frac{(x_i - a_i)^2}{2\sigma^2}} \prod_j^N (1 + e^{b_j + \sum_i^M \frac{x_i w_{ij}}{\sigma^2}}). +\tag{30}\\ +\tag{31} +\end{align} +$$ +

 
+

+ + +
+

Choose the cost function

+Now we don't necessarily have training data (unless we generate it by using some other method). However, what we do have is the variational principle which allows us to obtain the ground state wave function by minimizing the expectation value of the energy of a trial wavefunction (corresponding to the untrained NQS). Similarly to the traditional variational Monte Carlo method then, it is the local energy we wish to minimize. The gradient to use for the stochastic gradient descent procedure is +

 
+$$ +\begin{align} + G_i = \frac{\partial \langle E_L \rangle}{\partial \theta_i} + = 2(\langle E_L \frac{1}{\Psi}\frac{\partial \Psi}{\partial \theta_i} \rangle - \langle E_L \rangle \langle \frac{1}{\Psi}\frac{\partial \Psi}{\partial \theta_i} \rangle ), +\tag{32} +\end{align} +$$ +

 
+ +where the local energy is given by +

 
+$$ +\begin{align} + E_L = \frac{1}{\Psi} \hat{\mathbf{H}} \Psi. +\tag{33} +\end{align} +$$ +

 
+

+ + +
+

Running the codes

+
+ +

+You can find the codes for the simple two-electron case at the Github repository https://github.com/mhjensenseminars/MachineLearningTalk/tree/master/doc/Programs/MLcpp/src. Python codes to come, only c++ as of now. + +

+The trial wave function is based on the product of a Slater determinant with Gaussian orbitals, a simple Jastrow factor \( \exp{(r_{ij})} \) and the reduced Boltzmann machines. + +

+The Broyden-Fletcher-Goldfarb-Shanno algorithm was used to perform the minimization. We used \( 14 \) hidden nodes in the calculations below. + + +

+
+ + +
+

Energy as function of iterations, \( N=2 \) electrons

+
+ +

+



+
+
+ + +
+

Energy as function of iterations, \( N=6 \) electrons

+
+ +

+



+
+
+ + +
+

Conclusions and where do we stand

+
+ +
    +

  • A simple extension of the work of G. Carleo and M. Troyer, Science 355, Issue 6325, pp. 602-606 (2017) gives excellent results for two-electron systems as well as good agreement with standard VMC calculations for \( N=6 \) and \( N=12 \) electrons.
    • +

  • Minimization problem can be tricky.
    • +

  • Anti-symmetry dealt with multiplying the trail wave function with an optimized Slater determinant.
    • +

  • To come: Analysis of wave function from ML and compare with diffusion and Variational Monte Carlo calculations as well as the analytical results of Taut for the two-electron case.
    • +

  • Extend to more fermions. How do we deal with the antisymmetry of the multi-fermion wave function? + +
      +

    1. Here we used standard Hartree-Fock theory to define an optimal Slater determinant. Takes care of the antisymmetry. What about constructing an anti-symmetrized network function?
      2. +

    2. Use thereafter ML to determine the correlated part of the wafe function (including a standard Jastrow factor).
      3. +

    3. Test this for multi-fermion systems and compare with other many-body methods.
      4. +
    +

  • Can we use ML to find out which correlations are relevant and thereby diminish the dimensionality problem in say CC or SRG theories?
    • +
+
+
+ + + +
+
+ + + + + + + + + + + + diff --git a/doc/pub/BM/html/BM-solarized.html b/doc/pub/BM/html/BM-solarized.html new file mode 100644 index 000000000..cacbfbcfb --- /dev/null +++ b/doc/pub/BM/html/BM-solarized.html @@ -0,0 +1,1527 @@ + + + + + + + +Solving quantum mechanical problems with Machine Learning and Boltzmann machines + + + + + + + + + + + + + + + + + + + + + + + + + + + + +

Solving quantum mechanical problems with Machine Learning and Boltzmann machines

+ +

+ + +

+Morten Hjorth-Jensen +
+ +

+ + +

Department of Physics and Astronomy and National Superconducting Cyclotron Laboratory, Michigan State University and Department of Physics, University of Oslo, Norway
+
+

+

Jul 26, 2018

+
+

+









+ +

What is this talk about?

+
+ +

+The main aim is to give you a short and pedestrian introduction to how we can use Machine Learning methods +to solve quantum mechanical many-body problems. And why this could be of interest. + +

+The hope is that after this talk you have gotten the basic ideas to get you started. Peeping into https://github.com/mhjensenseminars/MachineLearningTalk, you'll find a Jupyter notebook, slides, codes etc that will allow you to reproduce the simulations discussed here, and perhaps run your own very first calculations. + +

+Furthermore, since we will also talk about Monte Carlo simulations, the exciting idea of Monte Carlo optimization and Machine Learning can be used to mesh strategies for reducing the curse of dimensionality. + + +

+ + +

+









+ +

Why?

+ +

+How can we avoid the dimensionality curse? Many possibilities + +

    +
  1. smarter basis functions
    2. +
  2. resummation of specific correlations
    3. +
  3. stochastic sampling of high-lying states (stochastic FCI, CC and soon SRG)
    4. +
  4. many more
    5. +
+ +Machine Learning and Quantum Computing hold also great promise in tackling the +ever increasing dimensionalities. Here we will focus on Machine Learning. + +

+









+ +

Overview

+
+ +

+ +

+
+ + +

+









+ +

What are the Machine Learning calculations here based on?

+ +

+This work is inspired by the idea of representing the wave function with +a restricted Boltzmann machine (RBM), presented recently by G. Carleo and M. Troyer, Science 355, Issue 6325, pp. 602-606 (2017). They +named such a wave function/network a neural network quantum state (NQS). In their article they apply it to the quantum mechanical +spin lattice systems of the Ising model and Heisenberg model, with +encouraging results. + +

+Thanks to Vilde Flugsrud and Alfred Alocias Mariadason for many discussions and interpretations of the above paper. + +

+









+ +

A new world

+
+ +

+Machine learning (ML) is an extremely rich field, in spite of its young age. The +increases we have seen during the last three decades in computational +capabilities have been followed by developments of methods and +techniques for analyzing and handling large date sets, relying heavily +on statistics, computer science and mathematics. The field is rather +new and developing rapidly. + +

+Popular software packages written in Python for ML are + +

+ +These are all freely available at their respective GitHub sites. They +encompass communities of developers in the thousands or more. And the number +of code developers and contributors keeps increasing. +
+ + +

+









+ +

Lots of room for creativity

+
+ +

+Not all the +algorithms and methods can be given a rigorous mathematical +justification, opening up thereby for experimenting +and trial and error and thereby exciting new developments. +

+ + +

+

+ +

+A solid command of linear algebra, multivariate theory, +probability theory, statistical data analysis, +understanding errors and Monte Carlo methods is important in order to understand many of the +various algorithms and methods. +

+ + +

+

+Job market, a personal statement. +

+A familiarity with ML is almost becoming a prerequisite for many of the most exciting employment +opportunities. Add quantum computing and there you are. +

+ + +

+ + +

Some members of the ML family

+ +

+

+ +

+ +

    +
  1. Linear regression and its variants, in essence polynomial regression
    2. +
  2. Decision tree algorithms, from simpler to more complex ones like random forests
    3. +
  3. Nearest neighbors models
    4. +
  4. Bayesian statistics and regression
    5. +
  5. Support vector machines and finally various variants of
    6. +
  6. Artifical neural networks and deep learning
    7. +
  7. ....
    8. +
+
+ + +

+









+ +

What are the basic ingredients?

+
+ +

+Almost every problem in ML and data science starts with the same ingredients: + +

+ +We seek to minimize the function \( \mathcal{C} (\mathbf{x}, f(\mathbf{\alpha})) \) by finding the parameter values which minimize \( \mathcal{C} \). This leads to various minimization algorithms. +
+ + +

+









+ +

What is Machine Learning?

+
+ +

+Machine learning is the science of giving computers the ability to +learn without being explicitly programmed. The idea is that there +exist generic algorithms which can be used to find patterns in a broad +class of data sets without having to write code specifically for each +problem. The algorithm will build its own logic based on the data. + +

+Machine learning is a subfield of computer science, and is closely +related to computational statistics. It evolved from the study of +pattern recognition in artificial intelligence (AI) research, and has +made contributions to AI tasks like computer vision, natural language +processing and speech recognition. It has also, especially in later +years, found applications in a wide variety of other areas, including +bioinformatics, economy, physics, finance and marketing. + +

+You will notice however that many of the basic ideas discussed do come from Physics! +

+ + +

+









+ +

Types of Machine Learning

+ +

+

+ +

+The approaches to machine learning are many, but are often split into two main categories. +In supervised learning we know the answer to a problem, +and let the computer deduce the logic behind it. On the other hand, unsupervised learning +is a method for finding patterns and relationship in data sets without any prior knowledge of the system. +Some authours also operate with a third category, namely reinforcement learning. This is a paradigm +of learning inspired by behavioural psychology, where learning is achieved by trial-and-error, +solely from rewards and punishment. + +

+Another way to categorize machine learning tasks is to consider the desired output of a system. +Some of the most common tasks are: + +

+
+ + +

+









+ +

References

+
+ +

+ +

+
+ + +

+









+ +

Another interesting article

+ +

+Here we will use so-called reduced Boltzmann Machines to simulate quantum many-body problems. For Monte Carlo aficionados, there is a very close similarity with what are called shadow wave functions, see the work of Pederiva and Kalos and collaborators, Phys Rev. E 90, 053304 (2014). + +

+









+ +

Just for fun illustration of scikit-learn: Hudson bay data

+
+ +

+Most mammalian predators rely on a variety of prey, which complicates mathematical modeling; however, a few predators have become highly specialized and seek almost exclusively a single prey species. An example of this simplified predator-prey interaction is seen in Canadian northern forests, where the populations of the lynx and the snowshoe hare are intertwined in a life and death struggle. + +

+One reason that this particular system has been so extensively studied is that the Hudson Bay company kept careful records of all furs from the early 1800s into the 1900s. The records for the furs collected by the Hudson Bay company showed distinct oscillations (approximately 12 year periods), suggesting that these species caused almost periodic fluctuations of each other's populations. The table here shows data from 1900 to 1920. + + +

+ + +

+









+ +

Hares and lynx in Hudson bay from 1900 to 1920

+ +

+



+ +

+









+ +

Linear regression in Python

+
+ +

+

+ + +

import numpy as np
+import matplotlib.pyplot as plt
+from IPython.display import display
+import sklearn
+from sklearn.linear_model import LinearRegression
+from sklearn.tree import DecisionTreeRegressor
+
+
+data = np.loadtxt('src/Hudson_Bay.csv', delimiter=',', skiprows=1)
+x = data[:,0]
+y = data[:,1]
+line = np.linspace(1900,1920,1000,endpoint=False).reshape(-1,1)
+reg = DecisionTreeRegressor(min_samples_split=3).fit(x.reshape(-1,1),y.reshape(-1,1))
+plt.plot(line, reg.predict(line), label="decision tree")
+regline = LinearRegression().fit(x.reshape(-1,1),y.reshape(-1,1))
+plt.plot(line, regline.predict(line), label= "Linear Regression")
+plt.plot(x, y, label= "Linear Regression")
+plt.show()
+
+ +
+ + +

+









+ +

Artificial neurons

+
+ +

+ +

+The field of artificial neural networks has a long history of +development, and is closely connected with the advancement of computer +science and computers in general. A model of artificial neurons was +first developed by McCulloch and Pitts in 1943 to study signal +processing in the brain and has later been refined by others. The +general idea is to mimic neural networks in the human brain, which is +composed of billions of neurons that communicate with each other by +sending electrical signals. Each neuron accumulates its incoming +signals, which must exceed an activation threshold to yield an +output. If the threshold is not overcome, the neuron remains inactive, +i.e. has zero output. + +

+This behaviour has inspired a simple mathematical model for an artificial neuron. + +$$ + y = f\left(\sum_{i=1}^n w_ix_i\right) = f(u) +$$ + +Here, the output \( y \) of the neuron is the value of its activation function, which have as input +a weighted sum of signals \( x_i, \dots ,x_n \) received by \( n \) other neurons. +

+ + +

+









+ +

A simple perceptron model

+ +

+



+ +

+









+ +

Neural network types

+
+ +

+An artificial neural network (NN), is a computational model that consists of layers of connected neurons, or nodes. +It is supposed to mimic a biological nervous system by letting each neuron interact with other neurons +by sending signals in the form of mathematical functions between layers. +A wide variety of different NNs have +been developed, but most of them consist of an input layer, an output layer and eventual layers in-between, called +hidden layers. All layers can contain an arbitrary number of nodes, and each connection between two nodes +is associated with a weight variable. + +

+



+
+ + +

+









+ +

The system: two electrons in a harmonic oscillator trap in two dimensions

+ +

+The Hamiltonian of the quantum dot is given by +$$ \hat{H} = \hat{H}_0 + \hat{V}, +$$ + +where \( \hat{H}_0 \) is the many-body HO Hamiltonian, and \( \hat{V} \) is the +inter-electron Coulomb interactions. In dimensionless units, +$$ \hat{V}= \sum_{i < j}^N \frac{1}{r_{ij}}, +$$ + +with \( r_{ij}=\sqrt{\mathbf{r}_i^2 - \mathbf{r}_j^2} \). + +

+This leads to the separable Hamiltonian, with the relative motion part given by (\( r_{ij}=r \)) +$$ +\hat{H}_r=-\nabla^2_r + \frac{1}{4}\omega^2r^2+ \frac{1}{r}, +$$ + +plus a standard Harmonic Oscillator problem for the center-of-mass motion. +This system has analytical solutions in two and three dimensions (M. Taut 1993 and 1994). + +

+









+ +

Quantum Monte Carlo Motivation

+
+ +

+Given a hamiltonian \( H \) and a trial wave function \( \Psi_T \), the variational principle states that the expectation value of \( \langle H \rangle \), defined through +$$ + \langle E \rangle = + \frac{\int d\boldsymbol{R}\Psi^{\ast}_T(\boldsymbol{R})H(\boldsymbol{R})\Psi_T(\boldsymbol{R})} + {\int d\boldsymbol{R}\Psi^{\ast}_T(\boldsymbol{R})\Psi_T(\boldsymbol{R})}, +$$ + +is an upper bound to the ground state energy \( E_0 \) of the hamiltonian \( H \), that is +$$ + E_0 \le \langle H \rangle . +$$ + +In general, the integrals involved in the calculation of various expectation values are multi-dimensional ones. Traditional integration methods such as the Gauss-Legendre will not be adequate for say the computation of the energy of a many-body system. +

+ + +

+









+ +

Quantum Monte Carlo Motivation

+
+Basic steps. +

+Choose a trial wave function +\( \psi_T(\boldsymbol{R}) \). +$$ + P(\boldsymbol{R},\boldsymbol{\alpha})= \frac{\left|\psi_T(\boldsymbol{R},\boldsymbol{\alpha})\right|^2}{\int \left|\psi_T(\boldsymbol{R},\boldsymbol{\alpha})\right|^2d\boldsymbol{R}}. +$$ + +This is our model, or likelihood/probability distribution function (PDF). It depends on some variational parameters \( \boldsymbol{\alpha} \). +The approximation to the expectation value of the Hamiltonian is now +$$ + \langle E[\boldsymbol{\alpha}] \rangle = + \frac{\int d\boldsymbol{R}\Psi^{\ast}_T(\boldsymbol{R},\boldsymbol{\alpha})H(\boldsymbol{R})\Psi_T(\boldsymbol{R},\boldsymbol{\alpha})} + {\int d\boldsymbol{R}\Psi^{\ast}_T(\boldsymbol{R},\boldsymbol{\alpha})\Psi_T(\boldsymbol{R},\boldsymbol{\alpha})}. +$$ +

+ + +

+









+ +

Quantum Monte Carlo Motivation

+
+Define a new quantity. +

+$$ + E_L(\boldsymbol{R},\boldsymbol{\alpha})=\frac{1}{\psi_T(\boldsymbol{R},\boldsymbol{\alpha})}H\psi_T(\boldsymbol{R},\boldsymbol{\alpha}), +$$ + +called the local energy, which, together with our trial PDF yields +$$ + E[\boldsymbol{\alpha}]=\int P(\boldsymbol{R})E_L(\boldsymbol{R},\boldsymbol{\alpha}) d\boldsymbol{R}\approx \frac{1}{N}\sum_{i=1}^NE_L(\boldsymbol{R_i},\boldsymbol{\alpha}) +$$ + +with \( N \) being the number of Monte Carlo samples. +

+ + +

+









+ +

Quantum Monte Carlo

+
+ +

+The Algorithm for performing a variational Monte Carlo calculations runs thus as this + +

+ +Observe that the jumping in space is governed by the variable step. This is often called brute-force sampling. +Need importance sampling to get more relevant sampling. +
+ + +

+









+ +

The trial wave function

+
+ +

+We want to perform a Variational Monte Carlo calculation of the ground state of two electrons in a quantum dot well with different oscillator energies, assuming total spin \( S=0 \). +Our trial wave function has the following form +$$ +\begin{equation} + \psi_{T}(\boldsymbol{r}_1,\boldsymbol{r}_2) = + C\exp{\left(-\alpha_1\omega(r_1^2+r_2^2)/2\right)} + \exp{\left(\frac{r_{12}}{(1+\alpha_2 r_{12})}\right)}, +\label{eq:trial} +\end{equation} +$$ + +where the $\alpha$s represent our variational parameters, two in this case. + +

+Why does the trial function look like this? How did we get there? This will be our main motivation for switching to +Machine Learning. + + +

+ + +

+









+ +

The correlation part of the wave function

+ +

+To find an ansatz for the correlated part of the wave function, it is useful to rewrite the two-particle +local energy in terms of the relative and center-of-mass motion. +Let us denote the distance between the two electrons as +\( r_{12} \). We omit the center-of-mass motion since we are only interested in the case when +\( r_{12} \rightarrow 0 \). The contribution from the center-of-mass (CoM) variable \( {\bf R}_{\mathrm{CoM}} \) +gives only a finite contribution. +We focus only on the terms that are relevant for \( r_{12} \) and for three dimensions. The relevant local energy becomes then +$$ +\lim_{r_{12} \rightarrow 0}E_L(R)= + \frac{1}{{\cal R}_T(r_{12})}\left(2\frac{d^2}{dr_{ij}^2}+\frac{4}{r_{ij}}\frac{d}{dr_{ij}}+ +\frac{2}{r_{ij}}-\frac{l(l+1)}{r_{ij}^2}+2E +\right){\cal R}_T(r_{12}) = 0. +$$ + +Set \( l=0 \) and we have the so-called cusp condition +$$ +\frac{d {\cal R}_T(r_{12})}{dr_{12}} = -\frac{1}{2(l+1)} +{\cal R}_T(r_{12})\qquad r_{12}\to 0 +$$ + +

+









+ +

Resulting ansatz

+The above results in +$$ +{\cal R}_T \propto \exp{(r_{ij}/2)}, +$$ + +for anti-parallel spins and +$$ +{\cal R}_T \propto \exp{(r_{ij}/4)}, +$$ + +for anti-parallel spins. +This is the so-called cusp condition for the relative motion, resulting in a minimal requirement +for the correlation part of the wave fuction. +For general systems containing more than say two electrons, we have this +condition for each electron pair \( ij \). + +

+









+ +

The VMC code

+ +

+ + +

# Importing various packages
+from math import exp, sqrt
+from random import random, seed
+import numpy as np
+import matplotlib.pyplot as plt
+from mpl_toolkits.mplot3d import Axes3D
+from matplotlib import cm
+from matplotlib.ticker import LinearLocator, FormatStrFormatter
+import sys
+
+#Trial wave function for quantum dots in two dims
+def WaveFunction(r,alpha,beta):
+    r1 = r[0,0]**2 + r[0,1]**2
+    r2 = r[1,0]**2 + r[1,1]**2
+    r12 = sqrt((r[0,0]-r[1,0])**2 + (r[0,1]-r[1,1])**2)
+    deno = r12/(1+beta*r12)
+    return exp(-0.5*alpha*(r1+r2)+deno)
+
+#Local energy  for quantum dots in two dims, using analytical local energy
+def LocalEnergy(r,alpha,beta):
+    
+    r1 = (r[0,0]**2 + r[0,1]**2)
+    r2 = (r[1,0]**2 + r[1,1]**2)
+    r12 = sqrt((r[0,0]-r[1,0])**2 + (r[0,1]-r[1,1])**2)
+    deno = 1.0/(1+beta*r12)
+    deno2 = deno*deno
+    return 0.5*(1-alpha*alpha)*(r1 + r2) +2.0*alpha + 1.0/r12+deno2*(alpha*r12-deno2+2*beta*deno-1.0/r12)
+
+# The Monte Carlo sampling with the Metropolis algo
+def MonteCarloSampling():
+
+    NumberMCcycles= 100000
+    StepSize = 1.0
+    # positions
+    PositionOld = np.zeros((NumberParticles,Dimension), np.double)
+    PositionNew = np.zeros((NumberParticles,Dimension), np.double)
+    # seed for rng generator
+    seed()
+    # start variational parameter
+    alpha = 0.9
+    for ia in range(MaxVariations):
+        alpha += .025
+        AlphaValues[ia] = alpha
+        beta = 0.2 
+        for jb in range(MaxVariations):
+            beta += .01
+            BetaValues[jb] = beta
+            energy = energy2 = 0.0
+            DeltaE = 0.0
+            #Initial position
+            for i in range(NumberParticles):
+                for j in range(Dimension):
+                    PositionOld[i,j] = StepSize * (random() - .5)
+            wfold = WaveFunction(PositionOld,alpha,beta)
+
+            #Loop over MC MCcycles
+            for MCcycle in range(NumberMCcycles):
+                #Trial position
+                for i in range(NumberParticles):
+                    for j in range(Dimension):
+                        PositionNew[i,j] = PositionOld[i,j] + StepSize * (random() - .5)
+                wfnew = WaveFunction(PositionNew,alpha,beta)
+
+                #Metropolis test to see whether we accept the move
+                if random() < wfnew**2 / wfold**2:
+                   PositionOld = PositionNew.copy()
+                   wfold = wfnew
+                   DeltaE = LocalEnergy(PositionOld,alpha,beta)
+                energy += DeltaE
+                energy2 += DeltaE**2
+
+            #We calculate mean, variance and error ...
+            energy /= NumberMCcycles
+            energy2 /= NumberMCcycles
+            variance = energy2 - energy**2
+            error = sqrt(variance/NumberMCcycles)
+            Energies[ia,jb] = energy    
+    return Energies, AlphaValues, BetaValues
+
+
+#Here starts the main program with variable declarations
+NumberParticles = 2
+Dimension = 2
+MaxVariations = 10
+Energies = np.zeros((MaxVariations,MaxVariations))
+AlphaValues = np.zeros(MaxVariations)
+BetaValues = np.zeros(MaxVariations)
+(Energies, AlphaValues, BetaValues) = MonteCarloSampling()
+
+# Prepare for plots
+fig = plt.figure()
+ax = fig.gca(projection='3d')
+# Plot the surface.
+X, Y = np.meshgrid(AlphaValues, BetaValues)
+surf = ax.plot_surface(X, Y, Energies,cmap=cm.coolwarm,linewidth=0, antialiased=False)
+# Customize the z axis.
+zmin = np.matrix(Energies).min()
+zmax = np.matrix(Energies).max()
+ax.set_zlim(zmin, zmax)
+ax.set_xlabel(r'$\alpha$')
+ax.set_ylabel(r'$\beta$')
+ax.set_zlabel(r'$\langle E \rangle$')
+ax.zaxis.set_major_locator(LinearLocator(10))
+ax.zaxis.set_major_formatter(FormatStrFormatter('%.02f'))
+# Add a color bar which maps values to colors.
+fig.colorbar(surf, shrink=0.5, aspect=5)
+plt.show()
+
+

+









+ +

Technical aspect, improvements and how to define the cost function

+
+ +

+ +

+The above procedure is not the smartest one. Looping over all variational parameters becomes expensive. +Also, we don't use importance sampling and optimizations of the standard deviation (blocking, bootstrap, jackknife). +Such codes are included in the above Github address. + +

+We can also be smarter and use minimization methods to find the optimal variational parameters with fewer Monte Carlo cycles and then +fire up our heavy artillery. + +

+One way to achieve this is to minimize the energy as function of the variational parameters. + + +

+ + +

+









+ +

Energy derivatives

+
+ +

+To find the derivatives of the local energy expectation value as function of the variational parameters, we can use the chain rule and the hermiticity of the Hamiltonian. + +

+Let us define +$$ +\bar{E}_{\alpha_i}=\frac{d\langle E_L\rangle}{d\alpha_i}. +$$ + +as the derivative of the energy with respect to the variational parameter \( \alpha_i \) +We define also the derivative of the trial function (skipping the subindex \( T \)) as +$$ +\bar{\Psi}_{i}=\frac{d\Psi}{d\alpha_i}. +$$ +

+ + +

+









+ +

Derivatives of the local energy

+
+ +

+The elements of the gradient of the local energy are then (using the chain rule and the hermiticity of the Hamiltonian) +$$ +\bar{E}_{i}= 2\left( \langle \frac{\bar{\Psi}_{i}}{\Psi}E_L\rangle -\langle \frac{\bar{\Psi}_{i}}{\Psi}\rangle\langle E_L \rangle\right). +$$ + +From a computational point of view it means that you need to compute the expectation values of +$$ +\langle \frac{\bar{\Psi}_{i}}{\Psi}E_L\rangle, +$$ + +and +$$ +\langle \frac{\bar{\Psi}_{i}}{\Psi}\rangle\langle E_L\rangle +$$ + +These integrals are evaluted using MC intergration (with all its possible error sources). +We can then use methods like stochastic gradient or other minimization methods to find the optimal variational parameters (I don't discuss this topic here, but these methods are very important in ML). +

+ + +

+









+ +

How do we define our cost function?

+
+ +

+We have a model, our likelihood function. + +

+How should we define the cost function? + + +

+ + +

+









+ +

Meet the variance and its derivatives

+
+Why the variance? +

+Suppose the trial function (our model) is the exact wave function. The action of the hamiltionan +on the wave function +$$ + H\Psi = \mathrm{constant}\times \Psi, +$$ + +The integral which defines various +expectation values involving moments of the hamiltonian becomes then +$$ + \langle E^n \rangle = \langle H^n \rangle = + \frac{\int d\boldsymbol{R}\Psi^{\ast}(\boldsymbol{R})H^n(\boldsymbol{R})\Psi(\boldsymbol{R})} + {\int d\boldsymbol{R}\Psi^{\ast}(\boldsymbol{R})\Psi(\boldsymbol{R})}= +\mathrm{constant}\times\frac{\int d\boldsymbol{R}\Psi^{\ast}(\boldsymbol{R})\Psi(\boldsymbol{R})} + {\int d\boldsymbol{R}\Psi^{\ast}(\boldsymbol{R})\Psi(\boldsymbol{R})}=\mathrm{constant}. +$$ + +This gives an important information: If I want the variance, the exact wave function leads to zero variance! +The variance is defined as +$$ +\sigma_E = \langle E^2\rangle - \langle E\rangle^2. +$$ + +Variation is then performed by minimizing both the energy and the variance. + + +

+ + +

+









+ +

The variance defines the cost function

+
+ +

+We can then take the derivatives of +$$ +\sigma_E = \langle E^2\rangle - \langle E\rangle^2, +$$ + +with respect to the variational parameters. The derivatives of the variance can then be used to defined the +so-called Hessian matrix, which in turn allows us to use minimization methods like Newton's method or +standard gradient methods. + +

+This leads to however a more complicated expression, with obvious errors when evaluating integrals by Monte Carlo integration. Less used, see however Filippi and Umrigar. The expression becomes complicated +$$ +\bar{E}_{ij} = 2\left[ \langle (\frac{\bar{\Psi}_{ij}}{\Psi}+\frac{\bar{\Psi}_{j}}{\Psi}\frac{\bar{\Psi}_{i}}{\Psi})(E_L-\langle E\rangle)\rangle -\langle \frac{\bar{\Psi}_{i}}{\Psi}\rangle\bar{E}_j-\langle \frac{\bar{\Psi}_{j}}{\Psi}\rangle\bar{E}_i\right] +\langle \frac{\bar{\Psi}_{i}}{\Psi}E_L{_j}\rangle +\langle \frac{\bar{\Psi}_{j}}{\Psi}E_L{_i}\rangle -\langle \frac{\bar{\Psi}_{i}}{\Psi}\rangle\langle E_L{_j}\rangle \langle \frac{\bar{\Psi}_{j}}{\Psi}\rangle\langle E_L{_i}\rangle. +$$ + +

+Evaluating the cost function means having to evaluate the above second derivative of the energy. + + +

+ + +

+









+ +

Boltzmann Machines

+ +

+Why use a generative model rather than the more well known discriminative deep neural networks (DNN)? + +

+ +









+ +

Some similarities and differences from DNNs

+ +
    +
  1. Both use gradient-descent based learning procedures for minimizing cost functions
    2. +
  2. Energy based models don't use backpropagation and automatic differentiation for computing gradients, instead turning to Markov Chain Monte Carlo methods.
    3. +
  3. DNNs often have several hidden layers. A restricted Boltzmann machine has only one hidden layer, however several RBMs can be stacked to make up Deep Belief Networks, of which they constitute the building blocks.
    4. +
+ +History: The RBM was developed by amongst others Geoffrey Hinton, called by some the "Godfather of Deep Learning", working with the University of Toronto and Google. + +

+









+ +

The structure of the RBM network

+ +

+



+ +

+









+ +

The network

+ +

+The network layers: + +

    +
  1. A function \( \mathbf{x} \) that represents the visible layer, a vector of \( M \) elements (nodes). This layer represents both what the RBM might be given as training input, and what we want it to be able to reconstruct. This might for example be the pixels of an image, the spin values of the Ising model, or coefficients representing speech.
    2. +
  2. The function \( \mathbf{h} \) represents the hidden, or latent, layer. A vector of \( N \) elements (nodes). Also called "feature detectors".
    3. +
+ +









+ +

Goals

+ +

+The goal of the hidden layer is to increase the model's expressive power. We encode complex interactions between visible variables by introducing additional, hidden variables that interact with visible degrees of freedom in a simple manner, yet still reproduce the complex correlations between visible degrees in the data once marginalized over (integrated out). + +

+Examples of this trick being employed in physics: + +

    +
  1. The Hubbard-Stratonovich transformation
    2. +
  2. The introduction of ghost fields in gauge theory
    3. +
  3. Shadow wave functions in Quantum Monte Carlo simulations
    4. +
+ +The network parameters, to be optimized/learned: + +
    +
  1. \( \mathbf{a} \) represents the visible bias, a vector of same length as \( \mathbf{x} \).
    2. +
  2. \( \mathbf{b} \) represents the hidden bias, a vector of same lenght as \( \mathbf{h} \).
    3. +
  3. \( W \) represents the interaction weights, a matrix of size \( M\times N \).
    4. +
+ +









+ +

Joint distribution and the Energy function

+The restricted Boltzmann machine is described by a Bolztmann distribution +$$ +\begin{align} + P_{rbm}(\mathbf{x},\mathbf{h}) = \frac{1}{Z} e^{-\frac{1}{T_0}E(\mathbf{x},\mathbf{h})}, +\label{_auto1} +\end{align} +$$ + +where \( Z \) is the normalization constant or partition function, defined as +$$ +\begin{align} + Z = \int \int e^{-\frac{1}{T_0}E(\mathbf{x},\mathbf{h})} d\mathbf{x} d\mathbf{h}. +\label{_auto2} +\end{align} +$$ + +It is common to ignore \( T_0 \) by setting it to one. + +

+









+ +

Network Elements

+ +

+The function \( E(\mathbf{x},\mathbf{h}) \) gives the energy of a +configuration (pair of vectors) \( (\mathbf{x}, \mathbf{h}) \). The lower +the energy of a configuration, the higher the probability of it. This +function also depends on the parameters \( \mathbf{a} \), \( \mathbf{b} \) and +\( W \). Thus, when we adjust them during the learning procedure, we are +adjusting the energy function to best fit our problem. + +

+









+ +

Defining different types of RBMs

+There are different variants of RBMs, and the differences lie in the types of visible and hidden units we choose as well as in the implementation of the energy function \( E(\mathbf{x},\mathbf{h}) \). + +

+

+Binary-Binary RBM: +

+ +

+RBMs were first developed using binary units in both the visible and hidden layer. The corresponding energy function is defined as follows: +$$ +\begin{align} + E(\mathbf{x}, \mathbf{h}) = - \sum_i^M x_i a_i- \sum_j^N b_j h_j - \sum_{i,j}^{M,N} x_i w_{ij} h_j, +\label{_auto3} +\end{align} +$$ + +where the binary values taken on by the nodes are most commonly 0 and 1. +

+ +
+Gaussian-Binary RBM: +

+ +

+Another varient is the RBM where the visible units are Gaussian while the hidden units remain binary: +$$ +\begin{align} + E(\mathbf{x}, \mathbf{h}) = \sum_i^M \frac{(x_i - a_i)^2}{2\sigma_i^2} - \sum_j^N b_j h_j - \sum_{i,j}^{M,N} \frac{x_i w_{ij} h_j}{\sigma_i^2}. +\label{_auto4} +\end{align} +$$ +

+ + +

+









+ +

More about RBMs

+ +
    +
  1. Useful when we model continuous data (i.e., we wish \( \mathbf{x} \) to be continuous)
    2. +
  2. Requires a smaller learning rate, since there's no upper bound to the value a component might take in the reconstruction
    3. +
+ +Other types of units include: + +
    +
  1. Softmax and multinomial units
    2. +
  2. Gaussian visible and hidden units
    3. +
  3. Binomial units
    4. +
  4. Rectified linear units
    5. +
+ +









+ +

Sampling: Metropolis sampling

+In order to sample from the RBM probability distribution it is common to use Markov Chain Monte Carlo (MCMC) algorithms such as Metropolis-Hastings or Gibbs sampling. + +

+Metropolis sampling starts by suggesting a new configuration \( \boldsymbol{x}^{k+1} \). In the brute force method this is done by some random change of the visible units. The new configuration is then accepted with the acceptance probability +$$ +\begin{align} + A(\boldsymbol{x}^k \rightarrow \boldsymbol{x}^{k+1}) = \text{min} (1, \frac{P(\boldsymbol{x}^{k+1})}{P(\boldsymbol{x}^k)}), +\label{_auto5} +\end{align} +$$ + +where we need the marginalized probability +$$ +\begin{align} + P(\boldsymbol{x}) &= \sum_\mathbf{h} P_{rbm}(\mathbf{x}, \mathbf{h}) +\label{_auto6}\\ + &= \frac{1}{Z}\sum_\mathbf{h} e^{-E(\mathbf{x}, \mathbf{h})}. +\label{_auto7} +\end{align} +$$ + +

+









+ +

Sampling: Gibbs sampling

+ +

+In this method we sample from the joint probability \( P_{rbm} (\mathbf{x}, \mathbf{h}) \) by way of a two step sampling process. We alternately update the visible and hidden units. +New samples are generated according to the conditional probabilities \( P(x_i|\mathbf{h}) \) and \( P(h_j|\mathbf{x}) \) respectively and accepted with the probability of \( 1 \). While the the visible nodes are dependent on the hidden nodes and vice versa, the nodes are independent of other nodes within the same layer. This is due to there being no intra layer interactions in the restricted Boltzmann machine. + +

+The conditional probabilities are often referred to as the activitation functions in the neural networks context due to their role in determining the node outputs. For the binary-binary RBM they are +$$ +\begin{align} + P(h_j = 1 | \boldsymbol{x}) &= \frac{1}{1 + e^{-b_j - \sum_i x_i w_{ij}}} +\label{_auto8}\\ + P(x_i = 1 | \boldsymbol{h}) &= \frac{1}{1 + e^{-a_j - \sum_j h_j w_{ij}}}, +\label{_auto9} +\end{align} +$$ + +where we recognize the logistic sigmoid function \( \sigma (x) = 1/(1+exp(-x)) \). + +

+









+ +

Gaussian RBM

+For the Gaussian-Binary RBM the conditional probabilities are +$$ +\begin{align} + P(x_i|\mathbf{h}) &= \mathcal{N}(x_i; a_i+ \sum_j h_j w_{ij}, \sigma^2) +\label{_auto10}\\ + P(h_j=1|\mathbf{x}) &= \frac{1}{1+e^{-b_j-\frac{1}{\sigma^2} \sum_i x_i w_{ij}}}, +\label{_auto11} +\end{align} +$$ + +while the visible units now follow a normal distribution, we see the hidden units again follow the logistic sigmoid function. + +

+









+ +

Cost function

+ +

+When working with a training dataset, the most common training approach is maximizing the log-likelihood of the training data. The log likelihood characterizes the log-probability of generating the observed data using our generative model. Using this method our cost function is chosen as the negative log-likelihood. The learning then consists of trying to find parameters that maximize the probability of the dataset, and is known as Maximum Likelihood Estimation (MLE). +Denoting the parameters as \( \boldsymbol{\theta} = a_1,...,a_M,b_1,...,b_N,w_{11},...,w_{MN} \), the log-likelihood is given by +$$ +\begin{align} + \mathcal{L}(\{ \theta_i \}) &= \langle \text{log} P_\theta(\boldsymbol{x}) \rangle_{data} +\label{_auto12}\\ + &= - \langle E(\boldsymbol{x}; \{ \theta_i\}) \rangle_{data} - \text{log} Z(\{ \theta_i\}), +\label{_auto13} +\end{align} +$$ + +where we used that the normalization constant does not depend on the data, \( \langle \text{log} Z(\{ \theta_i\}) \rangle = \text{log} Z(\{ \theta_i\}) \) +Our cost function is the negative log-likelihood, \( \mathcal{C}(\{ \theta_i \}) = - \mathcal{L}(\{ \theta_i \}) \) + +

+









+ +

Optimization / Training

+The training procedure of choice often is Stochastic Gradient Descent (SGD). It consists of a series of iterations where we update the parameters according to the equation +$$ +\begin{align} + \boldsymbol{\theta}_{k+1} = \boldsymbol{\theta}_k - \eta \nabla \mathcal{C} (\boldsymbol{\theta}_k) +\label{_auto14} +\end{align} +$$ + +at each \( k \)-th iteration. There are a range of variants of the algorithm which aim at making the learning rate \( \eta \) more adaptive so the method might be more efficient while remaining stable. + +

+We now need the gradient of the cost function in order to minimize it. We find that +$$ +\begin{align} + \frac{\partial \mathcal{C}(\{ \theta_i\})}{\partial \theta_i} + &= \langle \frac{\partial E(\boldsymbol{x}; \theta_i)}{\partial \theta_i} \rangle_{data} + + \frac{\partial \text{log} Z(\{ \theta_i\})}{\partial \theta_i} +\label{_auto15}\\ + &= \langle O_i(\boldsymbol{x}) \rangle_{data} - \langle O_i(\boldsymbol{x}) \rangle_{model}, +\label{_auto16} +\end{align} +$$ + +where in order to simplify notation we defined the "operator" +$$ +\begin{align} + O_i(\boldsymbol{x}) = \frac{\partial E(\boldsymbol{x}; \theta_i)}{\partial \theta_i}, +\label{_auto17} +\end{align} +$$ + +and used the statistical mechanics relationship between expectation values and the log-partition function: +$$ +\begin{align} + \langle O_i(\boldsymbol{x}) \rangle_{model} = \text{Tr} P_\theta(\boldsymbol{x})O_i(\boldsymbol{x}) = - \frac{\partial \text{log} Z(\{ \theta_i\})}{\partial \theta_i}. +\label{_auto18} +\end{align} +$$ + +

+ + +

More on RBMs

+ +

+The data-dependent term in the gradient is known as the positive phase of the gradient, while the model-dependent term is known as the negative phase of the gradient. The aim of the training is to lower the energy of configurations that are near observed data points (increasing their probability), and raising the energy of configurations that are far from observed data points (decreasing their probability). + +

+The gradient of the negative log-likelihood cost function of a Binary-Binary RBM is then +$$ +\begin{align} + \frac{\partial \mathcal{C} (w_{ij}, a_i, b_j)}{\partial w_{ij}} =& \langle x_i h_j \rangle_{data} - \langle x_i h_j \rangle_{model} +\label{_auto19}\\ + \frac{\partial \mathcal{C} (w_{ij}, a_i, b_j)}{\partial a_{ij}} =& \langle x_i \rangle_{data} - \langle x_i \rangle_{model} +\label{_auto20}\\ + \frac{\partial \mathcal{C} (w_{ij}, a_i, b_j)}{\partial b_{ij}} =& \langle h_i \rangle_{data} - \langle h_i \rangle_{model}. +\label{_auto21}\\ +\label{_auto22} +\end{align} +$$ + +To get the expecation values with respect to the data, we set the visible units to each of the observed samples in the training data, then update the hidden units according to the conditional probability found before. We then average over all samples in the training data to calculate expectation values with respect to the data. + +

+









+ +

Which sampling to use

+ +

+To get the expectation values with respect to the model, we use Gibbs sampling. We can either initialize the \( \boldsymbol{x} \) randomly or with a training sample. While we ideally want a large number of Gibbs iterations \( n\rightarrow n \), one might decide to truncate it earlier for efficiency. Doing this while having intialized \( \boldsymbol{x} \) with a training data vector is referred to as contrastive divergence (CD), because one is then closer to approximating the gradient of this function than the negative log-likelihood. The contrastive divergence function is the difference between two Kullback-Leibler divergences (also called relative entropy), which measure how one probability distribution diverges from a second, expected probability distribution (in this case the estimated one from the ground truth one). + +

+









+ +

RBMs for the quantum many body problem

+ +

+The idea of applying RBMs to quantum many body problems was presented by G. Carleo and M. Troyer, working with ETH Zurich and Microsoft Research. + +

+Some of their motivation included + +

+ +









+ +

Choose the right RBM

+ +

+Carleo and Troyer applied the RBM to the quantum mechanical spin lattice systems of the Ising model and Heisenberg model, with encouraging results. Our goal is to test the method on systems of moving particles. For the spin lattice systems it was natural to use a binary-binary RBM, with the nodes taking values of 1 and -1. For moving particles, on the other hand, we want the visible nodes to be continuous, representing position coordinates. Thus, we start by choosing a Gaussian-binary RBM, where the visible nodes are continuous and hidden nodes take on values of 0 or 1. If eventually we would like the hidden nodes to be continuous as well the rectified linear units seem like the most relevant choice. + +

+









+ +

Representing the wave function

+The wavefunction should be a probability amplitude depending on \( \boldsymbol{x} \). The RBM model is given by the joint distribution of \( \boldsymbol{x} \) and \( \boldsymbol{h} \) +$$ +\begin{align} + F_{rbm}(\mathbf{x},\mathbf{h}) = \frac{1}{Z} e^{-\frac{1}{T_0}E(\mathbf{x},\mathbf{h})}. +\label{_auto23} +\end{align} +$$ + +To find the marginal distribution of \( \boldsymbol{x} \) we set: +$$ +\begin{align} + F_{rbm}(\mathbf{x}) &= \sum_\mathbf{h} F_{rbm}(\mathbf{x}, \mathbf{h}) +\label{_auto24}\\ + &= \frac{1}{Z}\sum_\mathbf{h} e^{-E(\mathbf{x}, \mathbf{h})}. +\label{_auto25} +\end{align} +$$ + +Now this is what we use to represent the wave function, calling it a neural-network quantum state (NQS) +$$ +\begin{align} + \Psi (\mathbf{X}) &= F_{rbm}(\mathbf{x}) +\label{_auto26}\\ + &= \frac{1}{Z}\sum_{\boldsymbol{h}} e^{-E(\mathbf{x}, \mathbf{h})} +\label{_auto27}\\ + &= \frac{1}{Z} \sum_{\{h_j\}} e^{-\sum_i^M \frac{(x_i - a_i)^2}{2\sigma^2} + \sum_j^N b_j h_j + \sum_{i,j}^{M,N} \frac{x_i w_{ij} h_j}{\sigma^2}} +\label{_auto28}\\ + &= \frac{1}{Z} e^{-\sum_i^M \frac{(x_i - a_i)^2}{2\sigma^2}} \prod_j^N (1 + e^{b_j + \sum_i^M \frac{x_i w_{ij}}{\sigma^2}}). +\label{_auto29}\\ +\label{_auto30} +\end{align} +$$ + +

+









+ +

Choose the cost function

+Now we don't necessarily have training data (unless we generate it by using some other method). However, what we do have is the variational principle which allows us to obtain the ground state wave function by minimizing the expectation value of the energy of a trial wavefunction (corresponding to the untrained NQS). Similarly to the traditional variational Monte Carlo method then, it is the local energy we wish to minimize. The gradient to use for the stochastic gradient descent procedure is +$$ +\begin{align} + G_i = \frac{\partial \langle E_L \rangle}{\partial \theta_i} + = 2(\langle E_L \frac{1}{\Psi}\frac{\partial \Psi}{\partial \theta_i} \rangle - \langle E_L \rangle \langle \frac{1}{\Psi}\frac{\partial \Psi}{\partial \theta_i} \rangle ), +\label{_auto31} +\end{align} +$$ + +where the local energy is given by +$$ +\begin{align} + E_L = \frac{1}{\Psi} \hat{\mathbf{H}} \Psi. +\label{_auto32} +\end{align} +$$ + +

+









+ +

Running the codes

+
+ +

+You can find the codes for the simple two-electron case at the Github repository https://github.com/mhjensenseminars/MachineLearningTalk/tree/master/doc/Programs/MLcpp/src. Python codes to come, only c++ as of now. + +

+The trial wave function is based on the product of a Slater determinant with Gaussian orbitals, a simple Jastrow factor \( \exp{(r_{ij})} \) and the reduced Boltzmann machines. + +

+The Broyden-Fletcher-Goldfarb-Shanno algorithm was used to perform the minimization. We used \( 14 \) hidden nodes in the calculations below. + + +

+ + +

+









+ +

Energy as function of iterations, \( N=2 \) electrons

+
+ +

+



+
+ + +

+









+ +

Energy as function of iterations, \( N=6 \) electrons

+
+ +

+



+
+ + +

+









+ +

Conclusions and where do we stand

+
+ +

+ +

+
+ + +

+ + + + +

+ © 1999-2018, Morten Hjorth-Jensen. Released under CC Attribution-NonCommercial 4.0 license +
+ + + + + + diff --git a/doc/pub/BM/html/BM.html b/doc/pub/BM/html/BM.html new file mode 100644 index 000000000..42ad4c72f --- /dev/null +++ b/doc/pub/BM/html/BM.html @@ -0,0 +1,1532 @@ + + + + + + + +Solving quantum mechanical problems with Machine Learning and Boltzmann machines + + + + + + + + + + + + + + + + + + + + + + + +

Solving quantum mechanical problems with Machine Learning and Boltzmann machines

+ +

+ + +

+Morten Hjorth-Jensen +
+ +

+ + +

Department of Physics and Astronomy and National Superconducting Cyclotron Laboratory, Michigan State University and Department of Physics, University of Oslo, Norway
+
+

+

Jul 26, 2018

+
+

+









+ +

What is this talk about?

+
+ +

+The main aim is to give you a short and pedestrian introduction to how we can use Machine Learning methods +to solve quantum mechanical many-body problems. And why this could be of interest. + +

+The hope is that after this talk you have gotten the basic ideas to get you started. Peeping into https://github.com/mhjensenseminars/MachineLearningTalk, you'll find a Jupyter notebook, slides, codes etc that will allow you to reproduce the simulations discussed here, and perhaps run your own very first calculations. + +

+Furthermore, since we will also talk about Monte Carlo simulations, the exciting idea of Monte Carlo optimization and Machine Learning can be used to mesh strategies for reducing the curse of dimensionality. + + +

+ + +

+









+ +

Why?

+ +

+How can we avoid the dimensionality curse? Many possibilities + +

    +
  1. smarter basis functions
    2. +
  2. resummation of specific correlations
    3. +
  3. stochastic sampling of high-lying states (stochastic FCI, CC and soon SRG)
    4. +
  4. many more
    5. +
+ +Machine Learning and Quantum Computing hold also great promise in tackling the +ever increasing dimensionalities. Here we will focus on Machine Learning. + +

+









+ +

Overview

+
+ +

+ +

+
+ + +

+









+ +

What are the Machine Learning calculations here based on?

+ +

+This work is inspired by the idea of representing the wave function with +a restricted Boltzmann machine (RBM), presented recently by G. Carleo and M. Troyer, Science 355, Issue 6325, pp. 602-606 (2017). They +named such a wave function/network a neural network quantum state (NQS). In their article they apply it to the quantum mechanical +spin lattice systems of the Ising model and Heisenberg model, with +encouraging results. + +

+Thanks to Vilde Flugsrud and Alfred Alocias Mariadason for many discussions and interpretations of the above paper. + +

+









+ +

A new world

+
+ +

+Machine learning (ML) is an extremely rich field, in spite of its young age. The +increases we have seen during the last three decades in computational +capabilities have been followed by developments of methods and +techniques for analyzing and handling large date sets, relying heavily +on statistics, computer science and mathematics. The field is rather +new and developing rapidly. + +

+Popular software packages written in Python for ML are + +

+ +These are all freely available at their respective GitHub sites. They +encompass communities of developers in the thousands or more. And the number +of code developers and contributors keeps increasing. +
+ + +

+









+ +

Lots of room for creativity

+
+ +

+Not all the +algorithms and methods can be given a rigorous mathematical +justification, opening up thereby for experimenting +and trial and error and thereby exciting new developments. +

+ + +

+

+ +

+A solid command of linear algebra, multivariate theory, +probability theory, statistical data analysis, +understanding errors and Monte Carlo methods is important in order to understand many of the +various algorithms and methods. +

+ + +

+

+Job market, a personal statement. +

+A familiarity with ML is almost becoming a prerequisite for many of the most exciting employment +opportunities. Add quantum computing and there you are. +

+ + +

+ + +

Some members of the ML family

+ +

+

+ +

+ +

    +
  1. Linear regression and its variants, in essence polynomial regression
    2. +
  2. Decision tree algorithms, from simpler to more complex ones like random forests
    3. +
  3. Nearest neighbors models
    4. +
  4. Bayesian statistics and regression
    5. +
  5. Support vector machines and finally various variants of
    6. +
  6. Artifical neural networks and deep learning
    7. +
  7. ....
    8. +
+
+ + +

+









+ +

What are the basic ingredients?

+
+ +

+Almost every problem in ML and data science starts with the same ingredients: + +

+ +We seek to minimize the function \( \mathcal{C} (\mathbf{x}, f(\mathbf{\alpha})) \) by finding the parameter values which minimize \( \mathcal{C} \). This leads to various minimization algorithms. +
+ + +

+









+ +

What is Machine Learning?

+
+ +

+Machine learning is the science of giving computers the ability to +learn without being explicitly programmed. The idea is that there +exist generic algorithms which can be used to find patterns in a broad +class of data sets without having to write code specifically for each +problem. The algorithm will build its own logic based on the data. + +

+Machine learning is a subfield of computer science, and is closely +related to computational statistics. It evolved from the study of +pattern recognition in artificial intelligence (AI) research, and has +made contributions to AI tasks like computer vision, natural language +processing and speech recognition. It has also, especially in later +years, found applications in a wide variety of other areas, including +bioinformatics, economy, physics, finance and marketing. + +

+You will notice however that many of the basic ideas discussed do come from Physics! +

+ + +

+









+ +

Types of Machine Learning

+ +

+

+ +

+The approaches to machine learning are many, but are often split into two main categories. +In supervised learning we know the answer to a problem, +and let the computer deduce the logic behind it. On the other hand, unsupervised learning +is a method for finding patterns and relationship in data sets without any prior knowledge of the system. +Some authours also operate with a third category, namely reinforcement learning. This is a paradigm +of learning inspired by behavioural psychology, where learning is achieved by trial-and-error, +solely from rewards and punishment. + +

+Another way to categorize machine learning tasks is to consider the desired output of a system. +Some of the most common tasks are: + +

+
+ + +

+









+ +

References

+
+ +

+ +

+
+ + +

+









+ +

Another interesting article

+ +

+Here we will use so-called reduced Boltzmann Machines to simulate quantum many-body problems. For Monte Carlo aficionados, there is a very close similarity with what are called shadow wave functions, see the work of Pederiva and Kalos and collaborators, Phys Rev. E 90, 053304 (2014). + +

+









+ +

Just for fun illustration of scikit-learn: Hudson bay data

+
+ +

+Most mammalian predators rely on a variety of prey, which complicates mathematical modeling; however, a few predators have become highly specialized and seek almost exclusively a single prey species. An example of this simplified predator-prey interaction is seen in Canadian northern forests, where the populations of the lynx and the snowshoe hare are intertwined in a life and death struggle. + +

+One reason that this particular system has been so extensively studied is that the Hudson Bay company kept careful records of all furs from the early 1800s into the 1900s. The records for the furs collected by the Hudson Bay company showed distinct oscillations (approximately 12 year periods), suggesting that these species caused almost periodic fluctuations of each other's populations. The table here shows data from 1900 to 1920. + + +

+ + +

+









+ +

Hares and lynx in Hudson bay from 1900 to 1920

+ +

+



+ +

+









+ +

Linear regression in Python

+
+ +

+

+ + +

import numpy as np
+import matplotlib.pyplot as plt
+from IPython.display import display
+import sklearn
+from sklearn.linear_model import LinearRegression
+from sklearn.tree import DecisionTreeRegressor
+
+
+data = np.loadtxt('src/Hudson_Bay.csv', delimiter=',', skiprows=1)
+x = data[:,0]
+y = data[:,1]
+line = np.linspace(1900,1920,1000,endpoint=False).reshape(-1,1)
+reg = DecisionTreeRegressor(min_samples_split=3).fit(x.reshape(-1,1),y.reshape(-1,1))
+plt.plot(line, reg.predict(line), label="decision tree")
+regline = LinearRegression().fit(x.reshape(-1,1),y.reshape(-1,1))
+plt.plot(line, regline.predict(line), label= "Linear Regression")
+plt.plot(x, y, label= "Linear Regression")
+plt.show()
+
+ +
+ + +

+









+ +

Artificial neurons

+
+ +

+ +

+The field of artificial neural networks has a long history of +development, and is closely connected with the advancement of computer +science and computers in general. A model of artificial neurons was +first developed by McCulloch and Pitts in 1943 to study signal +processing in the brain and has later been refined by others. The +general idea is to mimic neural networks in the human brain, which is +composed of billions of neurons that communicate with each other by +sending electrical signals. Each neuron accumulates its incoming +signals, which must exceed an activation threshold to yield an +output. If the threshold is not overcome, the neuron remains inactive, +i.e. has zero output. + +

+This behaviour has inspired a simple mathematical model for an artificial neuron. + +$$ + y = f\left(\sum_{i=1}^n w_ix_i\right) = f(u) +$$ + +Here, the output \( y \) of the neuron is the value of its activation function, which have as input +a weighted sum of signals \( x_i, \dots ,x_n \) received by \( n \) other neurons. +

+ + +

+









+ +

A simple perceptron model

+ +

+



+ +

+









+ +

Neural network types

+
+ +

+An artificial neural network (NN), is a computational model that consists of layers of connected neurons, or nodes. +It is supposed to mimic a biological nervous system by letting each neuron interact with other neurons +by sending signals in the form of mathematical functions between layers. +A wide variety of different NNs have +been developed, but most of them consist of an input layer, an output layer and eventual layers in-between, called +hidden layers. All layers can contain an arbitrary number of nodes, and each connection between two nodes +is associated with a weight variable. + +

+



+
+ + +

+









+ +

The system: two electrons in a harmonic oscillator trap in two dimensions

+ +

+The Hamiltonian of the quantum dot is given by +$$ \hat{H} = \hat{H}_0 + \hat{V}, +$$ + +where \( \hat{H}_0 \) is the many-body HO Hamiltonian, and \( \hat{V} \) is the +inter-electron Coulomb interactions. In dimensionless units, +$$ \hat{V}= \sum_{i < j}^N \frac{1}{r_{ij}}, +$$ + +with \( r_{ij}=\sqrt{\mathbf{r}_i^2 - \mathbf{r}_j^2} \). + +

+This leads to the separable Hamiltonian, with the relative motion part given by (\( r_{ij}=r \)) +$$ +\hat{H}_r=-\nabla^2_r + \frac{1}{4}\omega^2r^2+ \frac{1}{r}, +$$ + +plus a standard Harmonic Oscillator problem for the center-of-mass motion. +This system has analytical solutions in two and three dimensions (M. Taut 1993 and 1994). + +

+









+ +

Quantum Monte Carlo Motivation

+
+ +

+Given a hamiltonian \( H \) and a trial wave function \( \Psi_T \), the variational principle states that the expectation value of \( \langle H \rangle \), defined through +$$ + \langle E \rangle = + \frac{\int d\boldsymbol{R}\Psi^{\ast}_T(\boldsymbol{R})H(\boldsymbol{R})\Psi_T(\boldsymbol{R})} + {\int d\boldsymbol{R}\Psi^{\ast}_T(\boldsymbol{R})\Psi_T(\boldsymbol{R})}, +$$ + +is an upper bound to the ground state energy \( E_0 \) of the hamiltonian \( H \), that is +$$ + E_0 \le \langle H \rangle . +$$ + +In general, the integrals involved in the calculation of various expectation values are multi-dimensional ones. Traditional integration methods such as the Gauss-Legendre will not be adequate for say the computation of the energy of a many-body system. +

+ + +

+









+ +

Quantum Monte Carlo Motivation

+
+Basic steps. +

+Choose a trial wave function +\( \psi_T(\boldsymbol{R}) \). +$$ + P(\boldsymbol{R},\boldsymbol{\alpha})= \frac{\left|\psi_T(\boldsymbol{R},\boldsymbol{\alpha})\right|^2}{\int \left|\psi_T(\boldsymbol{R},\boldsymbol{\alpha})\right|^2d\boldsymbol{R}}. +$$ + +This is our model, or likelihood/probability distribution function (PDF). It depends on some variational parameters \( \boldsymbol{\alpha} \). +The approximation to the expectation value of the Hamiltonian is now +$$ + \langle E[\boldsymbol{\alpha}] \rangle = + \frac{\int d\boldsymbol{R}\Psi^{\ast}_T(\boldsymbol{R},\boldsymbol{\alpha})H(\boldsymbol{R})\Psi_T(\boldsymbol{R},\boldsymbol{\alpha})} + {\int d\boldsymbol{R}\Psi^{\ast}_T(\boldsymbol{R},\boldsymbol{\alpha})\Psi_T(\boldsymbol{R},\boldsymbol{\alpha})}. +$$ +

+ + +

+









+ +

Quantum Monte Carlo Motivation

+
+Define a new quantity. +

+$$ + E_L(\boldsymbol{R},\boldsymbol{\alpha})=\frac{1}{\psi_T(\boldsymbol{R},\boldsymbol{\alpha})}H\psi_T(\boldsymbol{R},\boldsymbol{\alpha}), +$$ + +called the local energy, which, together with our trial PDF yields +$$ + E[\boldsymbol{\alpha}]=\int P(\boldsymbol{R})E_L(\boldsymbol{R},\boldsymbol{\alpha}) d\boldsymbol{R}\approx \frac{1}{N}\sum_{i=1}^NE_L(\boldsymbol{R_i},\boldsymbol{\alpha}) +$$ + +with \( N \) being the number of Monte Carlo samples. +

+ + +

+









+ +

Quantum Monte Carlo

+
+ +

+The Algorithm for performing a variational Monte Carlo calculations runs thus as this + +

+ +Observe that the jumping in space is governed by the variable step. This is often called brute-force sampling. +Need importance sampling to get more relevant sampling. +
+ + +

+









+ +

The trial wave function

+
+ +

+We want to perform a Variational Monte Carlo calculation of the ground state of two electrons in a quantum dot well with different oscillator energies, assuming total spin \( S=0 \). +Our trial wave function has the following form +$$ +\begin{equation} + \psi_{T}(\boldsymbol{r}_1,\boldsymbol{r}_2) = + C\exp{\left(-\alpha_1\omega(r_1^2+r_2^2)/2\right)} + \exp{\left(\frac{r_{12}}{(1+\alpha_2 r_{12})}\right)}, +\label{eq:trial} +\end{equation} +$$ + +where the $\alpha$s represent our variational parameters, two in this case. + +

+Why does the trial function look like this? How did we get there? This will be our main motivation for switching to +Machine Learning. + + +

+ + +

+









+ +

The correlation part of the wave function

+ +

+To find an ansatz for the correlated part of the wave function, it is useful to rewrite the two-particle +local energy in terms of the relative and center-of-mass motion. +Let us denote the distance between the two electrons as +\( r_{12} \). We omit the center-of-mass motion since we are only interested in the case when +\( r_{12} \rightarrow 0 \). The contribution from the center-of-mass (CoM) variable \( {\bf R}_{\mathrm{CoM}} \) +gives only a finite contribution. +We focus only on the terms that are relevant for \( r_{12} \) and for three dimensions. The relevant local energy becomes then +$$ +\lim_{r_{12} \rightarrow 0}E_L(R)= + \frac{1}{{\cal R}_T(r_{12})}\left(2\frac{d^2}{dr_{ij}^2}+\frac{4}{r_{ij}}\frac{d}{dr_{ij}}+ +\frac{2}{r_{ij}}-\frac{l(l+1)}{r_{ij}^2}+2E +\right){\cal R}_T(r_{12}) = 0. +$$ + +Set \( l=0 \) and we have the so-called cusp condition +$$ +\frac{d {\cal R}_T(r_{12})}{dr_{12}} = -\frac{1}{2(l+1)} +{\cal R}_T(r_{12})\qquad r_{12}\to 0 +$$ + +

+









+ +

Resulting ansatz

+The above results in +$$ +{\cal R}_T \propto \exp{(r_{ij}/2)}, +$$ + +for anti-parallel spins and +$$ +{\cal R}_T \propto \exp{(r_{ij}/4)}, +$$ + +for anti-parallel spins. +This is the so-called cusp condition for the relative motion, resulting in a minimal requirement +for the correlation part of the wave fuction. +For general systems containing more than say two electrons, we have this +condition for each electron pair \( ij \). + +

+









+ +

The VMC code

+ +

+ + +

# Importing various packages
+from math import exp, sqrt
+from random import random, seed
+import numpy as np
+import matplotlib.pyplot as plt
+from mpl_toolkits.mplot3d import Axes3D
+from matplotlib import cm
+from matplotlib.ticker import LinearLocator, FormatStrFormatter
+import sys
+
+#Trial wave function for quantum dots in two dims
+def WaveFunction(r,alpha,beta):
+    r1 = r[0,0]**2 + r[0,1]**2
+    r2 = r[1,0]**2 + r[1,1]**2
+    r12 = sqrt((r[0,0]-r[1,0])**2 + (r[0,1]-r[1,1])**2)
+    deno = r12/(1+beta*r12)
+    return exp(-0.5*alpha*(r1+r2)+deno)
+
+#Local energy  for quantum dots in two dims, using analytical local energy
+def LocalEnergy(r,alpha,beta):
+    
+    r1 = (r[0,0]**2 + r[0,1]**2)
+    r2 = (r[1,0]**2 + r[1,1]**2)
+    r12 = sqrt((r[0,0]-r[1,0])**2 + (r[0,1]-r[1,1])**2)
+    deno = 1.0/(1+beta*r12)
+    deno2 = deno*deno
+    return 0.5*(1-alpha*alpha)*(r1 + r2) +2.0*alpha + 1.0/r12+deno2*(alpha*r12-deno2+2*beta*deno-1.0/r12)
+
+# The Monte Carlo sampling with the Metropolis algo
+def MonteCarloSampling():
+
+    NumberMCcycles= 100000
+    StepSize = 1.0
+    # positions
+    PositionOld = np.zeros((NumberParticles,Dimension), np.double)
+    PositionNew = np.zeros((NumberParticles,Dimension), np.double)
+    # seed for rng generator
+    seed()
+    # start variational parameter
+    alpha = 0.9
+    for ia in range(MaxVariations):
+        alpha += .025
+        AlphaValues[ia] = alpha
+        beta = 0.2 
+        for jb in range(MaxVariations):
+            beta += .01
+            BetaValues[jb] = beta
+            energy = energy2 = 0.0
+            DeltaE = 0.0
+            #Initial position
+            for i in range(NumberParticles):
+                for j in range(Dimension):
+                    PositionOld[i,j] = StepSize * (random() - .5)
+            wfold = WaveFunction(PositionOld,alpha,beta)
+
+            #Loop over MC MCcycles
+            for MCcycle in range(NumberMCcycles):
+                #Trial position
+                for i in range(NumberParticles):
+                    for j in range(Dimension):
+                        PositionNew[i,j] = PositionOld[i,j] + StepSize * (random() - .5)
+                wfnew = WaveFunction(PositionNew,alpha,beta)
+
+                #Metropolis test to see whether we accept the move
+                if random() < wfnew**2 / wfold**2:
+                   PositionOld = PositionNew.copy()
+                   wfold = wfnew
+                   DeltaE = LocalEnergy(PositionOld,alpha,beta)
+                energy += DeltaE
+                energy2 += DeltaE**2
+
+            #We calculate mean, variance and error ...
+            energy /= NumberMCcycles
+            energy2 /= NumberMCcycles
+            variance = energy2 - energy**2
+            error = sqrt(variance/NumberMCcycles)
+            Energies[ia,jb] = energy    
+    return Energies, AlphaValues, BetaValues
+
+
+#Here starts the main program with variable declarations
+NumberParticles = 2
+Dimension = 2
+MaxVariations = 10
+Energies = np.zeros((MaxVariations,MaxVariations))
+AlphaValues = np.zeros(MaxVariations)
+BetaValues = np.zeros(MaxVariations)
+(Energies, AlphaValues, BetaValues) = MonteCarloSampling()
+
+# Prepare for plots
+fig = plt.figure()
+ax = fig.gca(projection='3d')
+# Plot the surface.
+X, Y = np.meshgrid(AlphaValues, BetaValues)
+surf = ax.plot_surface(X, Y, Energies,cmap=cm.coolwarm,linewidth=0, antialiased=False)
+# Customize the z axis.
+zmin = np.matrix(Energies).min()
+zmax = np.matrix(Energies).max()
+ax.set_zlim(zmin, zmax)
+ax.set_xlabel(r'$\alpha$')
+ax.set_ylabel(r'$\beta$')
+ax.set_zlabel(r'$\langle E \rangle$')
+ax.zaxis.set_major_locator(LinearLocator(10))
+ax.zaxis.set_major_formatter(FormatStrFormatter('%.02f'))
+# Add a color bar which maps values to colors.
+fig.colorbar(surf, shrink=0.5, aspect=5)
+plt.show()
+
+

+









+ +

Technical aspect, improvements and how to define the cost function

+
+ +

+ +

+The above procedure is not the smartest one. Looping over all variational parameters becomes expensive. +Also, we don't use importance sampling and optimizations of the standard deviation (blocking, bootstrap, jackknife). +Such codes are included in the above Github address. + +

+We can also be smarter and use minimization methods to find the optimal variational parameters with fewer Monte Carlo cycles and then +fire up our heavy artillery. + +

+One way to achieve this is to minimize the energy as function of the variational parameters. + + +

+ + +

+









+ +

Energy derivatives

+
+ +

+To find the derivatives of the local energy expectation value as function of the variational parameters, we can use the chain rule and the hermiticity of the Hamiltonian. + +

+Let us define +$$ +\bar{E}_{\alpha_i}=\frac{d\langle E_L\rangle}{d\alpha_i}. +$$ + +as the derivative of the energy with respect to the variational parameter \( \alpha_i \) +We define also the derivative of the trial function (skipping the subindex \( T \)) as +$$ +\bar{\Psi}_{i}=\frac{d\Psi}{d\alpha_i}. +$$ +

+ + +

+









+ +

Derivatives of the local energy

+
+ +

+The elements of the gradient of the local energy are then (using the chain rule and the hermiticity of the Hamiltonian) +$$ +\bar{E}_{i}= 2\left( \langle \frac{\bar{\Psi}_{i}}{\Psi}E_L\rangle -\langle \frac{\bar{\Psi}_{i}}{\Psi}\rangle\langle E_L \rangle\right). +$$ + +From a computational point of view it means that you need to compute the expectation values of +$$ +\langle \frac{\bar{\Psi}_{i}}{\Psi}E_L\rangle, +$$ + +and +$$ +\langle \frac{\bar{\Psi}_{i}}{\Psi}\rangle\langle E_L\rangle +$$ + +These integrals are evaluted using MC intergration (with all its possible error sources). +We can then use methods like stochastic gradient or other minimization methods to find the optimal variational parameters (I don't discuss this topic here, but these methods are very important in ML). +

+ + +

+









+ +

How do we define our cost function?

+
+ +

+We have a model, our likelihood function. + +

+How should we define the cost function? + + +

+ + +

+









+ +

Meet the variance and its derivatives

+
+Why the variance? +

+Suppose the trial function (our model) is the exact wave function. The action of the hamiltionan +on the wave function +$$ + H\Psi = \mathrm{constant}\times \Psi, +$$ + +The integral which defines various +expectation values involving moments of the hamiltonian becomes then +$$ + \langle E^n \rangle = \langle H^n \rangle = + \frac{\int d\boldsymbol{R}\Psi^{\ast}(\boldsymbol{R})H^n(\boldsymbol{R})\Psi(\boldsymbol{R})} + {\int d\boldsymbol{R}\Psi^{\ast}(\boldsymbol{R})\Psi(\boldsymbol{R})}= +\mathrm{constant}\times\frac{\int d\boldsymbol{R}\Psi^{\ast}(\boldsymbol{R})\Psi(\boldsymbol{R})} + {\int d\boldsymbol{R}\Psi^{\ast}(\boldsymbol{R})\Psi(\boldsymbol{R})}=\mathrm{constant}. +$$ + +This gives an important information: If I want the variance, the exact wave function leads to zero variance! +The variance is defined as +$$ +\sigma_E = \langle E^2\rangle - \langle E\rangle^2. +$$ + +Variation is then performed by minimizing both the energy and the variance. + + +

+ + +

+









+ +

The variance defines the cost function

+
+ +

+We can then take the derivatives of +$$ +\sigma_E = \langle E^2\rangle - \langle E\rangle^2, +$$ + +with respect to the variational parameters. The derivatives of the variance can then be used to defined the +so-called Hessian matrix, which in turn allows us to use minimization methods like Newton's method or +standard gradient methods. + +

+This leads to however a more complicated expression, with obvious errors when evaluating integrals by Monte Carlo integration. Less used, see however Filippi and Umrigar. The expression becomes complicated +$$ +\bar{E}_{ij} = 2\left[ \langle (\frac{\bar{\Psi}_{ij}}{\Psi}+\frac{\bar{\Psi}_{j}}{\Psi}\frac{\bar{\Psi}_{i}}{\Psi})(E_L-\langle E\rangle)\rangle -\langle \frac{\bar{\Psi}_{i}}{\Psi}\rangle\bar{E}_j-\langle \frac{\bar{\Psi}_{j}}{\Psi}\rangle\bar{E}_i\right] +\langle \frac{\bar{\Psi}_{i}}{\Psi}E_L{_j}\rangle +\langle \frac{\bar{\Psi}_{j}}{\Psi}E_L{_i}\rangle -\langle \frac{\bar{\Psi}_{i}}{\Psi}\rangle\langle E_L{_j}\rangle \langle \frac{\bar{\Psi}_{j}}{\Psi}\rangle\langle E_L{_i}\rangle. +$$ + +

+Evaluating the cost function means having to evaluate the above second derivative of the energy. + + +

+ + +

+









+ +

Boltzmann Machines

+ +

+Why use a generative model rather than the more well known discriminative deep neural networks (DNN)? + +

+ +









+ +

Some similarities and differences from DNNs

+ +
    +
  1. Both use gradient-descent based learning procedures for minimizing cost functions
    2. +
  2. Energy based models don't use backpropagation and automatic differentiation for computing gradients, instead turning to Markov Chain Monte Carlo methods.
    3. +
  3. DNNs often have several hidden layers. A restricted Boltzmann machine has only one hidden layer, however several RBMs can be stacked to make up Deep Belief Networks, of which they constitute the building blocks.
    4. +
+ +History: The RBM was developed by amongst others Geoffrey Hinton, called by some the "Godfather of Deep Learning", working with the University of Toronto and Google. + +

+









+ +

The structure of the RBM network

+ +

+



+ +

+









+ +

The network

+ +

+The network layers: + +

    +
  1. A function \( \mathbf{x} \) that represents the visible layer, a vector of \( M \) elements (nodes). This layer represents both what the RBM might be given as training input, and what we want it to be able to reconstruct. This might for example be the pixels of an image, the spin values of the Ising model, or coefficients representing speech.
    2. +
  2. The function \( \mathbf{h} \) represents the hidden, or latent, layer. A vector of \( N \) elements (nodes). Also called "feature detectors".
    3. +
+ +









+ +

Goals

+ +

+The goal of the hidden layer is to increase the model's expressive power. We encode complex interactions between visible variables by introducing additional, hidden variables that interact with visible degrees of freedom in a simple manner, yet still reproduce the complex correlations between visible degrees in the data once marginalized over (integrated out). + +

+Examples of this trick being employed in physics: + +

    +
  1. The Hubbard-Stratonovich transformation
    2. +
  2. The introduction of ghost fields in gauge theory
    3. +
  3. Shadow wave functions in Quantum Monte Carlo simulations
    4. +
+ +The network parameters, to be optimized/learned: + +
    +
  1. \( \mathbf{a} \) represents the visible bias, a vector of same length as \( \mathbf{x} \).
    2. +
  2. \( \mathbf{b} \) represents the hidden bias, a vector of same lenght as \( \mathbf{h} \).
    3. +
  3. \( W \) represents the interaction weights, a matrix of size \( M\times N \).
    4. +
+ +









+ +

Joint distribution and the Energy function

+The restricted Boltzmann machine is described by a Bolztmann distribution +$$ +\begin{align} + P_{rbm}(\mathbf{x},\mathbf{h}) = \frac{1}{Z} e^{-\frac{1}{T_0}E(\mathbf{x},\mathbf{h})}, +\label{_auto1} +\end{align} +$$ + +where \( Z \) is the normalization constant or partition function, defined as +$$ +\begin{align} + Z = \int \int e^{-\frac{1}{T_0}E(\mathbf{x},\mathbf{h})} d\mathbf{x} d\mathbf{h}. +\label{_auto2} +\end{align} +$$ + +It is common to ignore \( T_0 \) by setting it to one. + +

+









+ +

Network Elements

+ +

+The function \( E(\mathbf{x},\mathbf{h}) \) gives the energy of a +configuration (pair of vectors) \( (\mathbf{x}, \mathbf{h}) \). The lower +the energy of a configuration, the higher the probability of it. This +function also depends on the parameters \( \mathbf{a} \), \( \mathbf{b} \) and +\( W \). Thus, when we adjust them during the learning procedure, we are +adjusting the energy function to best fit our problem. + +

+









+ +

Defining different types of RBMs

+There are different variants of RBMs, and the differences lie in the types of visible and hidden units we choose as well as in the implementation of the energy function \( E(\mathbf{x},\mathbf{h}) \). + +

+

+Binary-Binary RBM: +

+ +

+RBMs were first developed using binary units in both the visible and hidden layer. The corresponding energy function is defined as follows: +$$ +\begin{align} + E(\mathbf{x}, \mathbf{h}) = - \sum_i^M x_i a_i- \sum_j^N b_j h_j - \sum_{i,j}^{M,N} x_i w_{ij} h_j, +\label{_auto3} +\end{align} +$$ + +where the binary values taken on by the nodes are most commonly 0 and 1. +

+ +
+Gaussian-Binary RBM: +

+ +

+Another varient is the RBM where the visible units are Gaussian while the hidden units remain binary: +$$ +\begin{align} + E(\mathbf{x}, \mathbf{h}) = \sum_i^M \frac{(x_i - a_i)^2}{2\sigma_i^2} - \sum_j^N b_j h_j - \sum_{i,j}^{M,N} \frac{x_i w_{ij} h_j}{\sigma_i^2}. +\label{_auto4} +\end{align} +$$ +

+ + +

+









+ +

More about RBMs

+ +
    +
  1. Useful when we model continuous data (i.e., we wish \( \mathbf{x} \) to be continuous)
    2. +
  2. Requires a smaller learning rate, since there's no upper bound to the value a component might take in the reconstruction
    3. +
+ +Other types of units include: + +
    +
  1. Softmax and multinomial units
    2. +
  2. Gaussian visible and hidden units
    3. +
  3. Binomial units
    4. +
  4. Rectified linear units
    5. +
+ +









+ +

Sampling: Metropolis sampling

+In order to sample from the RBM probability distribution it is common to use Markov Chain Monte Carlo (MCMC) algorithms such as Metropolis-Hastings or Gibbs sampling. + +

+Metropolis sampling starts by suggesting a new configuration \( \boldsymbol{x}^{k+1} \). In the brute force method this is done by some random change of the visible units. The new configuration is then accepted with the acceptance probability +$$ +\begin{align} + A(\boldsymbol{x}^k \rightarrow \boldsymbol{x}^{k+1}) = \text{min} (1, \frac{P(\boldsymbol{x}^{k+1})}{P(\boldsymbol{x}^k)}), +\label{_auto5} +\end{align} +$$ + +where we need the marginalized probability +$$ +\begin{align} + P(\boldsymbol{x}) &= \sum_\mathbf{h} P_{rbm}(\mathbf{x}, \mathbf{h}) +\label{_auto6}\\ + &= \frac{1}{Z}\sum_\mathbf{h} e^{-E(\mathbf{x}, \mathbf{h})}. +\label{_auto7} +\end{align} +$$ + +

+









+ +

Sampling: Gibbs sampling

+ +

+In this method we sample from the joint probability \( P_{rbm} (\mathbf{x}, \mathbf{h}) \) by way of a two step sampling process. We alternately update the visible and hidden units. +New samples are generated according to the conditional probabilities \( P(x_i|\mathbf{h}) \) and \( P(h_j|\mathbf{x}) \) respectively and accepted with the probability of \( 1 \). While the the visible nodes are dependent on the hidden nodes and vice versa, the nodes are independent of other nodes within the same layer. This is due to there being no intra layer interactions in the restricted Boltzmann machine. + +

+The conditional probabilities are often referred to as the activitation functions in the neural networks context due to their role in determining the node outputs. For the binary-binary RBM they are +$$ +\begin{align} + P(h_j = 1 | \boldsymbol{x}) &= \frac{1}{1 + e^{-b_j - \sum_i x_i w_{ij}}} +\label{_auto8}\\ + P(x_i = 1 | \boldsymbol{h}) &= \frac{1}{1 + e^{-a_j - \sum_j h_j w_{ij}}}, +\label{_auto9} +\end{align} +$$ + +where we recognize the logistic sigmoid function \( \sigma (x) = 1/(1+exp(-x)) \). + +

+









+ +

Gaussian RBM

+For the Gaussian-Binary RBM the conditional probabilities are +$$ +\begin{align} + P(x_i|\mathbf{h}) &= \mathcal{N}(x_i; a_i+ \sum_j h_j w_{ij}, \sigma^2) +\label{_auto10}\\ + P(h_j=1|\mathbf{x}) &= \frac{1}{1+e^{-b_j-\frac{1}{\sigma^2} \sum_i x_i w_{ij}}}, +\label{_auto11} +\end{align} +$$ + +while the visible units now follow a normal distribution, we see the hidden units again follow the logistic sigmoid function. + +

+









+ +

Cost function

+ +

+When working with a training dataset, the most common training approach is maximizing the log-likelihood of the training data. The log likelihood characterizes the log-probability of generating the observed data using our generative model. Using this method our cost function is chosen as the negative log-likelihood. The learning then consists of trying to find parameters that maximize the probability of the dataset, and is known as Maximum Likelihood Estimation (MLE). +Denoting the parameters as \( \boldsymbol{\theta} = a_1,...,a_M,b_1,...,b_N,w_{11},...,w_{MN} \), the log-likelihood is given by +$$ +\begin{align} + \mathcal{L}(\{ \theta_i \}) &= \langle \text{log} P_\theta(\boldsymbol{x}) \rangle_{data} +\label{_auto12}\\ + &= - \langle E(\boldsymbol{x}; \{ \theta_i\}) \rangle_{data} - \text{log} Z(\{ \theta_i\}), +\label{_auto13} +\end{align} +$$ + +where we used that the normalization constant does not depend on the data, \( \langle \text{log} Z(\{ \theta_i\}) \rangle = \text{log} Z(\{ \theta_i\}) \) +Our cost function is the negative log-likelihood, \( \mathcal{C}(\{ \theta_i \}) = - \mathcal{L}(\{ \theta_i \}) \) + +

+









+ +

Optimization / Training

+The training procedure of choice often is Stochastic Gradient Descent (SGD). It consists of a series of iterations where we update the parameters according to the equation +$$ +\begin{align} + \boldsymbol{\theta}_{k+1} = \boldsymbol{\theta}_k - \eta \nabla \mathcal{C} (\boldsymbol{\theta}_k) +\label{_auto14} +\end{align} +$$ + +at each \( k \)-th iteration. There are a range of variants of the algorithm which aim at making the learning rate \( \eta \) more adaptive so the method might be more efficient while remaining stable. + +

+We now need the gradient of the cost function in order to minimize it. We find that +$$ +\begin{align} + \frac{\partial \mathcal{C}(\{ \theta_i\})}{\partial \theta_i} + &= \langle \frac{\partial E(\boldsymbol{x}; \theta_i)}{\partial \theta_i} \rangle_{data} + + \frac{\partial \text{log} Z(\{ \theta_i\})}{\partial \theta_i} +\label{_auto15}\\ + &= \langle O_i(\boldsymbol{x}) \rangle_{data} - \langle O_i(\boldsymbol{x}) \rangle_{model}, +\label{_auto16} +\end{align} +$$ + +where in order to simplify notation we defined the "operator" +$$ +\begin{align} + O_i(\boldsymbol{x}) = \frac{\partial E(\boldsymbol{x}; \theta_i)}{\partial \theta_i}, +\label{_auto17} +\end{align} +$$ + +and used the statistical mechanics relationship between expectation values and the log-partition function: +$$ +\begin{align} + \langle O_i(\boldsymbol{x}) \rangle_{model} = \text{Tr} P_\theta(\boldsymbol{x})O_i(\boldsymbol{x}) = - \frac{\partial \text{log} Z(\{ \theta_i\})}{\partial \theta_i}. +\label{_auto18} +\end{align} +$$ + +

+ + +

More on RBMs

+ +

+The data-dependent term in the gradient is known as the positive phase of the gradient, while the model-dependent term is known as the negative phase of the gradient. The aim of the training is to lower the energy of configurations that are near observed data points (increasing their probability), and raising the energy of configurations that are far from observed data points (decreasing their probability). + +

+The gradient of the negative log-likelihood cost function of a Binary-Binary RBM is then +$$ +\begin{align} + \frac{\partial \mathcal{C} (w_{ij}, a_i, b_j)}{\partial w_{ij}} =& \langle x_i h_j \rangle_{data} - \langle x_i h_j \rangle_{model} +\label{_auto19}\\ + \frac{\partial \mathcal{C} (w_{ij}, a_i, b_j)}{\partial a_{ij}} =& \langle x_i \rangle_{data} - \langle x_i \rangle_{model} +\label{_auto20}\\ + \frac{\partial \mathcal{C} (w_{ij}, a_i, b_j)}{\partial b_{ij}} =& \langle h_i \rangle_{data} - \langle h_i \rangle_{model}. +\label{_auto21}\\ +\label{_auto22} +\end{align} +$$ + +To get the expecation values with respect to the data, we set the visible units to each of the observed samples in the training data, then update the hidden units according to the conditional probability found before. We then average over all samples in the training data to calculate expectation values with respect to the data. + +

+









+ +

Which sampling to use

+ +

+To get the expectation values with respect to the model, we use Gibbs sampling. We can either initialize the \( \boldsymbol{x} \) randomly or with a training sample. While we ideally want a large number of Gibbs iterations \( n\rightarrow n \), one might decide to truncate it earlier for efficiency. Doing this while having intialized \( \boldsymbol{x} \) with a training data vector is referred to as contrastive divergence (CD), because one is then closer to approximating the gradient of this function than the negative log-likelihood. The contrastive divergence function is the difference between two Kullback-Leibler divergences (also called relative entropy), which measure how one probability distribution diverges from a second, expected probability distribution (in this case the estimated one from the ground truth one). + +

+









+ +

RBMs for the quantum many body problem

+ +

+The idea of applying RBMs to quantum many body problems was presented by G. Carleo and M. Troyer, working with ETH Zurich and Microsoft Research. + +

+Some of their motivation included + +

+ +









+ +

Choose the right RBM

+ +

+Carleo and Troyer applied the RBM to the quantum mechanical spin lattice systems of the Ising model and Heisenberg model, with encouraging results. Our goal is to test the method on systems of moving particles. For the spin lattice systems it was natural to use a binary-binary RBM, with the nodes taking values of 1 and -1. For moving particles, on the other hand, we want the visible nodes to be continuous, representing position coordinates. Thus, we start by choosing a Gaussian-binary RBM, where the visible nodes are continuous and hidden nodes take on values of 0 or 1. If eventually we would like the hidden nodes to be continuous as well the rectified linear units seem like the most relevant choice. + +

+









+ +

Representing the wave function

+The wavefunction should be a probability amplitude depending on \( \boldsymbol{x} \). The RBM model is given by the joint distribution of \( \boldsymbol{x} \) and \( \boldsymbol{h} \) +$$ +\begin{align} + F_{rbm}(\mathbf{x},\mathbf{h}) = \frac{1}{Z} e^{-\frac{1}{T_0}E(\mathbf{x},\mathbf{h})}. +\label{_auto23} +\end{align} +$$ + +To find the marginal distribution of \( \boldsymbol{x} \) we set: +$$ +\begin{align} + F_{rbm}(\mathbf{x}) &= \sum_\mathbf{h} F_{rbm}(\mathbf{x}, \mathbf{h}) +\label{_auto24}\\ + &= \frac{1}{Z}\sum_\mathbf{h} e^{-E(\mathbf{x}, \mathbf{h})}. +\label{_auto25} +\end{align} +$$ + +Now this is what we use to represent the wave function, calling it a neural-network quantum state (NQS) +$$ +\begin{align} + \Psi (\mathbf{X}) &= F_{rbm}(\mathbf{x}) +\label{_auto26}\\ + &= \frac{1}{Z}\sum_{\boldsymbol{h}} e^{-E(\mathbf{x}, \mathbf{h})} +\label{_auto27}\\ + &= \frac{1}{Z} \sum_{\{h_j\}} e^{-\sum_i^M \frac{(x_i - a_i)^2}{2\sigma^2} + \sum_j^N b_j h_j + \sum_{i,j}^{M,N} \frac{x_i w_{ij} h_j}{\sigma^2}} +\label{_auto28}\\ + &= \frac{1}{Z} e^{-\sum_i^M \frac{(x_i - a_i)^2}{2\sigma^2}} \prod_j^N (1 + e^{b_j + \sum_i^M \frac{x_i w_{ij}}{\sigma^2}}). +\label{_auto29}\\ +\label{_auto30} +\end{align} +$$ + +

+









+ +

Choose the cost function

+Now we don't necessarily have training data (unless we generate it by using some other method). However, what we do have is the variational principle which allows us to obtain the ground state wave function by minimizing the expectation value of the energy of a trial wavefunction (corresponding to the untrained NQS). Similarly to the traditional variational Monte Carlo method then, it is the local energy we wish to minimize. The gradient to use for the stochastic gradient descent procedure is +$$ +\begin{align} + G_i = \frac{\partial \langle E_L \rangle}{\partial \theta_i} + = 2(\langle E_L \frac{1}{\Psi}\frac{\partial \Psi}{\partial \theta_i} \rangle - \langle E_L \rangle \langle \frac{1}{\Psi}\frac{\partial \Psi}{\partial \theta_i} \rangle ), +\label{_auto31} +\end{align} +$$ + +where the local energy is given by +$$ +\begin{align} + E_L = \frac{1}{\Psi} \hat{\mathbf{H}} \Psi. +\label{_auto32} +\end{align} +$$ + +

+









+ +

Running the codes

+
+ +

+You can find the codes for the simple two-electron case at the Github repository https://github.com/mhjensenseminars/MachineLearningTalk/tree/master/doc/Programs/MLcpp/src. Python codes to come, only c++ as of now. + +

+The trial wave function is based on the product of a Slater determinant with Gaussian orbitals, a simple Jastrow factor \( \exp{(r_{ij})} \) and the reduced Boltzmann machines. + +

+The Broyden-Fletcher-Goldfarb-Shanno algorithm was used to perform the minimization. We used \( 14 \) hidden nodes in the calculations below. + + +

+ + +

+









+ +

Energy as function of iterations, \( N=2 \) electrons

+
+ +

+



+
+ + +

+









+ +

Energy as function of iterations, \( N=6 \) electrons

+
+ +

+



+
+ + +

+









+ +

Conclusions and where do we stand

+
+ +

+ +

+
+ + +

+ + + + +

+ © 1999-2018, Morten Hjorth-Jensen. Released under CC Attribution-NonCommercial 4.0 license +
+ + + + + + diff --git a/doc/pub/BM/html/figures/Hudson_Bay_data.pdf b/doc/pub/BM/html/figures/Hudson_Bay_data.pdf new file mode 100644 index 0000000000000000000000000000000000000000..6a81e9928655c0cf601697d11a7d199dcfce26aa GIT binary patch literal 15045 zcmb`u2|QKZ7dI|LxMUuZhHK7{JKt;0JkLWEm21j8T$jutk|Hui88d4#N24J!6b)1& zMKTkq=xOlobLDv=|KG>w|Gw8}JNxXj_u6Z%z0O+Sea<=}#_F2;P|{eK$khS(m0K_r z90~WbbA>4=z!Bzw-i~mD8j(zN_c{hg7!!{<`ohtmgDG4|3FheO0Ckl4V}O;BovWig z8BU<|0RzZnN0KKT2c5zY#w0I$Ge6K&r~^L?M`#3) zwamywvLhV3Lcq+K=-}dc431wRuj%DUhGQu8|JXqxLBpy5S|l$&Z#X21ni(9S?&$Ag z?`Wz8CIe^YB%-IUH<9G%X&<;M>PmklB}i6qfxIFt3}N8t;6haO3IMD_b!*`8`k0b`*#6xyPGwr?~^TRE^ z^+MVGJw;O!!*tkLyn1)N|J7aoLO+Hl9<*)6NIqJ_VUo4=f3V4ep=g-!$ z^CGYxueG=})bldOJTU5GBl^W5Q?JtUTvG@`?-fMc<(8A*QShz-MIo zFg7G*(4|%NX@fMozh0}os-zg_^gFRSbbfvhk6WZLZy^6n3OP(rxAN5s6^sO0x2R+x zY2+P0U+#ycUEzY-F&QO=+p-P#lO3v)YjsVB568Vs3b&+#YabLp_VMu=KLFaU4;HtI=RAKAYuvcS-n5 z+_YFqGnerdNw~4v5?6tjr=Q0QERU(?6TaG=#_zT_; zlSRukl|c`?6kem+rf{HLcyGI`v$>#hsET85v~E$f*pB4ANe(by2hpeR%1bJ5M(cW4 zn`B3iFUnR zxvzg0N_QVJ*`xjZ88-TU-=?_!7jFd~ayge%{jw$;%1zB=%oo$jo5(=QGri06D!x{H z-znw#tsNN;XL#?v9{d%TApa?<`n3wRG3OvFN4eTTk$Z;-yzc_UF^6&){j_O#3;9y1 zsI`yv^D31VgIpRMvLm{adg6^7r)TkMz48 zoO$W3Sf}#P?&%igZd3ppX5b1+{KhU%p>o3`zovRWlrjZaT&W4-DXHT3pY5YpZj5N> zi#&MG&90@-h>LJKrQ4#``aW-GRr#wU?S$qBKLdRZiIg5n|6Kbsz5kuv{g4~e1Y*to zHu|r`^knsrmlLjUJzw8A9#nhuKJRxh=xDpvuiNfNeH;5+-+H|!)V&Oq`Z~2(rhSQW z>FZN2pRYa^d`D*4o0zS!hnHIk6P-V`h>p=6Rzfrt^7kR^6RIF_sf=L)bl@r!pP;Xobb@;;@SPd2=qq zHjr^_cKqiwRSb=E!Cr$>oPth8+?M5l?_*=lAC8_F7$hT)=6I@nuaouH=O3P_LYzJq z^!-Q9DD#oYWQ`=V2id2!PKP9I7etwKV`A_pY0k?x6`8eWDK4S*A2aMu(EQw+*7wMM z-$iqeO8Iw==C@uPOXj~}a79h)O?pf+RUh+EU)7-uF|m_fcK9gkNY;cf+71Qk8n|MA zjlR-=U$2Pq4Eyeao}iA}47qL;Mi#Xi0#Sd) z!2jvbv9eI){MVV|67*PT1enzpWup;D)37M--DTH{7T@BDhZq@|`VoyhC(p;}>Dx}H z+YgDD8(mWp7D~9KBgzvFV8(~;(< zW?isLzd8qmT^%(uXwXyUH#sXSZRtkoY7`bRvn4j;Q88H^<>Q8^%L7iDJHnRxKmVc* zMkDPv5P7rF3Wr`J90rduM;aq5%#mok2@;Q`w&RTEQj^YQIGx%_$Jx21O1X?wm)qfM z3l7hyCK$Ih{5?1N&uIR?G%JH#Gq*%;vXd@|n)}uRcLkSAGUfc_RC=s#RMGLttrpm-vL9lAJzk zt@Qm0L8fbdBGVlg;n|WO^D-(cp1vBk_1+%Lwrc z9=~VA7vRcA@~352Yqw=|y~Sf-RjrVdXt-r~Ijnhh-$-84LPs|LH~!3-Ir%EjAu(5f z!KWl{bWaub;dTa2jok`?8b#j2RWFYBVmW%8m<-dlTsPbw$3kT*c;sOC^mTPvy=UhN zbG+_6Tx3gwsmlbZX%X^FQ=F=;g^B>$Dq%5z_eek+YvsLeb!>(E7)H^G?n*0 z$Jku;E_pd8`1N-tDlf#;2C_j=c~i2j6%B*0Hb#awgZqR-_pNQ&c$?9{hiOR= zQYnoO_X=-`F|-$kJu44qkC44ve&+5dlPpieX~(RKQp_ICBDaIiMK?UjtSTh7zRZ5P z@QYrf1w!@E~N0yCS``LDLgObB{cHgD<-|0lXP=J#<$u~u_ z-yTK?@(Ivy?VIuA!_fx6bTe`?MoaSgv_GIT^Ly2n(QH%{$24Qpy{)aD{e|Q9f=j8b zaZy4B0;bw|lcJ>K&?@rL81YRssP)`#mvjR$h-xbgav1tiJv$9=e`X` ze%2FED{iLQr8TUnb*gJTpu8Q$a)${mdL(`!CvHmk9OERnN8_z(`Ps)(C^0EcAzQbU zflFFkOGD{QDv{Hlnt8XY7r3Q+;d1<3?$vWemI02g zj~7Mnn~U>Zl1X7T=RIH9#~58wXm`QP!Z~jEfa1rqM`pflY5XexeFK>{3r`4Z?_3N4 zXNoi~ucqQ~DiVc_YNo-&PuVCdf$;6sIBo(o%5_CPAj>OvJ_!+vQ0APpk4)ZN5w*RYBuL zU)vxET$B#69r~LE%x1v_0k_5!aPbxF09?>O;G%O2=hybartdcymz)cDceag;HfC3n z%MHfpKaUj{EU20!mLZ>zokD$_M_3N9@E&p2y>Vu|Tje8t#w70sTET1I3^Emg~n+YEawUBa3v@=HXnT;V@7&zJR! z@9rP?tQfw*0&E(itP8{jCbvD*K_Gt1W%G_|$}KRP$@K77FST}t#3pziLcCan=WDAT z^ANo?>%Ecl7HTJU??+8zq!imndiIajx0`d_<1V|DZp>>p^wXL44i6b2DA_^(Ho=kE zw`=iTKHCLbQLOD%FC9_d-a8nFrr5l5%_^77w;wm)lDf|Q7}GX$fKVw{1$Pe+;T&$I zQm5_XO?+&dA6Jo@*h1P}0!yMhWczZ0-{fqhWXcOh^5qu~ZasQ;W$ddzK@IWgUCFJ0 zQ06JU@}}{^V5yMns(T)N)|Wr7BXoDJh?CukmCiIH>wK!5qFv}s<>eel$))e@-3vdN zxcAyfY@ol*0vVaLH#W|o9x~vmj7auJ`LCvv^*&6$-8x;TA{Poz;ddL^5;>IRC9qV4 z;EVEOFp{_}BIllWstT4AJ7kFuu(*EaYfVz@w0jZiRj%VdPrGjFbki3O@~3aW-e$4L zdcT9gf{nuyI(Da1t?TY^TWY5d-hBM&;_2ZBi?)s$qtaP&ua=fZ0 zdG?e+^2IR;J-(Yv+08YvV`-reSP9q?dcSzRrfqL}oe#nud;2ZR-_`0nj#J>hWIox$ zo4{+rJ^D?+=yJ@2?r@00UICk6q$Ur?h}l!+b5%CiJ{?j0%5rl@Ws{1l&RCV0!YjWZ zBWX|G$84+)_2x^4Cet%dF!!c)85KEdhpc8pFjD1Ot>1+&&#HB59!v;xex7bQPOlr~AuWg9*&U08OL!sr1_Wh7i#Gb8>yVy^ZRm2QG9*9&1W#ur$AL{4U~fpLXc+ zn6fY@hu|vh$BC(_a$i)jqOSzl6@|beO8K5F5Wi;n;oe^BaOug^T0{Bxq+B7cA+Hpn zp$TqXe*^a&S9;`f^B6j7rBBOc2w(Om`nb-g2?*x++2bBAoi!BnfB62_RpgY3aMH75 z^lyAb^9|At)5z8bmX#lnkUgz*c1WxqE`xCPT2lW{owxA zSsTF(w6|RKiWIM5 zElm|gru)3n!AiIb(~Z8_v!^gQFM&8v}2 zdNH;xkCgd1)mu8|+&<3SP?Rq+LxzOT!=Fh?NX`cjZ`ty1xEJaK*;(Ln|>lBvba^`>0oXb5(c_Xp* zX$@wx$OnyHYjZdw`AQH@r(~n06ZzsiY`d#6%n^%TIS!TEIDTn=`mC-cdStYX;cDM} zEf3#mz9k;B`5THe%~6wzbukPdhE>yv!kM_Kx=Y3HTkaqJfO*k)d)qy`G+x?TWb&8P zmW)pSMynlSK8BV2_Ak}l?GH~)aJ^vH-%BHqHG>ps+sQ2+TA3dwD`z+oJ7!nbB7OfO zF|@G=b-8J$C7sa|2jeuwkD|G|ct5}Gjeo0yJ$r;2ecqOsB|O@7&~T~OP3xh_SwWOo z0pAFk{aNO9J6IlLi<~@6f}8Hbq;JO?rDf;C)&?6`+GYV63WHhW7!49sIGO<*`WP0S z9bU0zA)6ee`|HHMMEoq>OkDW+Q%D8{JLV~*$ep0?Lly;k-uQFhMEEom7+jeKc+Kgw z?9XH^y?T>3T{t|!WpH+i-S>HQ0hwST;c+*sZK)wNXK;^2t#DT4ZtKY?>8HY9CuIbhvf?>bzbO4V*!JF*{Ny{Ktqw#zhO#@(eazAMbym5PdKMIUuT3%U`q-XfPkJ7zHfB@a`9`Iv z`^TQCMONc8PNz4pnax7Gb#8w{`%PmzAkC{UMgNHUX__}oFncRiPAGHD29s_Uu%S`w zu#xd?cJPb`JQ!md=?a-4baxhZ{_+scEKtR(a->|;Z%>Jl>qxpPZnR5D&o1Up%m+^g zyoiJbBDq{-@d<1`GMxu;=*K+&_yqUpd##hqT4Pp6RQu7CURxwCuU9Ye)) z>bLHGJLBe4BnMOX)^z=JhD>a!Y*Q=frNt@IxYbZy9}N0NaMmL8g->(FzQ-2sR9k*g z_i&0P=8WkE4}F2EH~H)UzxLUIMxq`^GaNkT&s5c#)7l!6O}?f3@%}XdgR1c23gxZG z#b`@w=s2I!Aw2YMF)3LycC`kK<2`X4BeZ_u2MXRhrv^gRF!%eOv3aK>EU=|Z zt=D4gSARvb`HaM=?I*NZCn_y=Pnp^I#ZG@JG`P@#4&*NRVAQ{C!c6VC{G?5#`es~~ z-6+S4mL_3^4_G07v6oVN66B5eouXSG#J|sy=CjS=*@Cl9eHSOw3>CED%7f#D&m)Pna7p+O;w^8cNqwzU(7gle7HreaJ&VzXk#3%YbFOF_` zQPN&{unn$rN}ww=KR5*zU(yjM$=$@7oX7sv}WwMS1 z9bK7wr@{!SSzRPdDzV?hZTO|8Q*P7TjVq6$KWg{4$(?p<)39kzYdWSbHG+De@F4PI z{}SKsZ&x3+SJAg3`WE>POqVitsYt3l1Lu`5CW_X;##D?0U8TeGIQXr*$cot$y$>KxIVlCpgG{u{h+kXv5W)iV4t z@_1{Y^f|PLqq-aM*vW#{man6)5(;S#kGDD&$WVXFkZ7XOz4*xeGW=KDgg`Nxt?2rM zuGmcz?&Fi~HdXa^vUJeaiyCJhb9GHMWqdiFj=u6(J}`~5&`@KEFSwR=Th{x@KDUf$ z&6~1~k|H5^{~*$*yB80#wwG#ab3eZYGt6(}(tDM*bn(o^XFZ*BZ!awJomk$284lU9 z!J=#y0itkg5g>@YkV{kdwNPQ?C})_gc=+`Aaw+dO-D)qYeK9vg%*ytN2^I~!bpDJC zWXUlQIUw~)v^IGMiJwPB7nL4hh`N%0GcO~iXR5BGtj^CV^|?!DN6U!?XG=`7ViRss zY|dbODMz8Yblf`k22)C?a*a%ujxj3qhFwH{rvC}S%ssH_z1czt+(63BVmAzUq4uZg;Y@D(v>=(MaCE$LSxc$d1m7DS{r9&i%X1TV&?=YR zanP-|nM|^%Y#PyundCC~res_6gXK%&`|;g<0=LXwqy=wh=zO41c+ScoEx3*@0>Nw$ zTOYl!$S^LsH6O_tajWI+e&s4*q>J~QS)$vuLw2z&3Q{~)Rg#ac?GQI_vj~yL*VEIH ze5#%l@4U!$b0Big@w9`MYsHODNAXX$zc6l(;vPM1#1$Jo7Qx~qI4?$H7yPx)&e9T5 z#?JRbCx+gX_L;jU2NJ7MIeC)*mE}X+ej}~E#&(taXBwCAG8qV+z@v0|Yze%-Si1Y# zDw5QMxju!~G@KT44`gV> z#F#suxQ$(qCJ&+y>hDq`cBOZ;rJ--~57io+yy~86e(j{c!dBMWV#g~CVWl3#NvV0} zCW|9=54Y`4)Vp^4uEo6*s%FRwe;XBU)36n6v;LGtF+qie1_d?*fzkWf#QjyR2CFQ9 zk3Xt+@y(GZ3~?wPRW+{?SC&hW+a8`R)GK*0Cw?YYm!&PkmZh-Z*{);fgY0Cs5A3ey z?;5U{_=o*=(t*$?qnfuwoFlcS&0Gf3d^jJ_;OHjBCvEcYu+`_3TE4<2y{WEDa+4`~ zdv~mKYWcav;L)RRrQwd^?9##6Hz%wwJu)wY zA!uq*DHoVaMfOi7oyO>geIXqlHu-ed-Syb0z^pXBXvvCn^wz0Yq^B3{DvWlvBigFX zt?~}M$yb|?-pk2KweMX7-O#lTvgc}a9@*~nwlpg7dsnu_rkC{P#zdtimF;>qA6GDE-FP^bXl!lWPaEL$ zgspGzd^|g5<|&`gL3X$OiWB)J=o%9~R)b?9^Ka0m3r_*rXKimf)Q)wlirET}>ZFf( z3k_xWt1zKH``kz~a$igrKX87IPQGIc&E21aN%7sm@1){v`lt29=AOC~@8rC8Z@1mg zll$HEF4;VkjM3mKxwDrc?4=5)Pl#2^qC>v=@#-xvggAShMc9}*n!09?p8{F~K@6!L z{jBbsufDtI9J}J)%9anA+~vhTu0F2VnawI1>{XqR^5{`X6nBq*P)TjK{jQ$l+28;9 zu|=9Hk8gto-7HK-gOt5BHfpMOZXC(bKgMMPH(e4{G}5Y?yuJLjulgc4kFC9ShhADq zq5i&P^VGw7VPx_w`WpdzL$ka5%jhGuQJo4I^vw=M=8+8Rt54j#daf{0RcO!laP*6` zrs=gLImBF@VLR^dWT-`=@6pohCKiI*IW1FOi6h1Ibl>8TbB2#X@5MR_U@iGPZZ@c} z@|4ghBAGbr1>e7M<5Y|fV9LG9nsNWMG@-{P!CgPxtnM|_U}?$=6f1Y^<(<=-JMQM> zkILjV_bb}TekCz#Z5#@1^;^;{&=efa>p@l`g?D_9|1VlYK{#Gt(4GILgjOvcQ_9Pc? zvKI-CqU3bK5r#w$kglli=t?}`X9f}~VF)NI(||}OxlpntrIAR;jex&@YanY{9BphMs;y`xxg@e5XC0*3a&yGwYS;Rs5KsJo*RNL8g&NiN5n$#Ae2KsX0_J3D&9!JAe@kR!tZ(B0OC{M*^5d8xG78p${hj5{7V)!-6n| z%gDkJ2jQ{=ID!aA*uhaqAddqW3TS~7ESMIKZ~-+i#2t?CfFnG?#DF&-8Mp|h^aVV? z5q@wKi1mm7I3f_#a4>`&2}rcUJP?0Hb^j^>+!_dd;%^0_SGMM>8Bc%idLRpc@?!U) zF@SMM^Z!%LAkmiW=y3q7CzRfL%onH@hOp9JQ3GiFZvp=>EehH7b_GWm5CbR%0!Lx6 zfQlN?`#(B5AVb;3Ud8j6yCXE0$_kURXv(X@RXJ8Sx7Khx68Zn4!2bWz!GMG@EVwVQ zcqCi~1W!B$50eGZfdDK34PG{4K-LgcaDY$fI@BKn_LP((csdUH|E~gDy^aBWp#srZ zEjR!?fK34f$Y4PMgk)vl1Ohk*W{(4t6QBZ8xzMsW7$|T&FkBn}$1->foB(W{0A42H zk)RG8q3ak>Kw2CDjE88-0t3d&th7N#iZB=g3_3zgKte$UkHS#K;(*{QG(cZ4FQtyb z0!7GB3XUL4Im!|Mzn~qDLIVL;=EZ_}ap3lb!Zmaa2aYm8S`-MG4RmnxC?0 zkOBb@K%p>bP8=9RQOYWF&=ImT3O88j1PQ4KVjB_-I)Xx34`{lT0>k}YvwumgQ~-l3 zB+x)&FboHA1vrO>LuRnLT&qX0WUy66C={?D7XgS(VIQ)Cf9XTLAekt}1Smo0|5AaN z0?vbS?Kc5gyexdRK!&hl1CSl8Efie?yMS!sk3z8r$R<_`#a4dX1Y{3?7Kks(IiLaA z#%h5i$0KoI4FRu^4Qx|#9aEGnb?0`dAsX4-dhoD?_fP*EY)SSQ_wNi5d zwguH-Anag&6Ze2qp7MeT-N9!NEB(CS*xz%2fcSe3Z{XlouKL2k3w#O{5)>{ebNGR~ zdZk7NcDhmv1h>*kEdboJE44r37H*{p1gqc62A!|2nkKj}R;K&!`l`!^z5r1{t$YZg z0>-R{zkfe4LD&=Flnog|1-;r0_5a-sf|9>k_#-<0*9h_Po3H;GA!%2cH2}>1md^k< zmzCB1uPdctweN=MrT>B>O)nC_0KdN%0W0}iwBKLiz!0nP86yo$Z6D?TfBz1z!NeHz zwqdmK-0(Bp0i9zbR z_ar7$LyvKXp(9&Ig%4t@cs#!P30RG%kJ!|H_ov~ifiHi^E?%DM(3enq)#cDg6c&X<$)NBU z6k0|Si55d5#l$I}S2_3rhzH;6K_q*-dy(B;?BL!GPH;P-y&JGr(1k(@p!Yv}prkR< zC@Huznd~ixK>RmU+KY5d{0~D9AUQh0K=8uB0DAcS1KiA*3ly%|K^2&{(BAr1jN^m#RI^yzD@Qo8v>RAMHcAy-x|r_AWX8ZO%{s3 zYujX@r{L?`Hn2%q0_36Ajg|dN2VlDjL6bH8P#6LPwpQE7B%+JEBZ>04sg9=;*g8Y5 o4?qi3FE7YrQtDY$kL=mLByFm#-K z@7CY{+&kyaopa~Ry)$R#Fynl`=;yQd+H1Y*eV_Mvp0!`xxvhkcM~R0*q3~6d<+V^K z3_27F-2mqT{7s5$fFt||(^*zU2M2zEn7P4F$zV8Qjx!@ z!DZZEdMra&S( z2IOZDEx8fn@BgNcV3QF2`@5?wbn?hghQKYEP~=DXVUiyaG5o}Q;YFkW`=L||bhron zh_zt`{{4_kFfI=A8RC^o|Jz^ay}s#_C;a#Mugc2F_3Vnq#KgS(eBYqNQX!cpD__!c z|I*M$b-7frr{8K0b5fPIu5)*~7NcQ3`S%(_Xi0$+s?Wa&ijV zjuOrIo;MJDV7$A)bAco77K`?gfS_PtLxUtbI(o6|ZzJyz{rj;@@wr|W()&y*T4%cV zotM8TxVwwy>K8@zOP^EC&CPA@?sAozC`fOF50(<(eXJg^cbiY+is|NODAv%`0Gb{;M=qn+_N z;eIRWnc$FI_OL(QVW!d1(a|qC`KpAZBry$5)Y{sbS5JF;J6rTUE32TTzSj#?Gd`UA z9uMx{H~*e*_;aR7`q#HSY&ki(PxtxPCTq%?pL9jjDy2zz->IE|1unVFfvTC{X_Uidv;(U!n%yghaH z`4ub9#fx%NwNEV`JP3ySrJFVRi29y+HCowER{OU+$NpecVa+Qfq!39d#QUI^A|op+ z&raO39cOfG%%U|A_h{YW*KiJn;3J~-$(rX*3!N6L-wT4?vg+Om$R>(q!cD%#0{ag3 zTVN8uRTvH)-sRxnVA)tEKA%%}?+}A`85vCUDzP~D_}t%oTsNkR+YNZ(%1T%+jE)z_@-Vc5sA0JtkSMe4)87dYBKY9{)%&x=tIQ8<`7Cv=< zfq4mHQkie|$2cZXuvz+&$_f9dsHm83+VHFuRv%`6>-1>HtLN4mQBCDoCgi#h?;Q`* zyjfC&?Kt4mXlQ7ZGffaNq)SUnt?ljdyBjlR$j1#oK3)AjzcQ3%ZFS|k2^)NLe}8}e zm4+1Yy{;&->({R%j|MYmnY#Rfgf(u4wXm?T_3KyUHm>WF$qvb|WpEi88R=hrmhXQn z*|D7FdmSAEBj^?j34D2eyuyZK!eOdb7a5j*A=mNl3L}ep3X|0FPsX_)KbYWt+{Trp zV`F1RHNsf9moAZh|Nea*7O{ilptQVPa8bzl$&<@)PZ&mt%gP{nkN5uAL-=Lx*DsFF zC~8i{2nu@S!f>JRE8JQWQsUyq<<{R+;eKA;-W}~PNIE<>8dzc01A~HsL|uNqG)h)W zypD^FjSb86^IKkGlHlXmM>CBLFPT)_pFIm39@eSgAIj1ok7as+OaaCDNA=f%Oe&TT z8$l-Yx!Fm6cixlI`tVHDIA=jPwl+7*m^3`oC5$19tR#84x#`HQpeOY$TT^a0Tl0f~ zRg&M&uGJ*Jz~Ep!m00;FPf8o*zNR37&=x|-=}S>pQ$v>IMg1$v zUrkF$V6py|tGg!8%gwF6HeR8Yk(r50A4nnUXzq{u=@qL+;PdCtVchf4(r0%J$;rr= zVe0(-{W*F_NJu)m<2e;>-;V0uhOnAjSYU*IgVdmBs2^*i{W4F#2oJ(v>TvcsL*i!* z_}plv-Ro-alS(4l(Bk4^S(2CDC;KpPdDxE%&z{vj@qp*eHmNCv{if@A4jEGpp7MDA zcV6jfQPbIFm|~G>cex*ni#hqL!#S@y10Sr7mcZ(6ZFSM+btJrb!*!oOnL^Cvk^i#y zTq_o=7zO`*W&a7yjt*?us^zq}zNaUou{vL0I&>9^d54)Wz_F{B&lp*y)U7hFq zt~*S0H8nN!t3^#$Dy+XTuy*V*>*NNNpFtRy*C} zbC{Nhl&&|e*GIC!uaP32-rioF^&q$dpVPej+{y|&{6%0zMa9p7_s>U5?uCq>m6nxt z9L)IgPzqTy+1l9!B_}64Iyn);(p!(PNO#nP1m01%)OaIGU+c2MkY7*`1fzO1QCTor z<&f67FxlvBe2GTFKvYx|M?gT}YqlmM@ztw_zc86ZMD#_)#lvBeUMNM=JmwOBTnY;{ zT5R$BN%Fj;`}vtSpUclXb3cEw-MV$F{hCAl`~5aHR#qj!$Aj_^4#IZh8I|WSvLe+Of9t1D%8Tz?qNv5?vu>vb78Dfl^6?Q@*o<6( zRT?;K+}_zCfpB`*|Mo>_D2`^?19fC2Ad@p*;W5phTo0l3xicO8<;xcaF0Na@yNsMj z6Z~!^^DDprq|Q&BE1rh2E(zIu*D*YUa*G2YT<7|GxZ9-W3GxYuBS9Lap+6P2WBf>R z$HisuOwQrpLep$tTV2g_@@@;phm6PhgunLLo^_;(no6PiO{fx(HmuXc$K82InbEHD zn#(j2QhvOhisWPMtlPTRdwP0E;L(mudn)ZFAUm`d-mQ!4xqi16NsBZzG}aRjA3WfY z9uo2{EPwP>7HS=tfp|s%c8HG8)n9eE6b(ja$E!TMD?`{9E?oFIRVM<)-s#6DVlLBq z3g16_B&Y@Xm~j1;Ej)&neVN7 z5~>~N&{KqLf?-$-@W-Iao_hM(DmRx=ISx#w_S<_%d4?!Q-Em#3C@EndED0pSAs|qc zl|_g641j2}=t${Byj5lV3o9n1tl{msIh4 zpJSOJDQ5gjb7_Y1sdGFz+2l(L6%W$<ADnO*tr4^c)JdUCun&jWkq*tm_={crkaIra95JZ7DHt0RywaYaN#BCU`;0ciEZ z-8$Fqdng&W=x}9KdAWR^UIEdyYvFPs1n*%6I981-ZC^acCbT?1`(wTOQz(7oQfPP6 zskIEuOXeUZr+|Pqz)1ubRahT6G#)8J5)pG<%Puw9!t9-%HeGz-WNP{r0gsTgiW-kq zxoOlja!n`AG_7Uqr|Wa{)OiBy@k<~#8|^t+XuK8@a$xTha&mJcA1N}^zuw>1rz`bh zaF|B_)qA-RBy)rkGwLLmRXg6)$<+)4pHXtXqJ2d_ISZWXZ1&T4B?I>^jV%UfyRCA7= zXc|clSQ6DKOipHI1sGg7Ai!+9CdexAMFd56Kk2w}iCRo&@i-OWFa;$g0(dICp={rV ziGzK@_*UBY`j@wooV|U0L9brDs)l8ZnJ1Us@ib}a>r+2^+uhx*WtbQj$AkdA$jHcQ z_pS8qT7WnP9#TA7lOv6djU3)}Nd1q9j8w=}jhF9B5z@8h7vSfIHmsqzs*0zWQ0XRK z9@^1-I1gR{TFmGBmXIgA7sEcvUlXw(uI`8A3^T3Mt&^jrfFSPK7K}(ME-tP>Nc>LH zA+50ZYoqj#x#7w~)r|n*kePTqn883@1QCp&nw@^>JT9Kk6`wU+T`Qq}eLbYP!`A2d zzL@eQluiIn2l#_OAREGGp{Q*-0^U?aM|UV=*6$HdB5 zy+R6ZyZiGi>W2m~Qex`IwQB*&GDu0$GH!Any@VjwfM3HU=*cI zF92#haa|O1QwWl5VY*py=f5k>&ewAB${>zLQC!($kgkWI+3`k*%nO-)TL->3rb#*G`+ zC#UOm8+UTFm>17SvNbUv&C5P`z|wfSttNWo2K$I^t`08b`nEqO2TnUn>IJDI;YhAU z9WExCIYOCPd_s>23!8tsL9WEm{pk~m&R6tw>2L zTJ;M}J}-0m!LzirwxVMbP_cbwyRUOrf_sH5+49g(CDz==CLG|*{BuI7X6U@IQQfpW zVI1S+S1Bpe?ItRbm5xbD!qL~)kKcoO`!}(pJ$(2OslSA@QaVRy@3^7l$=|=vgp{V; zT^9-8Gn1P2#Rq205Lkm5DkK&BVSyQJX4QbvWn5i_p~cPJQcH*2e0aQC^v%a5A&yC< zBkYEd5LwxSzE+@KBH@DBk@0CW(Nz!nQgXP(9cTTvDR|EBVEcI#02SlwVL#9{MWX!l9#`<*gr|Nd)c2N%i#ftgTK@ zPs{nUx;E$9h^eU~AyUk@<~vY+&}38Hc!GJi)Wa zd+ZeZp8NDB3t*#oJ+|)`h_pbP)bZM(F{iTr-9dUm0e5{vLy6mlDMIJqUlZ0Te8q$d z{fM;{sWk6JBe#W4{K%2Qaw8~3gi?nn=xUKKzj*hK4!R6WXc0xdj|G9saXLHoNbT?2 zf!3UXk5AJoxvHwlXs02aUqqw}Xal}S19bH}U&Mb-H;6;$7__t8e-EJ5qv34g%OoTh zCHvvl=CJo1U0m9O@vp`83HJodmwhY_p{RA|iPMkg)W5JfQ!&^+FhE4Ydb@o%M;mDl z>i51IKN>4#KHgi+%Hz$6;6r=Q5W}Rh7!uDC#>x;C6-7D1`S!B)Gb25qiXyG3sHvU) z96!B(|NbLMw5tLSE+MoJjArh4)r=xw&Ug)VDAk1}dg;}w&gz29_`$*3RkjvI3Yt*` z28DfDi77rzHeoja9swVr0F)nsEpI-*(DLOA0q`@*YHDrps9S)Q(HDF)p)){v6sY(J zvBomxNW`1fxpFt2tc&NHHc$dE$D%vm8n1FN>wPPE4R#&-u9w;~!o`E(0RU8sV(g}` z1Ey)mXP1|;AqU7SC|G=Xa|4>+=2&Jm6pGgOw4{?CD;4A25seTZ-_6A9cO8#+A5&bt zYH_$ZXY}G+fq;O3GLd|*0e+RMc#1VOg4LjLF~vE(+~;VK{P6gAKGDpl1rQva%4L8p z{=iY1L+%0wtG>H}Ex%H>AzuAT*kP|n1J>g?>qVK`5p5ICjgRB^YLfciIV2vDmdMM{p2yfFPGHY3*~ zy+u`@*lmBvCMK07t-fp6K(2Z$wwfLhVCnO^_1GpjFpy4I_{!GS7L-?pLbIkXm_Nu` zF$2pLxBOjB_U$59wY9W#Dm}l#4er#r+Epxuc{iA_K!doz7pA2^fO6ZI?g+rTIPmT! z#@bAi*>4p)AL@7Ed4eMu477E;O=riPfc$`YEIi#Ra-3~O>rIoEo+(XfEGJzqHg7@q z+WyR{5Y{q0ObX;Cw8CwX*F=bDX`=!Cp8{A_zJ2?8xpTqwb@~nucCm#E!9*L(>@$ z5)#t+`|mQiI|9gfj$Xx?cMuR1gau;{hHE07IQN`huk6`;4cnJb+?ujdNq!1KQ>hZ z-AdOq>|#gQf7Nw$dBZ-CKIUQHT2Dy(o>9Gc@j`(TYZVeck?cFz9oEuR4wKdIOYCyf z#~axy2=Ov&STYv@KLac9M#4jDL7PM2mX*_5t4Xa3E^7GL2GtCeG^|0o~ z`3IYkLJlF3uq8=Ab=sT=X=YOIPHb;)&*5Dr2DG5V)sAQ@Z=q|d8X8^D+Re?*`T^OH zTVPi6g!#fH3KW#6wXrhRnI9NzV*ubFo-Ax_BLEuys8cs=B8N+$=;dR~MBtICYis9Q zgDxWRsC89AyK{8Zpkn+jV1TuWD#9yQu2{d|+%(%64^SlS>FsSVxz{z2tIG)lA^(v( z79SdY!3x|5kP8tRSr8Ppoc!0&tRowPi0Iv5iIK(G=@CbtE}(W^pqz2&RTA93jf(w3tEG>WqqI`8pq3+wfU>`@H5IZ+?CYLC9QBZK( zA>ZVbu0w1Z6q};~I{qdmw;C{iyaCjq)e$tUn|&fPq@KP|cpeRX#22=oOxhqB1%i}w zf@%huS?EFaCYAu~R8+4Js@=)EET?Nktr{+b3(BHH%CoPj`njvUT&We7u#MzCX=&5#z!oznB_{CO_7*Wna~J@ZjTkDLlF5rV9WW8SCikx(Btq zdl8h6ZVNLrdwX)|DuCexfQFfOJwJr*Br8UzvUa2v4}nW6$l!^DX;?^CwL%L=;>WS zwtqzn9cmUZ1WNGWkItRlAkMXl4`Jcz#mY9&V-y4Ngz!8FK>=vf#KZ(zE57#5&eqSL zRmKWWt?b6jZvs;`01OXwfVdM?4qZUp3?N{sWVb8wTFCE-DtJVipCd)IK>n2g@L*zP zZJU{CQq~9p5FCa;f1B_5e=hOm~%{u0Tvy}KJ&Rwe+r9|Hmhs)|f3Q($0V z_EWl%PNA>C(E42g9MzGo5bg-{)8WaMctM*!{d4k|rxnJqx6BgBok+q(IWPcg;i`2eeyNjNN9P9{Ga*Fi@BzK=GU_RUxInb&+6%aEnw!+Q;!jUc7nhaE z0X+n~Xy9^x+5&VfN`H>`bfZ84KzN*Thnc@k4*q%8s%1vZ^SFeVBxvQ|{*NS;1Vyc8n{q)7q~ z34mz9R#H;R`1lb)#K7U28$17QxwZd2{ z!#N~CIu<+3NL8>zFnjQLfgst9CPi6Y;ui3ZwfIcQLvevbd za%yo`QoweaU-a@nLjsluc0VvPGZT6wV(6KpI&a_y5?QwF5@a?0mlq%aE8Tvo7Lm?? zLzDykFBG`u_{E$@cYSvJek^wwHUaGr^!oK{3MsGZaS0;XQxO6$5OZKUqd>aVNye1~ zR%n4w*kPJ0Mes2L=no3<9J=z3j>Xlkzf(YPdjeDJgi+ zDIw%rC=@<~YTbGhO?86-$_*XD(Syv!pM1-T#U=UHN2mR#psL>Sb zE_o2tE53W3mky#bd9vX%Ff>vEil#( z3<(K`qU+UTJ6&&hcyxsD6c&#kvxAEGZK;Hv12$Itj^~ktSz{t0jRYGalDv8ICa|QW z1jr+D;0yZ>&Jg0UCL!f)yG{uJoM4U5pITlhZyQbLjosU^jpV?Ve_IbHFEaEg3sO>4 z3Dkc`9wkVjU(84Gk@U8tLA^gQfWjeTWkX(dhj`f`U3n#{kg6VO2t* zEsmVve2{PsWn^b6i|D+86}oqmSyvq=KD47WC@>cH?lHps;z@iTA8Sdwt-k@a{W8=K zOOUfI0m7#H{5ej4w)blu_$H2t+i6mFfz;xIKD6XvKMkMFcP>~>t)JI6o@xzbSjuM@ zL$A>~I9NK0%SN2@Ypg5_vMVVtdxLIli``HxVTtAsw-&JM-2it2t%vM_C$D3bs`fo^d?J;e$cmq-59S(tc_}*|AfS<20o`)^{v1RH!g`+_E)Xu30!x9=;MUX) zAfnR;HUkM}DBMCqOx6bVh0pug<p0so9x1ZJqMO=(|(hE@@Hz)nNgYl4sb03<>d zYzFVh`GN0wmdP0dBw*kgu^xVT^HrsHF_lTh=-Se&(69~gFm$I(K`pyC*BaFE7XNMN z@Z(?Af5ZTbQbC5-pB09|@rsB*BN5U`|7GP{3uXodX&Lv#*jP>hNS@D$&_R{VHmcyz z($sQ*kv_L%s0ANE zc7f;SEEKr?KF4|j>Xyy|pO%)^%h=fT#ZMr|g(@4vR+;_rBV&V? zgNuc(V|H#14QUs|o~qhS!BlHPvk1H}hF@vgCkT&O(9|Z^L4#6OSJ!>bg!ht_7V2aR z(81k5FCiAE>~f?Rf=t1W%T5p1Gk^3b6gX|$$!cPOhhKD}0-r*$SetGzwqlt8aRF%- zps;b^Dg&v+q#_GGg>U<5Q0a}U>@60%<2SdrIZjB@N5;W0gD9)sAvcTArfo7f>qnHy z;rk(I7#HZ6nS*eZarw|ZT0pPaI^4Q<|G@)v6!iYjK|jDiAzU0h64ZrvK#fC|yvB0- zO(?6zTXWdUV5_;KA_KUK&#If$>u50%De<7DAz~eT!yMo_l!#st0_cDQ!E9lxkznd6 zEG%5#ugvOR6jQiq%9u-$$Zj28mA92yzPJy;29di$psMm3Tvh@0#y=kvBr#LM!}eD+Ko3mbb?9P^nvjPCAS zgQ5-}Skj)UZ-uA01RYSW;B5${hxz|ECz|+s68L7r?0iRfH~rDdo+UVnpiZocN9fc1 zb*ZJF*8r&ill5VPS2_CZ&w&{Wy;a2D?|(gKvz9`rW0)RECAr#+q`5a0W>(Sv{!y*I zzFss>O!KM*tIL1F^nb=S7Y91g{n7wmhT6J1IhI#*FgUTWFjMA#&O2q$JdDPIl}Y}) zOPAAe{Q4$ks(A|7k%HG81>?s6PS?Q5sI{vL8>loT(0Ru9K{gDy#e$$Z1cCx$ZJwN@ z0>GnfzXLMDEdYT)o+Dcp#4oUffz*^&*@UQjkotj-3{Fj@2IT`hFG$H=fP}<8eo>7`4TP$>N()1am%X+#Q@mX>z;Jx&JYKJ>+lo6BEPOYVQZeCaB`d`%jFcR*kt z0BnPsgbr%CBM{L@Pb==aCL<%`ce@aS?B9wJ=Mc$)nGszUCvtUYXh`V#Iq1JY!{AqV z2L%S21CNSeH^BQJ^{#`A4;%xiRoaG9(D0iy`A|b_LqDK(z5mlEKNyx`BxMk6Z=C`o zGN?j`3=6^Q_;Y~1LW%>a#e+54WB2~LAazs?4P_KBIG$FoM-h7_>wosWe%d=FBn*W+kv*UXE7KMl-udS{pegD45dW7;b^Zi5i`SWKS zLPGtsLw!-lSrkyZX%2?!>L2sZ^()H$U-5a0(CL2F2<^TUGRF*iJ6)^JW#uv|dv65g z2gE3_65K}RC=^ggiEK1L(9|DHdyY1Gi%UpI056#4uy<4Ss>X{&_4UC9Sj)ZM)U9po&F#9v=RSb_O!#_SG3vxxqs6wfl@D+2?l!rB3uJm!$41u z2F-s+Vj{WE@d~TU&w-ow(9<7F^W>T~hykA!9CVS~5)2l|l7gm6z{<*su)=9R$L!!D zB9t-3fHdt7b)>zcV-A{lfY1C$WkEa*pzx13;OJP!esT!DI#>fC(gryWF@@Z_yKl(| zQ-V6Z?vqNkSFtRq^=u8}e+64|OzbU|=eIi3LUml--Tu}t&`5Ww(HmO-J0ms6dtfju zydJUDHf~+09VrMebg1{=djGx}`VLN~`b0^fLc!2jTv^!(4=!ZC11*^hxPSpkcUyr; z7?Qah$O?#i1LQh{qXt;eRch6v(=81;Bd8M4tF=M)|F%zi`7+`)-e^3zUB{{lVW9pv)oq4S?-L7%E&6mbjG~tuYIH#7eG-+nd6RCUD(M(fQLWG?}xu z+#7yH;GF2~mo6IU>ggYNFJ}Rl1nDcg0HL#?aSH{M+6GD6Y6{GpUn}j5!HQxD%%wrS zJ0BoBY%uB|q^;EiI16SG$piQh#2QS`!$Sx~r;Rnu3mpLPDewbU&!TiH4gQ9c4*El| z^7MLr7sBB6?x&eObox#0(Em0d+U>CTdz3B_d}DWQj0EZj$hMdupT)0&ohdz^*&`VU zbnwdM4Yo`HErU34z)1x3)FtS@7HdHbHyyPh#<7iOTB0)==ZMzL&x)6fZ?v?ze01C*r@5xC???4(vv_NQ2d10K~sc0)C|0x{V%d6pX;Qg)%ej0)r-r2#R6!|{A zb$aT^#?F4eaCB__+du(vb3{-I;Q`I62wD*oPb|o_01|XP`Z zMr)tuec#tOZn^yv!6wX`N4?31t!)_CbVdHn=Vy{A^U@G(skCIX=*Tdq`c|M5pxL)OyrR}BTY^R>y>*LJ4yZb6%vuEc+>#Pl=`Y>MX2o6m_1ifC#_ za-KIo+beb&c*lSBvi?XHXGDBF8Db%Ty~-#dK?M?^b=!7=B6#(|Uh@JZe8f9_2&RTR zd3xO71H=LXXmH#~`Ct??9BD8DK^Ee(Ioj4=cbCOktf%4u=@q<~i~xFgVLDC$RRM!| z6B^(<1x7*zRKC;o9td7Ssw0>kQGWk7y6{t#8E(A8|A{6KLD?Kmz@-PB4+FVm*H~}QmWx)4b zPrxb#$`Fzo<3Azo0XX=b-Ay>IeE)mUL^DRx>yYT;6|QDLASJ*;gVj?2?Y-rH4xK3M zBk(m@Sj#9=Wd-a(n3Iw_Dh^}5Cgaz#6de_ zQNR5!?>G2K%e?5m@~+pjol>`LRLics2N&!;ASEbpCCwbh(rS{Q2@f$gLo(h?NS~hy0FjOm z1161LSK&y2JK(c#-6D?$P>@E!j}k?b&Wc3&ptNuc#|6ki08)^$y}7x0=lVGaM|^JU zCMX$1-EaQ#<#s_mv`vt0)zsvm;dn|S_;*Tgw8E5LH>n2CCd%F2-5jzf5O`fse-LU1 z#xg%*=KIYp^xtSHVYG+f2z!~Bh)5`4k;}sBS1Da#F8&1kl!t4f@lsNf!U4_QP z*49=h0s88^OM6Jr%Uf}BaNucbYI5{|sJTblm-if3nGaF*!DZooddLUTP+Siol_*MP zTB*Ic83l&HQLykn`0^%)3U?J+*ikT0N}a5g$wmu6z4^;6+zvnn7bZW1G5Zg~DQOTl zP>SN|@gY`OkOh)!4?KacM8gG&ba^Z6`)SM2Z#j` zp#bJoM9l}KdG&}38Dr>fh}4>=YY6PAHg6YdRFjb5$!oPhUY%Y_YcRGOTK zi66JJ80;DulwF_@zmmjzb9!Sz;q6-}PCviFAu)4vy7P@g_dlJo@^Q2*bi3UX6&G=E zW6r&IB9EUzKed}{&_siC3A4A?&-d)Bo1naW3fOz}jVu`k%2)>)B{x@#{Frha78aQ` zWVW|mXlaRKDoyGxP>C@)SdQmkg&fV%1S`VHdE=`4iQPNos&cS_;n09(V1P2NL?t8~IjpX* zpC|Sehh{SvH<_N*=;y@Ne}CagfZH5-b8#xMdcTaKXxfRw3`*4S@m-t*6|+3*~YnF)A8Rz55`ikE#Y zynTDcHdsM^5g$Iwe7Z}r`AxsMH@Vplt%GM{1$VLrBO~*%v@k<@X}QyfdiKW$>$yGL zTz(1M$$mtdM?h3?)JvbC6eCcI(HA-+Ime_(87)e3(1~>gS4VnY8v`X=bP)&Tx{+`Y z&B(yP1BaV}5?-P(u`#?)(4Chay?s>7_Un86d5;gByiN?<4XBO=Pa9KOB9w#|4wYo? zM0IE=9(aH_@&z9LU~DY6#f5`;pH!VhUcc@MS+J5OV#rm7Gc0GbnM#CxtKMEH8yn#^ zhV<PBHE$ zXhHEznjR0aP%FsRCL~;_|5I5%wD`SXJIiw%VIeqTJrDev14L>VrBO_HDI+M zCML^mWuXK6FW6O+$G1VVLk?~t=e}st(@PUlxgD$>Pb$3{!E9Du<-o%;yKP0ss^odX z#K6G`G~0RA!~TFOn4!0luMWuQ88`?zY|#g^G-#po!h|ycpSruzQJ|!MfV6%%7edSH zymU)aNqKz#^u|C_Q@w!ItzKX1u8NK!ZJ!9W#PT{bSSTWD{KFFr0s<@t{VamnwcZB? z@;W{`qnJ$m{Ff0EACyHzxJNV$P=mqhd>v`-5$$+}KK2SOS_CE5=^lBpJ@r6+&3^QI zim<<-U$3T)<@%eW{)mijBZGzrnUU^jyw`px834T?%D#N_CIT3^_fU_a;gbi|cbA&- z@mR#w=4fg3aB_^P+W0;`Dk{{qYp}z0&fxgS%S3F_iV7zgl;f{BuTfJ52JgH9+)q%i z5Cr^VX=xseA?;x1;DghZfcT6my>9SY;qls0^4?b-qgIL?B~A_H2>$u#ot<&>*OEsQ zZa?YH-9`#a!(W7EXTvmtyAZ>{!C|uY=|wnE)dDgsFPvTS+WV&ebF_pJj&+#McpDfe#KUu2QCWV`*eIp`hOTCc8M~+$`rc~U)hopDgk-{KD{$ZyvBH2z zi<|<1UO7q1TddN4Dh!NE@8Ri>31U=N2I!A2E&Ka7rzjjRQVhRy1h4IhFBtjwbwSB26qJXvgy#!JqwE{pPTXZFT9K9 z;9$OIee{O5c6J3<;u+>-ZRYC*h&_E5LGbo$h0qh?ywsTQjyVfeo`DPUJI%LyjezOb z0$N?Mh~RQ|W-Rwr?swL@N3Wy=1g5B8Gg=U9dL+c4t&M8dFX`jiO>{q0d?>)hFjxgk zO4=!-AKj~xMk^}pQek&(QwvK!InR)Ummg^#7l)Smw)?x@y_-+0D2*kw_yWPO zTLAk2Zh81CuL#U-SOC+34v3d#p(9h%A>G`~saJQ_ij_EWbo%NI{gTY@@wL#~@{8K= z1nEIumMyV3I4`3E_fBhEp;mDa$a!Pm(K7SeLYnGuI2L>j__#peal4PAFG-ENRxVbgL&FR0OTb33?8yd zTs=@2!HK&|ec^EnM6`b4r;u@OmAVIZJKN~?)0q4Md(^nZUwzX*EaC>{*?c~*xQ}^w zQbvnIyrLhZE`tz&5H+CuFq4vDxoiMoed>G!Qs9~K6|&5xQPIm)rSV(;_DUfq8j>H-TvbvK-E(OMxKDz)4teiREv+%%%u0WF6;Yw)YDy zDr)*MfuyPIn7BOj7{~(=KqPn*S0ar%KL#^v+C7_Rek`E3Jykaa3Ta z=|g}Fe0(x#QsYw&d%w0*Org9mcxM;?75l&>M+9seIXOAf&wuQ}SRh!a@IdRPXUGuN`)HIsEu zFN!`@oj%Q#$kkJqqPzl#W-x1S)&By%pV-s#rVT&T5mT{#;UnJ}d3h|TU7-PB7xOv! zErN&}(1nOT+hqfp=T4K4H*!`FS{+DCf}@RhO?Na+L;T{HJh(sAIRAY-{jss;rM__o zpUl~1Yzhj;W_*Uh2D~&sHsp9cbR%#ko=A2UvM>nlqKKaggoK~s63;$Y+5ts(r>3-# z@iQ%e3I^}-5{ys1T(6h(QRa5s=Tr}nY*GPN0@LPYL~TbTRbY@?M@GoO)-MBE6(s1n zwY7+7sbl3Ci@}Reoc%{fwBF#wU}A2$k^>omC^@w-e+69Sf3|K=u=Yh+@#_ zbpyP}Q6RgvfE&(E91e1Uw^4?WSj(r<4#U>Y|D*aIwMq-A2K5+WV5rOQG+)1n2;`Cj z^*7Azzd-rTFoxm*x+o#Sko#j`w~mgA&in=yLDc5ijWWB}^*c51-#0R{CxwL6KW9OL zq_Dt@aJb-N{g&bwP1mTw^wm_r{wrGlGRcPxqHIhp?M3zCZqIWdw*{P6(oNduc?^QUjynQ15ff?_4GdlxNq<2)s;}9}uQVG;sIi_@1szhm`^r zU$%{dAEx8Yuy=k$x9H>tl;f0Ru7r}_3{)=iXWBVgxKG67AD8U0k@}{8r&pk7^7CXF z@q?wwgEu>cFV6%t63uD?L)x@8THI!CjucMc$kNJnFVjWmhqH|6l4c}Ct1lvhu5jP~ zAT5rA>)$UWKpmu&E1E1qNGTM1pz2UPcOl#KZg;UDnx?A9^McDS&0OmAjC;N$U;+m{ z6!XK(k#}txKkN3*qwS9$G*ypkca|+$(A|I1Z^W6Q9Uk4RvXAwPpxsDy=Ew{aF=nO% zudM)e07n4+HSmOc?)p|^V)uAkbCg?K317*K+kmcyfOg;XC|=FII)9($27@+UyjJ%s z+lj!!HCACgJ>u~y5o(pQn#)ALF#LH{!WW{Q#WK&^qVj0zaFrEKK@r(q1jo=TdKz(oyO?yCvhF9N%8c9VdyPF}Mrb#l6v-$-#iEhm{rYtG za;8?doNb~XX&?*TccN<-Gu7G(s#2md91H6Ry5nCd?oKzTD}9LD78)DdJX+>;d{?v0 znYiibzDRa>^l;;Mdtn@hkL^=$Mse|zRy(jA!7E47z+6z`ynGuttv;Tn(^lX#Z-F>I zntQh`f%w|RRIxEPRwrrcM}zX$zQ$|4-I{VM_H7!2%NpDJ^IMf)Qe$5o-7NZzs`^un?2SdUJzO{A9f1v|EAIf<05hYFZ-u z?zQm>0qPe_wAA9($Qp4?u+Ras{SlnjH8eIejgs)o{u6l)kr^+HhyCCc`N+tg@{$Nv zQfoail^geyAKLeo8XUrqc6m{^FD^`~>}FOmRdoTJy1As%p#BtgflXA6UEHqfcbnzC z%#ifQ+Cwp<>W;jpTOpHiTHSZQ=lgq~%zhD9SK7|nbQJZbY&@U-cbN&y4BzuJ$dn22 zV2Qxaq8tx5Ev7w|slg7Z23B``rEHn~^Pk^(yzW_PVqy0*6b7(mPCkaGGSq&?lb>#- zmTjclQyi!D_TB{M4?wM1z~M-2wUdB!9^XC}ZMw81SwC4RZHt~P+3S)G0 z$-O%{u`7=tzB?;%as1lm-hl;ScRNUJO^Que?_;e_p{>~Eg8@aC%s6ofO@Dv>r`^`A zZSG?w86RidOR9KgW%$_Apj)sfFI+fYokh9fVRfw>r7 zW*m>Bn8muuD%^ayh!CIh(@8Z8ohrLqnH~BQFsq&I75O55H#lDEIp!47c2iF;!T%Gf zjMn8hqI1{&&|Yt8DR$0^AiJ*J=_07gF0Q8Hf445*&}Z>7O`$hjr@z<^ogZWsI9b+# zK`0>uhFSKzwdE@>hZSJ`%irm8{vAxQNyTI%b&7}xGWIV^eXmrstER2RV(WuK+PY|a zl4fm+gphwq92G`01RbiVej5+`W+K#K-3qo`yFV>bBP%B9ndH~2?7v~kv+KGSZs$!n zgj~LIM!G(z@bR46jNoAEp%^7RZprVFHn4XNrYl?=vdA_8dQke$5939BI(^PO-B({* zRoQh7R!8KFQYw0iWTrQaD*k4QlQCg5;4KkzhT<16&sje^SXT#!Dw}>G6=>s2VESzQ zaRb<5a7t>wm&F1AgE(HmY%O@XsLK0K#pJ@SW5-#|*GJxU7#KhzUQ~*xY?(C+BXXCX` zIlQPYg%p_;`{%LD>PpT} z=jI@J6rOTqPWH5Z%hGpk!xusE!rLi;Wp)~K(zw+Er@Uz4=o}sB9=ssOt_O8F{81yH zgFMM*T*(E_^raM=LNq!qt_ZN8Y>nF#A+IK?>a=1uG%Q7%Ya@Ie)h-|_Yrb^r)0h2% zGpI#wr>6oLA6s8Lq(m|Kin;g=W?GSDKmz#6mgyB_NAeO7kI7awtE+69ga<$Pdm3#! z7A(PP11K;FPI8+nOUdjrtqf*f1jS`@1(S&xi-3B&;fyeG)*KSW{yNsw14VY-J3C9C zm-44-li?=a2|@e6nbADvasS&*$W7f_R+mqgUOONjKR7(X!p|QIB&uenDx!&LxVQ+C zap=%-b947!I08kQgq*w!W|vJjH*RF%Ab8XuTL*W%Qi%G+L#VTV61b5KjAy@YayT*5 ze47>x-bB^SS=y3M-Q5|d^*u|02@COiLp$7Y*KJAx^hh|T@emHAgNrj2d@LA!y=VAuGywED|Qd z3F^0nsb6Sy>yIAWSi)&ophv8rkidI!mP!}&Ifpp1*Vb%d4y zOEAji8 zTdP}Z=}VB$pZjFrZlC^nZ=>%KectbT3n{5sFbc5Tt0(hYA+OK?1hfDSsw{Zzg%|Ll zKn;bu6Ry1(F11nvqxGw0ByB-+0-ks)Fn1(UBk%D5DY4f2966SnV^Xut9Ic?B05Yp? zb#ns~WY~eE!!Wb0wMU^*K1OME7W>E34Dyrk5)43$tVsYk1N z$$oo@+#hTiB}`HVG~W2(ks*UTfP)6%phe!@9);sS_zDYhZ-jw9mRp9CDyeXC!A*>c zU>T$z_qow~(UMD4qK@w-`md6cFT03ReSwa}Z2J&i1~Z5J2L2fF)Z1CkaQNm67y?A$ zH8f6hH^2|!n!uskjiigEqhrt+o}zyN{;5{dRio#Nt5{LP_Z6>gm7y5->E7hqtfHwQ zgyg3W(`dwolbMXsOk7V0-e?d~|2p%F*XnFIEgkac6S}>;G6V(v&9II~o>BcM)ZOHV z%NsU&785MHV4KuJP9XjHv$eCc0169v6m-6Der$Q)IXEO_2&~wqURw%au~qW&l7QFF zAg5yG6%_}ivP2;|O5R9$r!;z>T)uX#2R!S@IURVc<(#ayPdrWZiXW4#O)S?fxIs<{ zd>ZttNy6^-?T_j-UE@Ef^i_}E);QDiYjuu~YqkZK*8cXNN6*yZyAOZ>6hq_`F|02f z2nTIvA?P0X3g`s{h8UMq;iLwXh-JuyA3;MwB17Wn$16CuveD`chW=KNQrHbkncOCw zupuly0QMs9JUmVxH8?W=J^Iy!>Fr3yi%Ig2gzeG-uj*a1w|Z_A#%*(h%yqoC|Jvs% zUx?g45MT4$&Qk`=&|~oO4v-E%fTg57}xK`Jt(s&UP@H7xXFH@f1r8Uc{8nb9yX09X)&yz-fK z^JL>r*+&EzxIUO5dkKQkX{%rkxtUiCbdCJbKBT9ovuUPdIyCGh%;AKCbI;;i?z|=x zV=Atl(~G%}SHJ%d>NHpX;*=TZ02SzXx5NPe`bgoECa; zX)2tk1aY+s#MQAu2q^*l|X6j9a?0u9A? zk}>6zp*_i?jjFHB^S+~ubKgqkveq5{I#<3eFd0HV2oiVQ)NS&)37pI?EWW~%P|$RL@Y*OAWanEn34F|Z;`{wv zUzIrx^$~J9p>H}izo02DaW>eA zM;s#Qy-=%xJn%gyi>BQz{VyRAal1BB2KSi;vdkvK9|Ws0r4(;Y!F#y?5ST@vF1$o0 za^f;`OX-w{41J2U$C@+1KJ zg;cBLvjxODZ4gSLemx}K?GxWNI@*Jz=I7UbsUB#>A9YG!v|c-BT6V0ccv?&@WDk+b z-M4|W9r&RCz&d;C)0k*_?rNUe&RERgysWJADG~8r|LX@+>AEO7ZVGl>qLaM}{jaJD zE4Nvd6=^ciH&+`19tQG{0l2dX{j!9tNYG~hTz(S!zaBPTUYstyfMl2PD4s=M??a^fYL?_sRQ-m40o*t(%lH#j}uTEKv+ zp$@oPVH5FUIEc_@a<7mN7s({^VoSF)@-I=~!EfhQXeN!V6|o$cLryE>5+|G5^-bI6 zCpMXzcijs*-AXuAG#$>vH}Wn`hD48`v)Voe61ck{1Vi7I@Sh1Fzc3Q1hSc-*XqoLD zvFREcs!BT|?c((4j*^)fA3OOKmy5AWM^s($T_>vV7NZts-F7q>Wx zEMO)6_x>T?M-8g*12E;s^u@t_0Oxdpae$xJ5X2eaQ4pqlZ{PDXG4lAk>&ftUR1-y4 zwJk(Q!zu3|9FD^O^u%4oN11awUGS0c8sYVWO&pwsO`E@BicjvNSE!wG>h%v_;MD&P z(x)*9e}TS$eoYk7k|!&<(7^v==F-e7s!z>x4>OLR>hwTq`-tp5O=M)CMn05{i(6LS z9JX%Ov===7RgyOG9BLIKS&}>dM(=ncV87)W-mf#QkR`WI(%@>bX@C-BvyW$6TN@w9 zp3zu7{4hL-%7RZa>;&n;JsBCDT5Kn@vnc8LQy%rCw46j9Z1J|3*~;r8b)G+8$UZW5 zPmAsVXhgg<%YWwshao!yctaM_US)j*4IMB$!yxX&+=0dzdc{&j_ql)nAscghX4_tj zWG`JP?vy+EnZ-+4S=kHm1C^xsp73m60)72Q(dwLwo?~NoT#L?znwF@C20Wg4)bi|* zTRzmu*Q$UM+#LEQlOFR?Xl{iF?RC&igJ&1R3@*7waNUG3?>}B$-I+envKRQhr3x`M@VpI}X zk)S03cWnIv8vmzP+TJbM=9<5r^YO*^DQ%j8`T36;`EQ3ys;rpr`c_dHa~h@!K2jE( zK*!GRf4z!JLaeEhHHsME(BUy{&V=CKVY6z~#H|D*bZB4!ZVu{AMTNtBl*VbXoT8!> zNo<)hq0ZMOhpsG!Y`-!?B5n)EP~!HtR}d$&4yn;U8o0D#^Hd;3*Lna+qNw))kM`c5 zpgD+Cj*d~+pN5#f>bamd@O3SmF@svNfh3>v9HbCrcAX%TV>t#H3sWalsWLCmpNv=G z5<_53H9M>Ewt&}Kx$q3JS+ft34denT>lLsa9U)or5M`RK@s*^N_>A9WdIhijc+c~< zjNe|AIh*c;|DF5))_eY#@V09p5t(Y7Eo-C1MY@pO!k>vCsI+_k-yVMkWeBi~E|k_x z&7uE=eS!hj;ya+z0j;OxZLbx^_M0bJfArF}oO*Z*m@ATJ%YXaDY*TORNw`jO&o+NQuh>=(q8XK;CaDFT^H@AQ?dq&v4-AJmO9F%r9@>Z?} z@@|1@oRWv6HyjC>@vQk&#-~v9IkuR1|BLQA9z-Gnpka|lXN5pw1v6I^xWnxRRQB!M zHz<7$>A*I+s#wxF+x3EEh8?yr^s8MG>}bwnQ_+GG16tH zOpy2A?eE&ov(ZOqZY6!m_+4OyvH}hd{hR%p;8&{TwR0vW(yNd)po-YjZszX3hpFFe zv<0w|@krv^JW5?B{{-Kva$4X6a+TuMR?D;t%TjRYLwQ`EQ`^~>b{zQ;6o$!WKz4|R zVt@v0^r4TpXeufY>+a_#@Tv=D%*NjB#Q*i5pJRE(@%q*4a+9Z}=_OE6;(39*wK8~x z8fuZ@p1tW0qNvfk8%b}-#()9Bo@aII1Yc1cpEOBs7@v0;0zE{MOS28*IRASmh26=p zX%Y{51P9H%pXvWzO?Rf<)j&*?PiJ@L<}4H>YVg6; zo#+mK*x1+c%Q$*Q#+|WO--xBVxPlvAsgD$b>_UYKA}M7m#~BPB1DwnAa#*|bb2#(u z#Ipkm&-!RNG=JC_DRZtPT(RQQ(z=1VhrW!!i&(dDG>okLJq-86;na-TP@LZXl=GL9 zJJgGuEUpqdUn=1ehqPSM9c%@!mpNaqbLW>?{)=luYi$*cFLH{LrJ z{Qu;~n|RvD9N60E6#>ck7r0C$vUBkj2jHn|q1C^Q-g7I7*RW)17&mymv|bS3^6$Kz z5Z9P1Uw8R0NyF1gEdQ*?&+qEhJ9JU^sw^1~IXS14WY8D&|9_fGZa%@O2i-~lPY7;| zR}uPyh%yVzs99cXoPZgC7NGY}+!buMtsvWp$}_H(bfExikq31|)82O6GZAF7)e~kX z`z;$@aKcvRv8?V5F6}iYaWeO=z_UfCJJfS2vy_=ltb)aIz`cZ{APy`o?0DUmxFPRsJS<2(l&wbGim{ ze(7hx(5GiBPtj`&5F#Nq;|s^m7VjlhOul~}lhMJSsIKQ)9D?Nm%oGIY^54?`%|}7l z69`?rJ#;<_r5ku{h8pokyF)m67`~@b;|jtRGjbyO`nRPW?_AB-dIkkj&$Hc5{4S4A(TM+} z1)zF}0)40k0B1w4QK9AdtjdWqaMuaaer$qZ=ieG)EE)M}u?I7(cjL#_-`i5UJ;@N7 z3FUIL7jRvv4!Z`jr9JfLP+&M&=tC6F{0={t-YNZ04>`U;NlmN^rp`*C1cKYs!Cqc82yDf-rn^Avf$` zJJfZ(@~lWEQ!Okq*IZ6y5!wA&Ev8}lHkkPOr4QS>q083yJ|P&I>L0Z@hX(#ASaOt@ zZZ8Q}{noOa(2@EuH_y+bbOjez&~Z(xdAu_Gzl`=6rD`*wlR%u$^K)eSH@KIN0Y-}c z{bk}Gr=d-Pn}cX6Z%HQH5G7RQlCHK1I6rmVUr|dSGf|d-HdVcAg1y4w^mZ^Qm-ZQt z$+gv017Gc@A3r)Alr-KGw%-b&co&nB6qA0BQIzWKTPHu0G{IlFcyJ>2OU#Fm!Y68l z%LIZzxQ@YtAlbiciKQU|cc@PpMGOIhlnGyq+lq|C7oQcc?=!7;%&$UU({J znguc$!h=rvsB&%Yx+hAG53x|Kcsu=mNL=QxuS}a3gakDNf9$)QGibt5XZZFX6v^Nz ztx((i1BHl|GgOboeQgJ*G-Kr~>R!1;fwwvFu+Awc$iwOi&Idm~4&Ccg=f73hwTh1Z z)I8zqT3J(Q)&7hTMMF_>Q}V7RNsI!0FdT}DpU?Ii{;e_}n!eEvN`N2KUCqq^jd&ke z$ditpbOjH0Ii)W+Bz@EUZ+$!7`{KkZaY)NBWD1%+yWt|wqsdVtW~HuoR8GE6wxn~k z1b|3MCW(y)AyQVv?=au}&u4Qf{3&Fyq4a&1{W|4*O-zyGt3P;^^%HG%WyHz)+IGsJ7i3JjNqVPB=fxO;P)txZTqi@z%+MXGVxH&o;k~5cPBtVrnGmW&;am`8cSFyP zGXLmVNsT^ue&q%Q9wsKricsh^3*k+Zb?nsiY<^kwgWm z0c?-+s#M7OzDKPvG9DiVC?(`g&yN^Ff2!__rXB=aOqkT5eDVZM+WoFu zCzLKqdmTa0xMkS*y|d@hr~i^}X22h4*Sk+QEj%k1&Iz2+E{~i&aZ|Tk9a<5@-1BM zWkaBoRx2{*H_xB(91@Ja_k)*u1@BUxQ%O+D$AmB|MCjCi&=R#{w@2SEZd?F|= z#tpl+w79*$Kt;0PpB}Uh<|6_=h`X+_km8W@{e~t`WQ8qhy;FG<4I}fzo^k>nDUxO^o@-LmlI{iJ(a19 z(|>{Zq&XmlV~}*|XU6?VAxX9v+RNj9tTHH-*c6P0G7Jdf0W5Q{+6cJLqGmeFD6FqG=rtp10#&*%YTmvu_xj?M`A%K2G@?wkY#eL%ZKyE-cOYDuI7h z_5R*oMkWo$2>8^xch1SnBTxUO-JKHuq_GwO>vGa|jQc8&ceP1kZ71@m{uXaF5W8)}{!q);0)i?b$On6u?ZC<`A(DU3;;S0M z-i~B{YxA5EzQw|eW2fzGL(U{_!#9gO(Ch^0GnU!m)?Y|eDWkhCK$p|xolkoo>SIE} zB=p4jDu8bAlwxJUQOi!AF4uM4;>oD9yj!nvEbLmOYSU}cVwgm{y1M#~Quv)zZ~8#F z66uMM^^Y*M_lnz-R0np+d~YROAy8lTts2u-Nm7u>R6BQaTv4de7JQj^z!X{kY!- zDSVsX6Sys;_4QFElw_>h?~2U+udo*zn_J85U@|SyPBW^?I$!p8SeIg230zL~ljVt5 z{%~P>f&{m2y{v&!)QJTH;9du0+2#0MNp#}4d&}*F{2n#5qX&K2p-J4cb<1T(mG!QR z4VTHhSyYaGk04vO{K~4r!2V2Ed2wsK+#+8Cw&1I|&+6} zR(1$kPx;|X7g6QpM9Bx_d_fg2h;Tk6)~A;0fNZ6Wkl#pA3b-pu%F54yQWcVmi21E` zv?Wd{{L_U+_!b3w2Jy+M>Aynx11v{yqTDocUZNzB;atxUct}EQjL+9tLc-c*{ft?B z*~9Yl*_a2_NWE*U=A(~^f}J3w_y*C#XU$L7VKD(FEDJc>PiyRqL85CdCe)K5OvI&| zyc$%55lx>$Qd0ciu590K35?schDyFkli7(2HthWqbUCMfQ`ne=Cjw&MuWT_dtev7q z%XsaTtVg}>B3f34RjO-E8~>QyfUo6>ozd1xURMTwz9@}`BWA~)AH@`s~hH`zV75S`7OfA$kI+Our$5KP6PjX9-GQCDpnO{n@nhzwUCQh3!3LTr;Nm1p2JXzbaj=Un(e&Qc_-(vE=YU zh0mHXGe7&yLB-iS z{h6WHtb4qn!>ge)Oo+e`DE z;meE8Hrt7#*F$Vk|zry*iRN80Ttf_zN$(N@Yc!2x@nXG-wJ<1@B5kYspOtJhxkx4cI^%HKp zI@bPAdO0^GITXvjSwR6)e(b>T1hoFDARItz78Z7)o`eE5#DILwzrLicb9J2E-S{fk zO@j~ggx8ZBzo$miSy0Jm3?$W?K_O}6F&9@74GSAmo=ca*x=0hRNK0Tc?7BvJy+_49 zTpjPw`Wd$kJ?O$IKo9%!Y*`uz-xAO`76Gynn!8hBu$@($O1Uyd`g#C!LiV{hJ4e7)8e% z7?jCOAr%HbjtqFz(7knlQZEC+Wx1VLIuE#5;44Hk&4w5a`l@YX6Z3kPgXdh7u{6T$ zhGTEk{Yp+=3aCX{@0f|Wsdhf$3Ko*Z!|6IQUc@ad1Zu4|>5ZAT=oR@YTW*SYbSdct zd>4#Kuk}(ER#rz4904A%9n20#Pyq&DUc%}x9;pEX?7cQc9}MXO+*_g7uUrhQ4O^Tw4&dOJlbNq-v}d0wr@?XW zu=63=v`+hZ4v(!?iO29KgPfam3(AxI(mnpy^Aksp<*;(f_@qn@FF4|lJd~u6$-Fx; zVzlIza(K0;c6FBBfUTS~sC^^< z+5k>oYFIdkaxE0YxjyVI=eHuRHY5cH>6u;0Gq}eoo+{7|Unp&N!v2eSc$shQFn>$e zaH$U)!}ZWdCo)`G{(bN7VX7lRkr#pa_cS-G`O-xA8iy>_+&E#!k%>PPM5NEm0&EmH zw=;wv^LZ&ef*wUEEI&Mz&gkQdn<^{}3`1W?NG)VSHIT(Zn4GvT`;#M&>JJqdVO{N< z=#&zileD%v%mvNT`3LL^+fu3{Sl)=a<(Ve1?OTn!yV)KRSN4Vx*(v&7@O&*=Jda(a zPU(5h@2_MtDb(a_r4*sW-gnB*!rmnkcUtsZ!^_dyg~FA<4N~kTFX-Vz#9MgeG3N7Y zy5$%DY!qe6b9ytVZjiQrbK4ZcSC`t%us7?x&fDVD z&;eaJE(~4flPA)3(wSI=h7<;szd+`B4fYTM*pP72d(ZZRd>a5IZ_yodpi1_wg;@J* zjr&|S31?G`SG@T0@^KupT!xxoC0WSuSjh1=$B$3>1LO|cbN}7VR#tK`v{rl~`4TqB zTUQB??!|ABF~$uL5Mf0FIB==ch2B1p^qq-_-c~Xx%r@9s4B_!NFWX zM;_o!Nxl=&H5_Ia;rEiw^UWa&ErH+%fF^X9(OTuB)DAhzDcT|(Snmb}H_w+cKFSm6 z)&7F6JQ+8fA<^{GYBDky7@$pA1e(^_roA8b3@EQ&Li~6;<{97sB+mDTzzs8)p!DgvwWC}0?9c2?HD5S?Z#=}AyhF_cqF&eaZ)5Gz(6yEqHari z5A{1!AqxUIxo=gYkbHOEGP{gvAs`IgoZ!9VkL)KtIT)6@k>B_e$`x>+jaA#yFq|_0 z<72UYy$pK>#-Jtg=nu{dpk)s*MlBePkvi+~8HPgkQ<&obGm=!xO`2+ zL!)ZX!-k=Exi`I=rim-4$(!{B$MgTh&W#JFzH&FTq0 zT7WXsz~6gsyM$Wy#oiHZsA-KW82p^#E0$aoh zr};=0eJU;XET)X+M;e;&LBqfq^|GAi{|+Mx-)JK7qG<7li+di~Ey|Z&!~oRj-+XK9 z)9R8()C2A9`D*&Q4-!$-(3$|Q3)<5MJgsoRzGuOt9KuepRd@5&OwLs}(~CHuxlngO z4hZe!H)Fdnst#mmD$u~@6CY{xbOt?8EA;0j$mzSL=wZ;-cGGRiDEr9B1hwwh!;>hd z$LU>7g@Pr}n0U#~D&V*Blqo|v-!zI^4r?zr2ZUF0FrgytTOum<-8Ej#kedkbD z8mP3=yzP55YZL>?>s5c44`}380$MVW_iE^>EGWe!$aCQyms9&B^r4|5pg{IY;e-2Q zcqEjRT^)s*i^MG#1g>k}HfsXe=#hADc7r$Wep=MM9h=H=%1rja@-vT+p{zNJ#6fBz z@xp!zSH47NGB*SC3O+(D9gVxIyIrnTNCLV%R9D?{Z$>A4(KML4r(iNoS8Wql%3XQf zcvaZ;EjHfdi2oJ|t&Q8=HWeekCLw%Yow1+z+c(FZGO;my{D>O)>f1!~oU6#X zJn5Yx^bt`66Uk;ECqsRwF#VOQKwSQQ4Uc+Q) z`x_l(g62JiKs-gv-RJr#IiHwRLAe0X5xC zqG`{}H1luNLhD!5{q$)0LZ--wc*B_6h({o^G#KA@3lINQ>lkP4lOdecUsz@R*SGYO z0~7OX%Vko%q%F0EC|bu{(qdExCZ*3GEqy}Cnz6`~;i1G5I~edypMmul(6sA_CJ|X< z=Znca*sNfeGIF{N%1+=0fw`1LSU4FyyJ36#31}qYq=sAIq>TkVZx!4v-!z{{x_;kl zJYLEOj+W$8)9I`_E4m7mVd6i2mlK=(bhJF#w&L!{zLg)YD-1-bDk-4T_xX{24`m|a zE$nkrxN7P13mW()=NQM)(>cMgb03hzQkZ~bZjMShK>92 ziQ?BOKj4Q=&`EGa+2krTEH27EoP|l4D2+#>MM;Uzc@6SFRDb&=H2`JRZpV^Ku?giMs4dA2M6EW+*~7h=d0}Q8^q{XYi`~` z8<-)R{F~6(3|pig_=;G#JoLu|8&8puBO!2Kk-@$k5wGqzysxS zWYUn8p0a0ZDQPS;)SW?c0rqEEsOrFFdu67qVGFS@MVIx(^Q(`2^{=*A$xH`!c4L!GCX978dvVxWiAW_;zI1GKqtncLT0Bbqnwh=(jgeN0UXQ^SELGmxe z9LH$43fdU}p}YmyOwZ=&FKZX{fIgccY%6#x?a;XKFf>%ypI>TjUb6lKQ}8_=k50&N zVL~F^3g0(m!H=i}O!Q2jCH>tSUc$d`u(; zNgL?bk|pjh6$8>7T{{ekL|Jdh((m<8F{ zFRcHDP#*aF)7P&!G1 z^$|}NOsfG|KBQjJHEuzYWZtj85brdazr)Ul`idvi{JQ)4nm6=RRj+}Q;K27OF0Irx zQ(w+vdND*JEa*jz%+Ld?y3|FsX94f#dw&LzajPoiXfLXg0qMdJ1TwsqHw9^w|ffqu!+F|3|Yq;l1e6+Uj&UVrU|2 zeDO%GtERp;14O33wI3-w3?|rF$t_^sb?l5$d6S5TEH{GVZ&Ea}>7Sf@Tb3l^5TtIU zHt+pw0eTe+Jh&kM9t?f-NV49JX?_FwBlg_fhlhCOWN{4_MSU5|Nkfa%r8UYAszKcs zInhGT_V}&exk~(v#Rt>Agkxj3AIlh2CqBYcNqPpdtFuLe{{r-5T9BiGJq4{m$^-L1 z+5`m>r#%=ffB+rf&__?Bmux>%8Jp-K`MddEoPF_%30Dgi_}me73G7X7$TLn&+A3$(kyp=(cm-fE-7rC z-1)G!KJ4l$&ah)6G6x3o9z;<}O}emrhlvZIqMHexLjHQH*J7e`#GC7Gzq1!u-VE|U z4Jq^?Pg_BCc`ssUev26Ag7v40rdd0f){VH*FSUi zWs9ChSaRWEu~EuAHO38GebKjrYo65oRkc2vAL2B`24Kkf&H9A~2k$~1YY1>j=!z^LzR;Rou!z9Sj9BsQ#v!x3ti8Q{jcAw! zD@*okX~8^aPh72Z*EHRKUAyU1Q!l zp5v=SgZ?n<#6&!wm1ZC?kdZ?X`Q>C^0E}x=tvdK!5VAOqy+99g0$AR>MGQDgQoXn8 zI%%GH4p&-f#20M`1TSZ&DVdC-&YYsXdn6CLUY!LJx0M+qn4aTX^eFeb+mwvdbEt^; z(q5dJ9@l6TC$CyYgm1T`1l%%NxpcrMN-C@w8yjSPg?~L%6+cbVaP6}4Sfh{$+{uWX zRDQoK>2hzs+J`#fBqEHXEG**7wfCX(MfNvFI%Idg$!m=}GR!}Y*@8grKkxDP$os^{98zOLZR@ z;$z8(owgGE2Ohw9)4t+|9nQCX^$xC7XxTr>l?OjFobo^3?8gaxkluT@su&UX^;JGj zPvoL51}5gymgXxo11iahk%D5;x|}s%h6%b%-#}W19@_z6x6r#3QJv0W??Fflty;4? zVuwzf=g(#uD*9m_ySbYN2|TmWC0)iZ*sau}Y(Vvuj{jUq+ABYczX>O**wRVuQn{63&S=0vL$4Fp-K0EI& zf3lIOz8p^DVb(44`&Q4zsI{*>&H`}<#82Svc}YT=zp(Oyrpb5h*VEKHtedRq}{<@|~g#1J_>CpwLj{p29k9=V4QQ@nWv*#6;TcA4-*RNCr zm2O0r@+djM0z`+CP|Wzwbo{4QJvq1PcruEFB9sW*l>MMiR5F1;kt3&qV+j^ABWI9sD(8grBgn8kI^HU(Gw`4X!S~rHpUUs56jr# zA797+mKi&JHN$c`|A#t0XvG`cdx3xRCMux839cF~0l^oz8&jpl^ztlhvcUwuijpdB z@JxhFeU3Blh$Kf_3k=~44AwfnfGHYn5cjLWAllD*6-?-|-NF<_#}A~34wO4{gK*MT z%vaW?h9yu}OHBDm@9=(9BBhefv(Q7T2;Q~O?3=JpQC3>CfRwgb(!*18Br-SzNf2hi z*I~pXr~%o@*YXRaeDD%a1WCavgWO6EUKjrL>jl3}lkt2k2%xXJar(^cM))?MYPh{I zF$Yl#-4yxx@&31Ng7cx>qBkyMK4nc+W z*WJWhKo#5^e=uBx!YqXhGWhufNeM)pLr_x~dghS`u_mUt+qdR7$s3ooFw_XK11kDe zMJJa{uE&b?1UV8mQxbowi7w9V!IiEV9{#k9cG4wb;LhujAcHEN@i3E|Al2XjC?yi~ z!tD-x)1xWDN2siuq|CaQL^eM+8%RjZ%ieTG7ckf56lq(W^oPID$^f=OjWC>Zi))u0`h%d1wUxu;Y#*;x$4v(-Dby&tAg_jhaP6Cq*#I7oo_m$@ zcf+O~w+;nzIJ@LnhV0kElo}92B5sb3nr1|^D98n!UG!xN(1oMaO?@z>FH4H)e}Lp+ zZTLqII=E!TtT1Qx>d%VLTjl!it>7$x!QC9{ifBt5_$E@JB4O-ASb26nd;YlKtF z`#|QsSO_)&ejGzGUs8cf{^SZ^m_B8j%LfUv2;4*_24vU^)8RWMUY z){L(lGt)Nav#iz4b-&xz6FmPFN6*)oq(;;07GK^-J#T5AE!BI=TMg2z^Kr z^GB~efWvg&rA>vuTe`RJ{bL2qcaY+eW+|t@6EH?ixRWqAIe(RBuUb0Yq{4q90rNnv>a%2VUc2Ed?036UiSn2oZfq&Z&S^>w%nH%jmz1t{5Wap zR{rDy^!QbR{VokrMiw$U zMDSdh3gO61EBU$*2Pco1h3$vSE!|(`^q_lSaX_;Q)RDrFQ5Z!Tlimn@2bj9h8gBKLab>=gNvRy(=anE&i&qH>Zavxs z7b4<**dkp2{z^(>k`D^8zS!dHxVarz)a>uOz?9veME57Z@zL}zZT%TwYA`*G55hTi zupI>vcra=u7j(syp1E;*k3h?V@5yoJgq^swce!^7i!5_zrBj|xFHvEFCF z8C)NIC6^|+U$-fIa?eUO`(-bsfDI)mND3ZmA_pPTUnqdJJLz|MF8AmWz#N{WUjOuq ziTSa@OrTCl>_Yo<`m$r!)6J*3AvUU(Msh(LS5Uq8p)(^~<5guu0|%Ay`5i8C!wX5n zk~0FJEndC)6yuz!O#mKCm;r zgi2~bJBJVrkrg7egr~0lRl;sV9j7+IZ4a4)P*Kl8hdab^sssY?3_Ps zr>=!w*LT86OQIQIkEZ?(=HhXzva>N_OTu?5D$HIBiq5T^7LHPB{s5H;u(lCeT`AA< zH8>eR`4@<`2+U)Tl&35y)37sO5KD1%hwiqJUJr$~lksbj8y2s+yP^4ZyxiK)E*);e zBA_0kqE{(*Pw00#Hv^chyNh2Gpe>+UTj-M?C`3I4W)|k=UwqHKq>dLz0T5!&fynXR z>y=(Nfz~uj*OCFbbs5j+de?k4zy{B5``tG}klbYa8sNSeub!w(_DhBTK?0it$R(N8 zp_RB4xQGPvz3#JTH|ThslBW)c6`O}5%IsRuyP1y7gHEHuIA)Gf#Xj-+4b9%=GgTrfw-vT7gacUV4W=)YfPJ* ziWo&(k{A1`lWL$P)3Bif+UCbnzw==zY6t$rpUuS5EL6X za!}R>1?P=?6e}YpG9~5YCB1-6s@7xG_pe1_O`4#{nwDRrB^5=(ey$pb2jw5kc>kc_ zZnj+Vm*3_0|3A%>TzR=?hB=>V^bjhKxBgx#V@{)O z=k-4=(}Pe!gEm^WuOLmQHW3Z3c9H+}Hg!M$U{rI}Gc zJ^86PNE*Af@4aRjuCyLCI7`>4b3$MV`bvNn<7rdDN`#s-FkHQ?o*s@9dTxx^@bK31>%5iEv47^AVLm*ikl@q?^aI2pfwo zyq-ZoeTtqNN*mqU$EPZvb=I-FhhM}fecNsXfp9*A==fdG>ac=ACm`^?55yU`WLjQ= zJ-gCoY|zCLNdQ5vCVMtrmGgF!(<6sLmcR#~aUF67hLi5j!31KtGwdd`^j0lz?Yz1S zC4%Bp$)B*SVlM3&`pa0~*aUVowTN|05OWM*w}4{#O z%-GGRee4ijfbC8RrYS1H=(Q%VR@9=yR%RH!gEbS zbN0v)R(^ZJ6J_T50d6^E{S;90Lhr)?hC$w>rdEcwG}zb4Vg3%C1uCz{zdHQY_xb`Cd-B;Va>2 zpSlaL+U48b&Hv`!X{1zW+bY9&9PmS;kvi{aMdP6rhyVwML^l1JSB=K^C$evUz6Nj^ z_;G&|-S0XKC6pg2_s^l1UA1=R4H7haGW+UZt<-}NohXo_IE|P_iySK6_ zV#bVf*04dR$eP!_QTm8JK+YTsnR5f*keEIck9YM}RT1)&P&I~T=md>UEkGK9P5doy~zPucxl z;BpNnKuNmK%6{D~irg4umShlK`>)#Yoo0gap7I5w&p$Fbd320~Rxt|#;(v&dLmYH_ zg>}4ai!%8tJ(fd<(G`D#0w_Y&HbKi0- z8zclnaYe<-s)^o)zXgoBcvShSGsP?tvU0A|na>yKnnKD^7`fv00VN6N6braTz`Bc0 z=py!CWm6FVEPlK?2}kZYXv2onHBIEE*FoZC2ROP~23$v#+8SQqZA{`PuwVLlyWghJ zuKR&`cHf^cCx@vlpOXvUDk6MNgD;L@%~cn9t;()W%S&5mw2Sq-(AUb!`q%d#{mlO< z>Doc`{mxz1lrKEGHzwU@n*1cLNAN`a1kTdlsjt+YK>s!8tdH|)dzpuCvK0!n1de}& zr&5YaC-WA&vK%ihvv)3)x4fq7W)`G0zh%CxM!v>SPfU6PAuXS4#3}mAXW~;97W{xn z*?b`aCJ{Zm7;6L4R3}ohq{F>};CWLY#w$cXi)iTGZsEU7gu(Wp*Jt3w*Q?ixH{s{6 zYxbfnJNyESd8g>#jP&#$9G>Zil|ePS2dk-2vno2M>!nB^8MSCgv95((SfFZ4z3JtJ zQ>WLqNfg%A>&eejjT?jr{890mSWX8#UpxdL4rZyiXN?tzi}~q^%5YICCBT^r`R~}I z-jeXnY|O$BBNBaMAo!TB>=J|J__Lv|0^tDWvBF&+5vYtFlB?Vl6)`S_xo4m0#QiIU zLRDqnzFlBK?e)WaR3a>pm)K>L8-ALIr15q?76cnxlPfJAF(~t0^HKw>W<9#&$A2i1 z=gt44XY=(%+*Sf|#uZ9&oh2bYy$`NJa_RF|uZ~gVs(Wd8cw5tj4~vRp+c$Hr%c!zG zEk$;(nuwYpK1opu2VW!e=XGIfX87WY>PA&-WBJ^GV&bk!CSAsVVF%x)G^B}d`tC@JqMF> zSu?E&CZ@T-NyUqA$-LAXlkE@5KR8e)4|85AG3Fl9gspL9vE_^@vfp~-x0o0Lykp(j zVoziuZpE?9*BgY{q58x%I34OOGd$B1*)sc(6BTG53UfUKDRRd4e& z5Z|+XX%A+O%U1gI!Rtd;BBCVn2}8huDCnT>7mMH z%2IDW*1z#8r>L{`vN4#`19$TE&(=?-Y5Qxe?A?HKHCra4R%G{5Ye%HM=rm8)K0g^8eAZ2pModqe z>3z|Q)6ok0!u;F~7R=qfgf!K9o8sBUD=wP^KKnNnB3zdep7kY%~1fSyo`uB@Z-ALRo=7^{y-}v?rOkM*^x~Pj??#PY6fvezY&?+r* zWf9Kx^PBlel@?*)`rv^elk3tuT=5!dn2Yu>6uEy+#_ivwNOi^Y+*R$$2ZdXt%*$O| z+Lh5AJVwT>H%Ax6Zpef&V-0=?*srcQ5z=(F`fGsr z{NJI5f->Z{lYfI97vb!c(La1)hQawnz4nDUR{Z19vUozD-C$gC;_JA=b_+~BK8h;X z%_qtyZb6~Il<(FmUL9JzWFB}i)@)pG3=f!FX4n1#x9+;)?Sxi_T!xJM>mCA;Z|+hD zIL?E!*+0c(P>lK%nC|=imWkZe$fT<1j-^|iSo`bj!?(*#YfycGfdP>e0g~P&jW5o- zyD*aZuK2vMsLZv})Hjl{zrC}NV4$E-PeICzH>mjp7K;9GG5hFFeY4y4q<)B)D8X}Q zaKWzy@EUZj+)bk(y)h`Qc}a{M$go#7jAm7VH?^FV(6^EqlElfywLYB`Z`$wNuf)pD z&!0BbY>xl5I9XUAo=&jz^2^uTO5Y=~r>4YA@#f~3ERp~|l0I%uIylI3?BIk&&Yv7% zZEm6}qd47{iFuv=C02as>)||(q+yC&aq*tie<%NRrYbh;r7_3qG|S**xDr8@z*T5H zs%%fp^tH}KiED6dY@eI-FTWgKUa8v{Vvr)p2He?#&W_zARwhrDU>3oCZ{(V5w#cYegTnT0Kq9yX7$ZmM8vR>D^LOdJ!+JIWPpu%WT-Zyfzk}O8G&9)r#pw z=-HY1n~Y)Z^Rx+nOpR&R5uQ?JM@zp;-p4)8cK}WbC z(zL!Z$<3%NlgB}?Mvb6w!v8Zz+Jxba(X#F+ptK_R4Wnxk+j7GBcjbVuVrJ;U+{QnOr3-zA;0Igt;{5pvR4pK(CkG#@X~K=_JgYH)x;2#pg(4@-w(Tlyvq zfVl(q;R_HH!La!Uzy$%t75ztO_EJBxfAvoOct~woIK%K5C6mG@NVNXfz0~8OkO(CV zWDSeN#=ldz&hAs$O_$ik0qEYhPI_$n7NT2xC9ZPUUwkJ3@Zby_&+<`lB*5_1?oXgN zWD*sndI)VO08YawZ}dnTsC`n>2;0c?Gz_PrS#>aZ{ab%OF(e~&FlrX|{Ev^GdrN;> z{G^eHV!c+|F{VBJqXr-I^V6QHu4G;rjrgR3%cIjCvXhg4=`0L@cEE66xat5-<1?=v_Z8Gpgg6c>DG(TBir&&_wTIqHP>T#>N$ZqP)d#LW=?8 zyRm^ZfyX?JzN5k}`UhA~Z7M(|j;w zfS36TpnN4pb%U>W$4jLzanv$Fsq2Fdg%?h@7dx!)wYF+ti-?YKHB7XaHn)SKx}8PL zpxSlKJn8Rp1pZb1-YV;FQ%!Et=LE4hIA0R1^!$0`MSLhhTnqc(d%Kxin;?ZZ;h&tj zq^iu%82~f<0g5F|1jWPv{C=(H9ve*7$=Ixdj(jUHO9DU;U?{GN4)2DAB3+gzip#Mc zMIsLkl_QeesBf{bX7dy-~n!IGn0-16GRXmYn`A!xsFotNtj-1{bj79iUdc1JiymWRQQ0>jB7X?rjVF z!VigU!Falj8W)sNPDxf(UWgM4GS=Mxr?j_@swdR`h zJ3n9ac5gKO(msi+T3R_Sw#MnzkK(XA%sblV3QZe_S)2n_!XpJfSMX9XTLkc`ob7IRslLmh-zK}|j4;i!g?2hP2e55^lR_`YUH=x;0U0KTwhu?;C=*ZSU_X~q&=2Lbtq zRFDclQaKWArfr-~noMLQL_|bpNAe2`-&Bt7sN1~)};PX~WjnTG5$sO8;*Y$5sZGLId2 z0RmD~FZnT8OkQ2BFu5;0(3ZY71i^#=guYn3*JK1%QUsE?)uns}@T1M!ZNgii@$yIU zQ|lZ(?z9)HdzVJePq)ag+r{hga_g(&_O$(?ya(no)~w^rcC3n)E>kj0vSJNnSE!v77(?6h~#f_w4>BWW9u;%Zn0x75JwPG zP$b#TDzr+B3g>cx3~v#`_B!=v04%!rX+YUa57ORWfSn`M_*(3#EQNrn5y<;|#4pr+ zecR9hRj@q7SZlU9t;-z{P?0y<-t+3nUz+)D$^Guc!=HM%_e$@}6@W6mz<~|39jp~@17QjG!o1C# zS4^{Opa^7-)_;E=fD4-}6u_zhcgRPS0BI1E{Q%UA0q?k74IR=!9ft9xOwGAcJ1-v{nm{`JC@Lf9vs{t1E*S$q+)liFY47&j9QSM}D=ZY2SGcRr|8W`@xfOV)+T2N$FzRC0k!s z1AyZ$MEmrQ+rk}DK0fuHH5ZR(R5>vEwqAa!FLm-1xmVW;^$xzPM$zQy*_UG7p^h7V z3fY%;mjK%h%8^7AGa!z>=oFIiNsW4ip^~g^IAtRTvE2qPi&e0eoE`? z3=;Z&<32jC0NH9!k-y(g|7x=o!k5u?J}FD`G+(b97`nr(&#XtU(JX*I2CjD#f&@4# zogUc=zjg!*vT;Z$RC*h?$m(ip>V)jfIsOU%h;c5p#Xu4rGgRl%-Qc8xeDZ?z2#tv`}f16qe6>{iYDWb3V2c1d9=+BbKkUCF;nur6ZXHhXgxLuVl@)S|d+>AeHjDb)QCs;svc| zVSqi4v{H-U7qj*jQ;h(S=SU7zOz@z^o+VXv@L(v8jS0DCIN~YEDBp=x+!UfB&T(h^ zNnBYelwR#sI~RoScbl|5v;8fZaZ%U~=JlCrnr{WWib=PRUCQG7RHZLXDU}rz2fk3+ zx_G2)`2q3a>TaIFPOaV2Ql~|L3bxe1L?Snw8ih1|_jA@qzi>ev>AD{!u5^Eat%!@M z)DUj_r+kz_wP(8{b(@**VidAi_Y-z-)>XM44j=6mP3;`xo7ckDUdtx}!>d5GyLAt} z`-L4tlt!ruRjR0#qTJdL>g6c5VDj8#H>Q6`MU!OfIFB32z>B5({COSxN-!Y%$t=YX z9ANKl>p6(1Ydp6#AE;mKEloGIJebDfSt2v-mnA&RzD^2jOeDiFz z+=aILHV0lIeshv%4(tw?E3ZAE5x+kABR;s90p1hTG~3FKn^c>@V_dpbw5yXE&)r7m%{X=& z?AfV|pgDnoB<}I)I#ZgiPQgp{E*>5E?Qpj@kd?gk^r+A@M|!kI%>w@l4b~0XT^r?I z-;XiaU`mpk3k+o#AI2AO_<=7T!afakpi)n=uX*a86s?ve<6soLZapj>{J@A z6j@A1nUWF9Y_a95?ry6SFImw!$py zh_zPL!iMZ62$1b|cXun`YafVe;QE{-*Sj+L<-4<`BkPo}e0oK(9M9E$pOdb}Pmy(T zjRSA5?i6l{ilFQow2JxQ{e)n&=4f%}S50h0Kv zE{8U3Fv}hWqiqTZ>>r#cdD`T6yLDWid(fawHxTUak4n%E5TS-3QW11k4MOgE;J0Wy zfB;Lo6alo$Go~r@IO|kN8mR3}%GNs+xeGt*o;Q0SC}NKZwg-GhIqy_I!k)d+?0JKO zH?+^X*D_GpBT7wKCT4e*gn`asN~NzqL8oZt_ft55t*v=~=;En)+~}+}BrUb&R_Q8T={BiDWI@7gO6!MO+y$hYW*-Qn zv#SAJ%z*)C`gt<{YU)+oc!NedQB`br7N6>eF>Yec-xz=I+&mv57*C_QnLn&DRS7aA zndus&($7bk6XUV7aZ3;fwBJ)D#n*igJ=&3;)cE5W4Sqf||Mw;*I9pZZvG6qIDM9t9nPJ3W*xvN93AKtJ8cuDSM{B z0#T_M-B-x|kf)1a_@8NPgP9>1w&rNi<@kD~>3-iTg_HL31xM)eM>~e0yn}<2-YnCs zS&Yu2`*LMHhi~1fRMC{8`;Zy1C4KX#K*AAZ;FwV=Ur^1yG*WgbmHDO<6Z>gjZ4vpj zFx*?~vprOv@Q1cMQ7KYiQ5H(xv$o)Q#XvH)-0W-IlKHKfwQKw2Wd6F*uZla9|&NAe&3em2KbB4RDZ*E3xd^M7P5oT>=N)OLd=qylv zsuXKPI}ZvUz++evw^c>MCR)hQ0YAFfM7Enn**=6<@e+LGH8RlIRgHwn$4O9Q03Cm| zlqMNE=5zBw1avk_PRik*UhmE5Zx@F%3IKLz7`QFZaYrQcVCaV|W3#Ri&<#k0d^MVN z(%FD(fFOVUuT6ofj=898 z-mcM@sv4}|s=Ulm(k@Sp^apVf*~@o-RN-MWI{kifO#&Ieww^sDLHm4e#0csM&(uq= znb=V#oVFjQyKR|tnAVhJhz=c3YHRBdv_E^icz#k~W+)SgFP*~8>g8s6ErDo&4&KFE zyEE)ZZ_)xs^LX@>CG?Yc*rmvIbZs@U;`r00%?xITv-ox2_B}{VCqLd|rbG&TJ6@b1|Nqi1CWR#i>L#b$|#tL%k9 zmX?GSbjLN{=RKx56BwDGCa0m8r_N$aCNLlvj0QfiKLoU1^Bi%FPr!sw2D8!8q8LdI zb<}#={d2P>yv4!rnz-@yBV1Tkjdi_tQhlqkZl8R0uD_45zf~5WGnCp@^~l>d^SMI9 z`j&pP$bG&}oe)u>v+dwQvp%)}C0{KQ|1cg`{OO5&>;zqXqYeoHZjv+6gWU-yl0 zQy8(~PNZ%>Q$Ce}hx{BDPmVM&%kPh+P@uB%`XWB!DGts?wcqVrYwNU-*j0u*0`1bi zzAL4Z==u?Tf7LZ`F)#{Lmhx4QN_vQnj`7=@Jyf=OHWM5<4~CTfj0Ix{H9Tgg@!8+E zMVE~UhDUfz9VHWr8`#(gkAoqVk}S9%K`R$eB%|pYg{H2mF3(0)noz*bu9D*F;Zb6^ zwN~gZpGZ`JL}T*j(!dS)eHX^BZ%{2h)~bgM=XAF*?4mC_V4zqjiKk|NE?JE%0hQgK zwT5yWb`fVR4cmvFH`=gNx134FHG|WrwP&{U)q`IO*v`%&pDUx5q@|TRe2qCqUF$Sb zUJzkRxW{vZ2K;>k+0LII$b=ZmtQ0oulX}JX=fqW{A-~Q2C}yG}lZz9!@&?luOn^Y~ zbF$ogqNyh5R8l@MzaH_4nVTC2QUas4UFhYS%Kr{|%S+uG@$t_YL-APCqDb5ZEfH%Y z*syqQHK2leZL}m_PSG>_sVm6FWXd(7CO4IrkIY&*)D(I2-ukE7pi~(;d+jp zrqc2X&xQ9x23p{+>ORxLDC(g8hzN15liObL6X$Nj^9<1MtSI9)pFf(f$8Bf5&{SQn zg>1cbR6-}3KfP!X(BElRZ$A^NavTB4m}GvFK+bH;Qo|_;(i$@9i;KRB$V*0qG~~5k z-H5;qpE%$7c5_5I;r-<~5vVHQID#Q6YW#JqkYK+m08K6B`!GX}5)rpz^Mfei2W&#n zb7y1HbP@cCPg009udR(34;oeG?^vTnRw5~ufHc(awW7Q@%&?2jKELdCrxzr;owLqr ztR3^MsJQecC0557A*f$@gvs$NegiiaBUunTU3g4Fti})q{23V7y}KCLG_*0u%F2SX zNe@8i>P`;sh9T=m7kaUI-@hkr_YjnfQUq?wro?1kn8uvxBwzA&MZWSpyM zX+=exg+tjfB4LF0zr?W`xo!}{9|h0vWM&5LKe1fbT>R^mV=ndDLM8Ug?hMxFGVJVE zR7Y7q*7ux;iu<19QYGIFo^323QuJ=Q(4u3~beFCDwq|KzDfah56cZHogp@6@72rJD z0LP0YC;c*e1{`q~DBY!40J+a5TH#wqCn^|HXxRDV4uK4fIIkJ)(rm-3^!&(w9U4_1 z7&g}ka|_c3Ra--3YYBeV;8@JUg1^K9}WK2r_ zcoF8{EB(cN@$BFEy>l^%nsL;qR>?n5er_}O}-E&Vn2B{DWO-x z3eWHsq&lF2Z=9*Db_Ii&@(NfD&%OK@wTgZ;?_o?IWb@hj6E#GIP=<$OGfX7X4y+HPIbm{s^8Au$ck&eu5h z(`&t?v^=NlrTU$BF8gxbJUr|k!cM;%@OpQFe}3M=m&E9;Omep^8RZpuC5*vj73_~0 zq80S))4VY8=^$EupwS#9Z^QeN9PPm;Fun(m@)2n~$AxltJ)6g1)(q&-(KhkLCA6Jy zMEGg;L#3V@(zdztGs$fOVA+N6py=U~LOEvRuEnrd0bd z-1=*^oT8NUC4K6xs658RBG@XXLrP5ke29>#KnlH&&8l>F`(E|VCT$kZNo9r@Juju+ z6M>ta8koDlIgpL%cA*Ms?EfVbI+pkQ1~Zx&C7(=8Aj164Mg&U+2@_Klq8O~|E99!1 zz?P~c+1#9m@*a(jwOj!P{0Dz~o1TtN{3J7}ZAw<<=P)Lo)tz#^JVjN8JC*M8Hr!K{ z+#;TDb*==1(XA?jG}LQsmMH~?JNR#Cm;9JdLQgC;)B<1oa;B;C;n%vQ4>Sr&4Gp(` zZuWWxxBGl))+x{%d><-udcjCZS!lKQC@_4Z;(@6n~l#Z7o!`j-Ga-z--jDIJimd4qQQ(V>}`5u;E(9D4#8^oi%)liY%XyZ1^a8|P_W85;!I1bY9-(yV_ z{wgBi)ib_HpZnKt&S(;??UZA*wkR<^O$6_W>x4uNAJnN(+1n6o)%P!I*&K%}Em9s_ z&Gu!vj4`ZAw!)OhA}78^bq=$&x>v$IFyNBT^U|**)7B1@FANfh-WX^El=44BL*VpA zub!WuFUK4%xu|hDeH%Uvdz~4mQ&1y4H*foNX#U;pfMXi`X7ySZ^u^3nE(##VwVym1 zm09~3r*R)$HvLJ5D`)R*>dyHDWn{WYH}k+9*uT4`3pFOQn15~AG0-5=c;{;oy!azs8RQaR1!y+n(hAdL~{}0N3qzTJxFh7CQ+PMNSX9R=!`b<>_(fcpijB zo8~Ji$+2^;OQ6S_7vMspZ)QwyOiBi>!YgzFI@RvcLJfe5g4Gz_v04r~$V(<0?a0P2 zFX(Xl&e^`dmrHQ#w2Up00OE+KZ=Rx1jN7(1-1!15I-x zY1vtg(9r0AU3+!TkwADHg*8^gpZz`Vi&;3`VU6xW*!nk^3j= z9#8wv&RQR6^grY4?ES{%LrUtBc~OrYs8wcgsD_n)ukYeKK62+P+NW$mGgh?DsS|VA z@p2mIaY)auWgvjr@{9g9N25PF#?tbSXsPBlN1_!rxqP~Xg&KkUfPiM#g*O;bax7$K z|KH(ITQ`ij6ZXh!JoRh=SKUyS-k>=4uNClUo1H-l=2a7lw^}%M zyeov%X>X}RfOe$N;^J>DKHX}kUi;DON;Cdr8ip5>Ws94PtrewEf!C1>|pB*R_NJ=J_8fljdW$44|K?^VU%%9)I$MyCl4bsx~ zMn*oSo$jcK9`?lH<+BpO;`)li*n2-@WmIS~h5# zcb{+IFBEa&dp!@W0jJ{>ys9i#-jH@T_NSOkUhdI8|8v#DJ#^jBqQ4D{HVq{{kWA-f zh}zy7UFkm@f#Rq!(x7gmJK)*0P?`oxG_qNA9H7;en}`ku$j^Sr_l>?W^4lcEw?wxh zRF`oKH)FdN5rI10oqQXlA56PxOUMz2fqU?)XDcM6O^I2ZPBmp%y!!gNM~Tkqj0ZGh zdt+i+>Of`eF6CcanYBkT{!f0c&WJS6HhSHBYnQ+`r>80~s*%UaZ?cu^^6C7`yP7Bo z8vO9+@WAzAZFu9&lr@9No; zX({!0r*H%uY*d<>Rm7KWJzm-4<_;X=>hJp)1?g2b2ZZ1Ml5b6o1>&=#E2}G`gWq+tv z;c&~7`qPd3v8cE3lhv1h#T}{dm8VX|k=Xrsv&sWrGtoAb-#0f;Jqw;mNv=?qmW&TZ zkO@iDc-5LttH0){m(LJx7L^^(L3`5Z+o0eRJJ*Dkd7Cq)_F=8`@N2pEe||Z-s6OkR zH5u5TjXI>5ttH5Ak9s!Gi0 zpmKZd7G<%v+|1Ygi)`BvCB@v{R`d6(@$wK z)(d3Uyzg~P-rxB0`nHZ}z(nCEtX*c0AC@96Os95xKcYnl=oZujbadFNo3q2h*eF&9 zA97Ho#hf*IdD<9{9E`e)N=w55BlcAKPMjj_<(oz@k$#n1_Zn^YjH~OaSzBAZ1v1>@ zOYfl7=VH7TecY=-56I$z;rir{KUtN`V9(<1AA_9&8B0uf~svV?O5Canl2~N}*Ob}Rm`>M-q zEC~tOESwZ;3!UFzQ8p%6ha)!TQD7f?>0pHHy4 z(7V6<^f>ftoWEqMrP~Ig99du{DSz5yGL_f}lA%bacm6VE20BEyS)TV}ldKM)_BK&Iv6H3~beLa2hm@k3Or`wf@rW$2*dP@J6_M;fG*BlY=Y%m?ljXC)@apQ>>@5y^s3Fa7MUHy2-1ZACUDnKmUze zAK_b~Z~|BK#d3FI)D6o&_<;H0MKzr`>aM2g;2R_7`%stDx3GWQLNI_J#(T!fiTZVC z-6U{%O-PaQt9Wv5j(n|gWbjww%$BZ(6QtKW(g1O`j_^OfV@C{K_TQHqKn?ESg_J|F z3H2LDj0wnyQQyW=L(6gP-)C$@)MfZKfF?pM>3N;UuoBZA;V+{pZC%lBq@Q;0-n}d4wn2Pxapp{o|HAQ(2NWa#>-;hO4JnMTwlvCh6Gs{C3FZe`kjoKM zQ-@!ipDiD*4(=ZwqWnj0In9MDyrNu9^+ibjo78)-Vb(wzy(q|7f@c*7J%4lZo-W0&l^xP~+S=`*j;pF6mxv=tkMhncB zzynwKB_;n$012X?a7~^ER?$^JyVSaAVbbpH?z`WGt*oXtc-)eo&lVCI+BGpj=H%oA zDl#0v3t3H$C^y_4w0_+69P5dT%fLZgVxrCQi0Vwuhw59?Tar5MJZSY|okJ&fe)mFVp6@5|Ia7<%h_dPx5*VZFXQHIRJt z^l1G6Osl`cq(#{RBUYL^eZjsEcTkAZrOnMW2WG;oDf(B*z3Dh34$)n6y)(chs(!YM zblaXJ06IZ8z!9}ULxuVyc+B=4W=U49%EV1)-U%?L!M{QQueYIY_YJE(7*lx;Bigcug7{qxXO>7_&W9#R_6MSGp00=^JR(0Bdx{K)fX9j<@xB zBO8jC+m0qgfFmI$0#+=}K?h#F;y|F9eTPND*c=-0<{i&&p6{GjvG-b*SP$i!wLSx7 zZ%|kmHVRB({V8a5-Dr>~Yii?evc#mir#UN`E4_na9&GnXfY4hNYCLF5W*{Tx-T(Cq z3$gRHnsx9Pz5)%QqoX6#SdgEeqa4Tb#NB=9Sh_$hW%zIgRv7yAlJti8OQZ`vd?f&i z!hyiL2GFEeeZ5eJrLG$q4|mW<^%QQ zU+j`qw!NAg1}Mn8u=4^vAP$2Y8%52*r5KQmC>NsVwY4`hGgGnSy^OYu9DqW4XwlQq?cgu8U$?yGJ9j1@k;81a0Idw0&=;BxAE=>xj= zs*rGhQK*rT3z9M}Xl@4Wb0{n-H6zj`$|mzyV>acE@}dh5Z^RvbYeg) z@UV9xV(qWMTZ?h6U>3yFs09!m(bm@XVF;VS4WQQn zn}O1HAn`8rGlM@Egn(E7?b(0_x@iG#lm(cd(V->+7(Bx+5CQJ%i*RSKzb8-xVA${@ z861Lwf*t{p9hMC3HqWOhbIb6^$dFgB)EZu>*)VS8@?{TR4_Fl?--R|5o>M8NPHtP!-Zut?<6Mu(4zkC$UaFn9qx zII+v}9xQvX8rh!ik#TBr6(pvMI%ggJ`lSyGQz^s{TVShq4`liSzoZ@xr5Cidx9i8T zD0~L~XaQ__dPpR)VG1;angG{y80hYnYi?;N1x8FUI9i$P)W$c|Nlb#LS1xKD?2l*H zhXq9s-h6#<82AF(V!l6KD!139C&50orY&q!>UC!3=l7v_wQb3;PUe28{&^iTy^o5)N#fj}5d1%dWdF4zrw0D#(_f%4h}x`K${=0& zFK+w4i|q#VTc7RTgU<6z33r}Z*OHPFh&otccmU>pWek2XBM9$Ua!vgJz>Ml;c;|m! zBP9xsnWnAJ2)_9q5#V15{-2*q60~4+UHjApTh$Hcg|<(-yawf+5V2f;JR=OrqoLdz zpnP10y?}Ba8{$jo_Jq(_7uMI&+J{;Is-x%B%mjC-^|5yW0RYDP$9?g?IDP20F^a~> z$OvmB6nvRF&VID=?LEEt>xJ5TVTVFFeGjAv#=zPLkZcBH^OGryfUf?299ZbOfoZK> zffOOCs#TD1fxl0OU?K`Fse-;0p^2 zD0Gc#GG7E>7mUtt!a2>Bn%t0(lw^z*^&-m3$|4Zo{LlE!{ZDOC=RHD!v=$|5Wve;ypHU1OsH9c2_HUuxa}*3 z3e+{bSXgqENa^V#;ThxsO3)Lwd+7WRSOnnuRRRJ)3ts^nTp=c~{t$Cr!w0T&{OOT} z?*SH^2^P1uUIK{e&6B0~Utwj2E|AQ}77P+@mWOxcZHiFSYmAlGwCOr7l>k0M%;N_= z!V=IyX|ozJV5ABfbTDDSLR^P4I_kNgp`jU7TVX*!0kw(~7A~%~M2;YGwFcubX zJ{VdQv+j8BTTBEQ=h`Wfle6ek99aT17hJn_ixQ%)dYe>vc^hlgC2M!gS+v>!F8L}U zp&}SZzp95l4gnf^Q-ZS@$TYsYwjL->6{lV<;Zg;?fctEI@913sOh!;1Kx{!kGg= zdaw!!k-{WRz3m}jP;L~Ino7+tjY~pe7h>!A?<@(SG}gc4bol?v$?N~~YYt1n|LF+# e|Lcp=FERX*?Auc=zu!T?kBWk(e5I`A^Zx~LJ|V~e literal 0 HcmV?d00001 diff --git a/doc/pub/BM/html/figures/MLrbm.pdf b/doc/pub/BM/html/figures/MLrbm.pdf new file mode 100644 index 0000000000000000000000000000000000000000..d6c37aa6e3a91bfc55a6a460787ae07e444caf6b GIT binary patch literal 11007 zcmb_C2{hE-_e5fnHKe2uWnX5WF(S*@w`@sK84SiYjIozgBt;>Fq=ceHSyDvF7Rgf4 zCMrovgceEif1i;eo$u-V&fhuD`^;PJ{_ee>&wVnMdIlk2Pao3lYbglTBQq%f%vZf27K z-IobrXPzx7OeU4#3*q6P2&5&0?qWk_LJk0ro&kVQ4Pin^V;_K1_h(Q3*&9RKAf$~4 z#g*pk4iRRy+i!ICq`EL6A`AF$a?mI+Fw20BuP>b$08oJaWIzhl7yKg405D|G1N|VF zO|lJy)T0K`T&UKD05$k!%b@rM_)!>CUzgBX&S%yGL`ZY0D~+N{4*_stWH zechQJ5E|BwHM}u^$)Hkv5Fwc<_HFcJqCnN|-O)#e*xe&SyxWQdah`dSrxL@oHEpGG z^*Hsk^`|~eRvw=UA}Wu5x?J~T#pIJ$M0!I%^SWP!akB5|WXAThTBGCUz05t&TXYAy zRr0KdJ*swl(25-IX;NOjrmWv< zoI-fg8niFPDX_Zx690-H&sIT<5!ogD^MwV^;|G)ef~Br?$;UNmnw8MiuWHYolF}7eCOL+1m3wQF)+s~t|;E@#kIZSrp4}De%iK|X}v%Cvq#@KPZjUnT4rt`LMbTYxx1)|DzU`v zU85F$TX$Z`b9v3_Yg_jPhQD%e@98;Ux~3yzr9%UWd|oOfz@=qVJk+FW;}TW%slN)( zw<5fu@(S_92mU9e%N5k7)ORq}oJRDImn&RV=CT-$BR1O6ovx11+Srx1v@Zn6LU-;Ox=SwO&h*`0#i>YqMnAPWWv3+R z02ivPOj|IjUh$jiN{Sky?BH<^rz@9Jgf3jH)^l7L&Yvhck?r}$GC@#BP`BcMR$2%7 zJJ(hJ9lW-tuePsi=R5hv=oW(MsKl{Bev9$JaqYtrXPvq)-W6GH9rNwsxy2evuR@zH za1L^=)(^@Jxg8r@F%Hr4R~>ILK0GO?^F~9LbMLxP`4jvO&zw~t+=$p0&iyT&rUqf& zf_LJamklebl^G3KXIod_)>hP?ikYOGG$c{mk8AIDK;tw{mLe z7VHzfaW8bGNp3^Uuc_aX*G**%9CsfqdGlzx@_I!?pnq23hLbxhdv3n?{$t?D_u(%0 zfX5nB?Kf6l_@H;TKcxKol7@{>>*1jF({=s~Uo4jj`(88_qc$7U&~Q}w7l%wF&2z@) zE!US-L%f{rSqd79UQBPWI(K>Zs#VguE?RFb)Nk>LJnOx@!Xxm(M(#OREy#z zho9Ih))@EoXx~Ryxs>rALmMznc9F)hZ1Tv!^x`ub5KWe6`rgZz&g)+XKhA2oTYW_j zDcr2kmaSCxOn;&{J-(t?z5ZM+>V(6m{LDl9&5vfhQ8X35x-_fyO6;4|U3L6KT>g^4 zc+;9?-JKp=5G{Uz)9b4Yp9JGI#E#p(>kvC4W+l=;A+_;z%v+P*of`5|j^QW+QK3GY zd)kLf950S-*ZsnGb!Bmlj;HaP5;={>fngg}eZ^V?_+86wzgk$mA8NLar8N_P&QYvxc+f&jjob5QpA9 zIk?}t-+A---Ms7R%E%|mjmdYHVhqljXQ_oWM#N`)3Qc(vGR5xwF<|{d!DHrTGykP= z3=Tg}@cNe5^%*2-FJ9BGkQ{Z_v-57NIlKJLq0A}7%5 zlRO%yTVZS~;%|G>-)!_;aa;VOJ$YBWOCBV?mx_-vHI|V*7Ga|}U8>-qIPg?o__9&o zo}=ctm&#h?B`Mq6|QDU*pJ1G$!$ykCBhhclzC#PbfQj*TCzCRk5JK}I_TP(%W!J&&Y#I0T6Nl#+DAa>}Uxc?SG zFD=ctIaZh}R^t5T?mI`jG1edN0kHQQyYG5=5Sck`n}I)cLT%i3I4cLZlme*4a@%9i~X2Z({+MgGMv zV{pWI3N^R7?xP2e!}H~iFW3gXLbH}yZ~HD)q4EZ_j=#teIOZga zbGkq`mKE!)#JbkRW*@HAIc~cm*nGKip-2n1ambEXtX=|nhsX%`Hn8b&J`qc3amtA+ zN=c|^NaiDw7H@TW^mdiip}k7U4|$lUAJ$deeR2NHmmne;IrbvIB4iiupy}nBfxK|# zorSt+}xHV4xhgB17mDtX?QjN|$Bc!@U^+}B7 zZFW7-nQ+qJuwI?ObA{~c!m;wo)H=sT-&m8;?A2Pw@`#w0!uE4l4bKt|8Q_)HzJbV5 zVRi#|hpoC?YX^g*-m2S(#`q~ulNT2!2i{Rg=Je5d{&?*GT~tHx8COX^qomx-_w&tR zhoxDM)E$a^mT2>&2J|>_vDMY`JY}nU_OwS-L4M7qii5^A7snjs)?1dfpJP@MBgL1j zAUV}0UF3V@L1(Mo!2jvALH!qzQ~Ao!kJcw{w|dY`MwjLKxhypO)$0@ejMW+zX&=1U zyS!p__2XWoSmA+;5uQMGs(EEi*3;hGx<|P`So<#;kzVDyZMFUt?~QMGO0^V|u+F)X zb@zA@JCa%jmmlN$+`e-1TCDh`_`3eA8{Jwpq?3b_Vxrgfz4(i2;1~CQQ4L|fX0Vv6 zmLU7Uu?y`Nl|}Tsaj~+G8wIiDx~gJEymfcr45U;hbMFBMU7SwxUVDK&H(56UV>J%x zV_a-ar6RBYLJRyA;9qD#6VUU`B*9du7UCt3XX+Y`7;^2*Vpf=ZxOq{^yd?5$kv3Pb z9B09m#lrU%BYjLOmTqw1X>AA@Ao${i`Zxn4?Q&mvBwOz)ucxIjM?DAuN=q*auXf`L zOU@BLC7>?SGjzI-ab~!6!qD&Yj?acg!+bt5z7M^2 zi+nkEx%5Kb>!_~QZP-(s!vzvUQ;Q1%)T92w>A$9;81x^hm*sUqGMXD?089H?cMcKT zn)8}}_^iv!)g|Z(C1;p5CC8}WNxGo0arFjM=a|Zv*S@X<8AW~M(aSR9?TFD`vHi=( zyE{`$$hc!3q==yn<=Q=$qgBZ8@SMc6S z7d|A-YhundeUUM4{p^FL1pV}ZWadcW#OqJZZ7P)(nuql&swNJ4`JYio==m9VjvZuD z>Wdr6xt*MNO%AUsY=u2x6GRU~rg-Fg;n-KWJPGpj!)vz(Ibsf4fBK7f{xv05oA1%| z<3Og4;?@g4vmE;rDXRi$mE&d#)|N3>OvTs9?Je0&MaQ#{Whaq={ZjLYX;Ww zC*J2=_T3S^^64fz5r-fLc z&*xmtPLJsryq16NTA*9X16uQ)`rRKrHe-)!*5KdCjhGL7J*iPzFyL_N($eHz+E>(0 z7+a!uU2@)&lNq#II&;N~4dRCB3TyqHeG%`+#o63$Y`)g;YH>F^G4wB7{&QxBouApM zSd$K~L-RIIAtIdE2FFITB+WV|rM-K^2asMV8+P)_Kw^6GtL;C8BJ1SE&PkD$Ouu zdFs9)T5EHk>7erF+pi`dkT|W^_1qkJ8Wr3Ss(KjXL)G)5xbMqtsQ=Rc_(&e-wt)s} zuG*rBbj2DDlZ?CGr=cH>Z>7#+1kMz`HIch&B@+Cu$+4uo@`N$QVO;-U%ktL2n)J`X zX_)gZ>qAq8^DOkgiiekRE<5q+-4n0$XoIVyTS_uJ2|;0uv8s$M{7nT$Mj{U?5EeO& z%S|7re$6GkK_6dW@`Rw-a2oKdjcGpC7L1hOxGyd+miL2{k2W~@;$ql##EpeJE- z(rtGcS#g1G*&_4O>Q~QFN~6-#bQkyAM|LtxD!xgW*}u9RHL6Gv?c8Ucg)fM5D?RMu zS{pK6V-_2|zgEcd_^uS5bH|Q6c*e8#jG=kZP))G|;e(W3&IDFD`POaub?h+~F6$8Y zFGmXQAyr*ISk=FjBwBrddLl!a*T+NVdf4IU>h{c%JW9i(tVbVzEYWue&-sfM{x!!# zWB&+^N1B&$O7W6E*uE4R6W>q})0y@7$E873hK}x69qIL>DDDd%#|Z89Blk|%5`Pdn z&%2~3uc-*Lw&Q=B5K}oOVo5_;)kSYHTTQ04rrl{w#avzWvdnzn1@9Ewi~E8!xcJM? zQqOZo6!=iyDUb5jY}$V9_A;#m(~H4Xn{MpZwLu;G3orj`Hi)+-6?SX$4(>v1gCc`k zGT$sNh!}ppKY0g&TxxePE-guZP2$`lJjvh_AEwl6o@39_Li8n4@T1VCfYlb;(ii;9rZ(c5(LGBGk}&r>5_Qiyp+nNcr>ljG?`Ar0Q~$=lfvKE-lklqf(qWTt zMoM00PlI!eX?LZM79P{56Y0A&uW5f&y|a6J3#4Q7P@cB4{oDx``3!Mtbg)DQFYdNP z%w1uP7_3n8r(N<9J4|IOwxGnma zi&x$!KFFcvf*^+ZhnO`y8447~bA~_i+URD9Q4;fSs#|Om__#5>cH^12r9+Nw%NolC zA5vH39#3hAi&|?gWo?xGPPPE6lfapC-YO?vqCVnn^V$=XO=~@QCml_@9t8-uuEebu zJ+DE^Gu)in+*H3*X~|S&OSQi3%JUimLVh>9$Wxi``-naF%S-A%ecc{VT!|8~AML-g zD#859S0b|atLCf0iSHcwc@Cm~5#T?kuz$SiXdcnw<3NL#vCwEO0fiu4`<{|b8@r}= zm}YcuZ|9Cfi|Uf;`JQ~o_b$7AD9<$i;fTV)SQEa+bSJ*N-21ED6T_Um|UDeuK z^!A(J^(4Dp?fnK7G9G&k-`mihrTPokao`ufQ+Vf?Qz=k>vS9OLTvB&wagvwXnden+ z3I?YiZ0gyuWWDtn=YRvaIj zll1m2k?#xjNT&A>=bfZi-X7=IX0s_>C>r$KLu?w^T)kaW;^q?@{ zqG$kw{#oPH_XX8IxaO&$fk5if+}x;eMH18!9YE0$Li$k|G`cGU&KEL_%Ai9?x-Tdd zGK0Yr(}O_;&u*ah2>#Q8z;ggC1j2(|RGPa76C#ixq&`H%B9O)qi3lMn5E_LBlc2*w zNO#cTAtVhH98nO`7eEG2I0y-V0YaG&8hlwphCVF`EZCKm-)(|Dz%PS2$R- z8iWJ-1C??}4R|vGmVhAPQ4kTllwpW?2n#+p!3_`Sgy-S)Sn%S`>aj$4{eLyY>^v5% zg&PcGw&6jRgGRvuP{V-+0Fl%nA`$!sZw7dPoCr5O2Evf=2+$w`I3#%RZlnh6lnBli zQ4J6P4h5@+=dqxHsu2c z8H)pgP-8Vbs42nEBqE>}j1$lpz~BsC9KeeQhY0>&4bS00uLgt#_pFA2*AZAkfO{A} zOKGqG0S!Q)2pA_G>|qIImN>X)SsIK92me69B7(_=S%Z7fSo(m`%`^o5r_O$Xnt1>O z&VaxGj=?rOOckIU-VU3=th#1J;Vz!5c0`!hx_3-nY#?tl7JiO*tgY5#= zD%@}T<>cljANs%_!#)={AqWVO^#tMp>z)i85?DnjFgxS?fOc6cSW$)bK^|P@VFenX zZq|%Dgkkw`IDi1vv&MWO3@fO?p#<|28yv1MEPROBSO)woYdHgAea8Sufm{8F5Ds8K zuB>@*spu!NnMeot1>*=-@B_TE)`SAvnR#MGN8lIW;olB&)<42mfDC|<&20NUKQ9iT zXKo7V02s3|@7G-?qzjN13>c|{nOz3rPOOuMPBvSbv&jPC4FHJ67E-+#jWV? zESA50%1Y&7YoJVmCkSnTR@8E8t4LUc+K78=jsMIQ(U}24l=V}VUCrA zFsN<_I6@*o9Q*kP!W14Dx*POs3{D#1Kgf4>j6{MD$sc1l!oqbp;QIbthX+~VpJN2r zJNyBUfCWc({ul~H0EYd?7#4)dKgTer1@J&-2$VQ~FNpx$`5$9yX#4`+h@`)a;bCv{ z2RsaFfeg@Ua3VN=9hyV{K4AVBAbA0A7#v{%&A?-WTy_3lEL^1ja}2*wR#@UfnsI2@ zKIg;3!C~RgF+A)E{#ZvOEs#BqgkK<2Jc_trFCL3pI7UP*kSPJR&|W~qTYv+oasfOd zYN4GHF}MY?Cjo>0%^Q(5;1(pf2MF#iXZil! zt*za9Td(R>y;E~)dV0F2XL@?3dU_^WMM(w=oeUiW0%6I?N~wWBaO1DPE-Jz+M&a&e z76d|Lu#uEhu`x3Pfn=joQmj;xb%+N~FNOBQ#qcE}RU{KX&5;@wYKi%;doM#H)DwFjIp_?z2{eM2<1a6SE^!54D1@O{X>MA=2dM=PBsDWho`jVf*Q>J6RDF9xXq^)n#liLzM6 zvb0*t!DWO&+}=%5;4%>VBNqn`qLJ14z@b9I+fP8?q?NQ$=U*Stf-8vTvCC&O1>|-D zFZS^Jvof@``kNm};N*-OF$3`Vo`Qvi8U+7O44DyLGY`!z7>@%XwVQ|!M7#m#Cyu)S zx}!m+gNGY}Z#70?1W6lH^P!fDv08&5d4yTW*&9MF@KvG98*C)V3BkCEA!%ZWYr#mU zAvL(91>#&^@r$tc#XjMZ*F{j!@D7oqhe+gGGopEezAt(5R>GXJ4pPm9)%lUX zlmB^=U`7O+n2d2uq|+!;SaO$;xRT67w8oP5D;n?CXjforq1c6AcSo4a)FLg&AITlj zAK`l7KT^r`Gi}B?#kUyij<*{y=*QYDal-I&?hsgmKMhXkcI-{tAU|8Z6TFjq zf)@|zrq#ytLvs$J3`3V--Xt@LJ&TZ~E5{v=Htg5l@-Rg@e{+uPoOKs0MQ@YbQiQ99 zY{S0Kx6ifzb{{K4@ej>%Qt0sh-_E}b7n&F97rvaC={Y3IEE4uG?MCg$g|7?D=s5*S zn<`)x7sj1$+}{Mh={R?^MDTqUQuO=b)p*B$u6oXomKl~G7AaLARV!8B4~e$AQyP2IQrWOLUZyR=>Tk_D*a_p{&3Dk&+%7q zx1e{9_-Rg;KWG-V(Y@MJi9^gu3zg&~gk8B>{WPU3Goa=r;92B(5}5%R4f$19SY;7x z5#AB25rO13hvka6^IAyD&3FkBR64sQ|3f^ zSw?u~TzXvHzP+*s=VJu_(#nJ%?B4qlhp2>TpQx1erEaz{w%NWZ%dk(EptZzgJS^pl zX1Zo}%2cKerLF(0u63b3fu*|R<(SqSxVP!I$Ctui(c4*nJLUT=k(V zejPGT=Xb?nL19f&`li-ft#r0(erk?&rY;^A#$S}=#lLXO(Ow&u%WizIRt%l~Q|ewk zlP3D>1#>*sfOI?KPsUuvY5i0Egt>zGdy8Ag>9q!{8I5?e31yfn>>lRplH%(5H{Nm6 zzHAvCinYE}4{_IfX&OFOT2?-yIf`08*UJ$%v{_ruh7)YRB)n+eQ$BK8np30EqG2Ri zg|@cepv>abieet=be6hI?)66X>PaI@??}h|dfL(xC4136-=1Fk-ibh8LN9Y;azlSi zzUWvFNR>(1DDC&d-6Bw0jMeqlHKXuSb?7+g8~R2$F>ZgY>8&jIp;qZ)v}3BNIOo)8 z;I0?vH=dJu--j0&8NWbfQ>po#wfxzO=jLCiKxtdNPu#9YY-&dG+xMq0_JHf>9}ac~^ee zg~YIQ@2{sAg)|qoaRa4RAz%835vj?6oWais)oZ8M2b3+lKF0&@gNL6V`TWxyv{(58 zXSQ}*^^Wwin<0*#SFo#Q!16-xF=$_Gkg{ITz;pc61E-3#gU_7XZ71is^YGUD))A+g zl8WM-+?#`p>w_Qex$%QZZ{K_Z7d5NUnn(l{vmfnC(No#e%)@$PfFnUBI(zJL!ma3~ z_l;MG==nuLgDBDa%iQGLG-W2pM+iA2H`h6<_0e=SW;l(Bo!AoN<=b8TZp!YD`oP7P zId=3nkS?MQ!-w{}xG~taW8C`9nD>ORf7+dcN!MBD*dqb;0`-FkvR{_&kZwPlmf!Tl z?DOLF<$~JQ^t7l1%h#@CSBS=|<3wX7s|Epqd>BBWpfC{V;T0A14+L`O0D%sSK_G#2 z5Qx|*t4&qt)z#`%QPhwI{_9&^UG?ykXij~#K$jHjd>htH% zp0CN#(bLMx)5yq^gToUQ)sv6UQ(4)QhQ<>F@|2PB#K7=`gY!f}^2EjUBqa1?V)CS? z_r$^RL_sMiDB$DclaZ0>?(XjL)Bg54PjGH(O6u(tnsuiCDVQqANP(h|C}m#@Xil;^ zuCMujU|GF_18?Niifm?9Mu*;c0y7zb=V%`ORL@R( z4L&c_lGF(=gZBz;n9%92`-@}5gXmC;$|X^NINON*K!aZ5{_*m9Txnm;0yy`(J)I=j zd*^U$L)`beG7kgo|(44W)5(C`m2T#dW)4H4~rdD^+v5F ztTS}_XJ8XFc+C^ma#38#iXl9y2mR za9f4F7ga)VSi{7Lr#z!a!J~Ra(&FvnkIP+KMY=(E6Z86XtRIdpTrU3nw1|ZN8q!`; zc{d1?G%pI52H0lF-?@Z(W}YdX%drP-2W|((^1`o!f%aVwvWS|Msy(7lP2s&Z84o5D z=yrQ+?@`5wWjzPC<9kH19}M$CJiV4WOIL6RqJ~4|5-?_!sD542kQcJ%A*>9Jg%;xE zaRm)^|EMJ)-*+A$MYBH-PKJuZZL(l?GVbvf|of3eGxWK(aWySWVTDSWl!C$|`n~pUur@OVB*0 z&zg~RopNAt)5Tl1$Y_)yYOCH&-t(63&5$mI8@9nVviqeH3VI_6r_*AVb4I>E&}9%A zyar0`msOSRA!j3Rf?|1Ofmr4&Bj-i1!{yUIJ#EIn{TuBle!lV6OWN|yBB6{o(GSMiyv{c%;?*Y7^Q^`{tgSPU zB;WXCbY!`vw8ES2@nVR6>q}Jh;8e0D|Cb|aHa%MYgGHiVBfn5D=+lRope?UOXxca@ zFEVe=QT>P+4ULN5S!Fxlx%2H7;+>do zUh_xn*NN9l9Rmr=GwTc^U$=6QfGAhLdhaYKjrJ3Do0lIl+_);S?z|Nr zxT~T4`FuPdRYQI{B-=$vJ3dJDb?pBJD>(AdK$RS4_UyDoUM>?WPDH##dgMNi!sB#Fh8;JQC; z7bu)xu}c3HTu{^&Oo<<9454dRAM%$-__ULbJ7bZp6wzUAPE)u9$6aB;$f2%;?wBwO?R-Do>(qA{|ErDagOkDf?ovK_6JwH_31Y3;bLmHV% zDS|HfUpy0o(4e@rT6wd@$&!&>k`eK2+u+YgHYpaNY~HxB`5O2iat07X)F;!sB73+P zkT>6Y>kOJM>c*0o5i~+$W5x%9`k#jQynNkCoL7)%d8+PO9x)lIk=PO7^5%Nc*eX!T zC-|O5VMuXX+Vll;Er0Vi66%rMP=UtE}0wt|IyHFpq&`$k2`L#$m8MZQ;(= zZr7D+TA*708gt-MM1hlu#OHyOzj+oSsF)03@*s+9+>bYOV-DU-y`C`VtmP{uF?T8z zU$VR-^HfWF$hIY* zn!|i5i=WCCB`jYe;$+c)xPbz@jWvWw+1hbZ(GB%W1NzkjhYl`J_@o$3t{6gIH!5Bu@a2@l|J}`@i^Rd~_rKFO=`k;uKj*;x$p3Z+$mL-dA}f{~ zg;17=;@c$zg*0j$D~tBy`LiQQyHT_}XFgVyfNwk@B}rObMgr)nWRa=H8DiI85u=JI z*$^6@rz#V=_(`q*=|5+a%LHPf-B}OrdB}(^j9ROd5XAhg33qg;)po(&VodF`LXEuh zZ(c5;q6k_Z8-_Ji@<--lL7iJzDHmR7c%%8q0>q@AbDX38kjP-O#0zgZmC(cx+oNM~ zdcQZY`o(a=^K(VQk8PR>Wj1+F7!SQy$|P# z&ij@B$zk0&p)c_pRO*5FI1zD3q4G;32t;TtCnc`owRD^vox42X9>k3Z_mxJ67%Q5b zh6;2%(w=0hP#0gN)lA6W$f^_BtYu#s4+A&q@!I%ME2#@=!LCGJ)+aD!PpTS5>78Bg zL>~9TZ4ea3O5rTm#yal*G~))-MDgW862)M&!C+~4=oiGLH)st^_}1!lQzf$hKUMV8 zg7uQ<8t+wMRE-BOBSW<;rs`T6np-lqGMeq~ffj$YiR#Lf5;R&*PGiQ*YTL@(eV$}B zzr6%ee?M^iF1z1x0SjNYzc!cKmJx0(y?vi+%cUmp^ZK|K%Jpr($LV2VF7-iJmLb)z zkyV2iZC}mvB8W0>$};?$vAUXoZmzIy^dHr4HZUTnmKYj1nx$Te;Ro#EW$w2n(l?V! z*(pn;|2}TjT1a2RQJbTV@rTC!mKvSzpnlV7z|b@x_Dn{1y6TZ#2PuGCytCL zVm!wZdEm>~WV3}nk<|S!`2T5oEzEhcE%x6-%XpvP{Fi!e>#r2hpY1M~mMVg{>?i83 z?1Z^yO9$Q8_XSksSF_CbPi+xsm01hM(HXZp`NyXx6-&VTu>bJNL3MMAe-<~^IB3b2 z={#J&O%zC0{*R6`i+@(Sb0`F+yh6DRe|xSZ8+70r`$Goc#G;ZpVoI@Lc>Rf+oAyat zWqp=M2x5roBDGEjHNy9sZuj_h6!mPc6iYdxO|Rk;XBSH#re})wh=i#Br__;J{^9BMRJyW{_o%O0X97p+ zW-b{VfB#^~Q~8z4zUf=y!}PC&TK=17+(^|g1(>jMI*9!T`M}J)8*7+-bYo0UX8~i+ zZjB@Z*w;n*^m4tRTxQ8FNP5j*-_!t?enG=cGZ$*ho#hbkQZ(+SdO~u*hNQ~)u3FH4 z>HwOGmeph>;|gUkuBZRgV%$X|e~|xONa5!nRi&Y6>Xw%NHnEY~#e)l$XI5|EKxSDH z9@Qz@-DH*$!)1WMLj}Fi0fs5~rN)Cu0>Qpc=}V+*=a;+y7@qbL$40(Z}$x`J2mSy4Ku(w)5b7Ratk5`4M_ zaM!mgU$+>CO8PTASzk>OY4PY^lDwXH=k}x&j8G+ZJXK`l+STnT!@3^+OD$6=*f(RT zLDuBWP0_I@Pb)iRAj9h#F}26;5?;wg@#pyumCtmJ*MG1fISnJxY2vS~e1*il9T#-1 zf(#;dw;J9v4V8?gKW=5&&0NI}AyRCODTssh{8JkC3~h5DD3~j)-(R&*mf*%k@ye7l z?)JGB^?)rlr8Ex=jr=a9wU^y0w%X5h>dnMhSBj}YYf%NGyNc~cv`5}4&U;^3&P>UA z=}WNOfTTKmXQ`L8tGp~&_LCHEG{IR5D)QwZKZ~cs+vTeJx);~UM}-aj%ghH`Ll_ZV zg-(q6?S+aq?1oOMjx@3&ZEJg!8z(M$CPG9hX#&E%)eyFWRO+vUL7@A$fYezx0?i@F zV8V~b8rjk`vqgo4I($K+UCK7m!*?x%uS*9me8C)z*_?LA$I+!=A>n*HBm2`L#SGP# z1cQkqo)OD#57}tvbZ+in3$quuDyXAKDJrni?m;Cd)OKpNztoQWpj_|mETM~Na${m*Qwz($ z2o$?xVv^7$duADgQW_RHNhm%-2nb~$wYUr%4b9L3Q8D-#MGP#NyB3cRUDO1yiHAbl z?ER~P-{#X>*Fa>GCHzY-xO5E2?75Bg)H)*`0oGRvUE|JxmtN`Rex&)BX=9k5*#xh) zEe0kni)=C!-{l4$`7Sko@pad&Z;S5yhGf$O*+TyGrq6<6Uq3ct+08oF6~k|`kKZU} z$_W-`fu_9B^+#KfIPV_p-8>V_JeM+jFzpA+`6Ve4gtMG%(>2cF1gh$FG{fCA&vcx* z38~)?u{-j}k-CK953aOEh=+x1!@=x(+(X=5@w9pw&O@CWMu|#|MJ?EVn9P{Sbydl1 zgMRGkYxObE-`sD#v7etXv#@_)i2}iuyvrFJS%@|3!o)q?71A28L@W{ZFz^XtDoL=g zu*dI1ID_KXU7hl&|FhN_WOy5Gc1P}n)ZK5BAyY38jz-&@z*-~eEe?{YlLtf7F-whS zZb-C-Ed}%*7+|0TQnq;WM*BK`qh_#KOWb09ym^X!Qg`F#N;W02=}dS`i{73>#kX-f z5RU1fnMr;-uY~^kJuS0E^4HpwDU>P%dj$R{4z4SBX3OM6hlb#Gcd8!rCAz9Z(0b+N zcs+?UU;l@tqUy;YXoEBPV6~AkAr#-r{$azRhmAz9l+*D}gxMxJaV7M$#QE(?i2Elr zu%4uxJa4|O!b~APS8!=s|gUJ0}cq2Q77hKCJ^}3I1 zUh}|wl~4Acha-26@-`*-yjL8$GpvupQ!k4s|IHNU3-B}Ah`t$(T+e6ogww{>@ErF~ z{asQFCd`@7riznq%5Yd3PB!R3l5=`zc&LMB(O~mYQ{h&BNPC?I9Gh{6wqoamd~1Qu&uI1XjAVl+=-Ql8W;Co$ z4*#-8{q#4ImY}lH-ygyRGunkjRoY;t`0JCKztTeu&9vfsSs$nbJL~541!#A7>uYW{f zhMziV`yxkX;>=5~*xQ2;y-`c#jYD`&1$bpb?gz~weXMp>^G4Z%^#NV z#+p~pmb7~iXMEPu$3=U+E7|mV55K>($uro>KOa0m2gDA4=JGb7Hq3e}Ex!&?P zVbOPOZ+aJSXV|c`ns!|?T>DOt3sP=(1h=x9D}623&DP7 zV&oXol6)im-hS;ouDkN_(844Rs)oXTZ~J*bThTiHrmJ>g$xJb(VZ1{Mi`8PnF*cCNMqk5*R}fn5G347C2Nte{>ORawg2zX(9Ia7g2iR zwcDU$l}f{~6U#$41P3}JRF8&|mJQ)Bevh@bemOlZhLbh>W@=YeKiJx6JHdOGpsD@& z*A+tKIBESPlHosz`qY?$@pmQ)e%?QkUaXf0R#yZ?D{D%tbMj4dSIm{SJ~hTq?IzL* z<+Se~LqQ3@QE(hLZsw zb^M6EDE7x>QmGS?Ew7BHcIEDHBQ`GIs-aG-U3I*lAwIBZT7TZFfnJ1VC3r~z*0eJf zMc9So1e{w7H1WjmJy6kVEA;{uSN;1DbADp*ZmKWA2M}(&J!+{zb2LOb-R6BtkEVoe zH?W%pA5Ic|OC{BpfMFyd&csbYCL;!Yl?zD%zG5d%*na5FwXe_Y@N^`m5_!RE*}uyi zcNwDJ-$I*V=ElKZ(oZ+PpmEnf<89tI;}w{S%W!7*FI{^L7Y!pXR4EL=b@aU;tbg~$ zEFowW*D=C_S{mDzNapQ(2v)-h`N|y-ZXlFYL_VAo3*9_!n3^M>Ng89AQAOPJ{D^~& z*6rg2j0-87He{fHO-wMr;7vT(0{> z0dL-tdW4?{xuQ7~U88-tHV1D2HmEFRcciC7X~N#+p7(bzn3gs>r*;B(c{4m=W!khf!QtG+S+Y1T(aM8)+Ahje4xH5!9{h4 z&!&K#6949(-1FUw85wG1KNK}VOM>(}sgI}aGtCu}P1T{G$z3TP@S%;!y(nd(o}r*jJ;X)d{~-{sf#L=;8P0@j7qV$nO-`{9iR>HOlZ%ZgQOx~I6u&JU8gn=wVUX<=ytYtVA`pj*IMc6 z_$^JtdgSrX#SrdWn`Cs6j4s2;jI(BX%l|qcyT&BZ*{HpJGT=14WeY ztM?{~Mt z(=H}81~75tcHfQfCdica;ejR1v!kj+&(hDm(A(UMZaf|GCsd$&CDw9 zbD-EpA^9920_(f7`^!S<@opcBRYyPvdLwhCnaiHk zZILATyH~MUiMqk3z1Y3=>)6z-r9$%ZgK^%DwP~{r!52d&t9=ewGP!P*Tiw#p2TE2e z3dL#0>3nOG`1Vs~pZAlwRYd;I@&9r;42EY?$^OYlcJa8T#E*?g=dAo^$gMy5zg{+4 z>wNn<{ef0@!Sr@{{E=vHb97kLRX=#WO`BUnF0$diEa5Tax8;zV zqLhz1RJQgi(X7R!WlwT2?3_mT8+L@aSOMW+OMz%;@;K+NN7q>vwd(;F z^!gNl-c4HD&D_MzLcq+$;xz+tad7gna`3QnebnIO5a51I{I5j;4vyZNsPq4g0qSUO zW9j{WXQ=eGY2yV td-e*~EB%}QNv~z$;%ejQ08)|q#LCIWLqj1W;PqM$l9N`Fs**4c`ES^n#f$&| literal 0 HcmV?d00001 diff --git a/doc/pub/BM/html/figures/RBM.pdf b/doc/pub/BM/html/figures/RBM.pdf new file mode 100644 index 000000000..50bb768e1 --- /dev/null +++ b/doc/pub/BM/html/figures/RBM.pdf @@ -0,0 +1,1623 @@ +%PDF-1.5 %âãÏÓ +1 0 obj <>/OCGs[22 0 R]>>/Type/Catalog>> endobj 47 0 obj <>stream + + + + + application/pdf + + + RBM + + + + + Adobe Illustrator CS3 + 2018-02-16T10:55:58-05:00 + 2018-02-16T10:55:58-05:00 + 2018-02-16T10:55:58-05:00 + + + + 256 + 76 + JPEG + /9j/4AAQSkZJRgABAgEASABIAAD/7QAsUGhvdG9zaG9wIDMuMAA4QklNA+0AAAAAABAASAAAAAEA AQBIAAAAAQAB/+4ADkFkb2JlAGTAAAAAAf/bAIQABgQEBAUEBgUFBgkGBQYJCwgGBggLDAoKCwoK DBAMDAwMDAwQDA4PEA8ODBMTFBQTExwbGxscHx8fHx8fHx8fHwEHBwcNDA0YEBAYGhURFRofHx8f Hx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8f/8AAEQgATAEAAwER AAIRAQMRAf/EAaIAAAAHAQEBAQEAAAAAAAAAAAQFAwIGAQAHCAkKCwEAAgIDAQEBAQEAAAAAAAAA AQACAwQFBgcICQoLEAACAQMDAgQCBgcDBAIGAnMBAgMRBAAFIRIxQVEGE2EicYEUMpGhBxWxQiPB UtHhMxZi8CRygvElQzRTkqKyY3PCNUQnk6OzNhdUZHTD0uIIJoMJChgZhJRFRqS0VtNVKBry4/PE 1OT0ZXWFlaW1xdXl9WZ2hpamtsbW5vY3R1dnd4eXp7fH1+f3OEhYaHiImKi4yNjo+Ck5SVlpeYmZ qbnJ2en5KjpKWmp6ipqqusra6voRAAICAQIDBQUEBQYECAMDbQEAAhEDBCESMUEFURNhIgZxgZEy obHwFMHR4SNCFVJicvEzJDRDghaSUyWiY7LCB3PSNeJEgxdUkwgJChgZJjZFGidkdFU38qOzwygp 0+PzhJSktMTU5PRldYWVpbXF1eX1RlZmdoaWprbG1ub2R1dnd4eXp7fH1+f3OEhYaHiImKi4yNjo +DlJWWl5iZmpucnZ6fkqOkpaanqKmqq6ytrq+v/aAAwDAQACEQMRAD8A9U4q7FXYq7FXYq7FXYq7 FXYq7FXYq7FXYq7FXYq7FXYq7FXYq7FXYq7FXYq7FXYq7FXYq7FXYq7FXYq7FXiN2n5tKdMcz30t tJeagGghhuUmih/xBahPrMvqusitZmQw0jTjFy+0tTiq+xvPzF/xDNqtzFqUMtxa6da6xcR2PI2o Se+eaKyT0WW4WNmiHPi54uWrtsVQGhaj+cmleV49PtLOW1TT9DUWFs9rPJM8qWCMCB9Vkj9ZbslO ElwtVH90T8RVZnqV951ttFvRcvf3TaZ5ksEiuoLZ/rFxpQltJrhxFaRgyoBJKjcE3VSDU1wKkmt+ efzBt9R1PUrOC/h8u2ME2ov9ds/SPo2U0ErRKJbW39MXFq0oUNM8leojZSpVQ7eZPzvW3vZTCfUG myXdnCtrJKzySWizoqelaNCZYrl/S4vcAsq/3ZPxEq9ltIZYbWKKadrmWNFWS4kCK0jAULsIwiAn /JAGBVXFXYq7FXyZd/8AORP5iy68mpwzwwWsey6UIw1uUJBo5P7xiafaDA+FM6UdmYhGuve4/iF9 ReWtVfWPLulas8YhfUbO3u2iB5BDPEshUE0rTlTOeyw4ZmPcabwbCV/mT5qu/KnknUtftIY7i5sx F6UUteBaWZIatxINB6laA5ZpcIyZBE9USNB4L+Xn56+fZ/OtjZancpqFnrN7BbSwyIqCH15Fi5Ql AvHjXoag/M1zc6ns/EMZIFGIao5Db6fznm95X+fP5l695M0/TLfRVjS51b6wDeSLzaEQen/dofhL N6vVqgU6eGx7O0scpJl0prnKmO/kP+bnmrzBrR8ua46XqrbyTw37DjOPTKji/H4XHxdaV9zl/aOi hCPHHbyRCZOz3bNO2vnj87Pzl85aT5uu/LeizJp1vp5gZrqNQ80rSQpNRi4ZVUc6UA37mhpm80Gh xygJy3tpnMg09E/JP8wtW86+W7m61WKJLyxnFu00IKrKOCsGKGvFt96bewzB1+mjimBHkWcJWHoe YLN2KsZ/MnVtQ0nyVqV5ps722p0ihsJo0SVxcTzJDEBG6yK9XkA48d+2+Ko7yvr8Ws2MziC5t57K d7K7ivBEJhNCByJ9B5Ijy5A/C1Pl0xVifnrzP5gjOuPpF++nW/lpbMz+lFDNJcz3TKzRMJkl4xpC y/YAYs3UAfEVZf5d8wwa3bXMsdtPZy2dzJZ3NtciMSJLFQn+6eVCCrAghj1wKhbrVNRsPOFlZTSi bTNZilW2QqA8FzbIJCAwHxJLHyPxbqy+B2VT/FXYq7FXYq7FXYq7FXYq7FUNqOmWGpWxtb+BLm2L I7QyCqFo2DryXoQGUGh2xVE4q7FXYq7FXYq+ZvPv5WeWtH87BtOsLm50GJFkv7GOfgRKQ8rRxyMr sEEKhzXelfiA3G0j2rkEa2vva/DFvcPJfm/ylqFha6ZpTGya1iS3h0u5BjmRIkAVFDE8+K0+yx98 1siSSTzLYjfO9r5fu/K99a6/G02lzKqywIWEkj81MSR8CGMhlC8QOpyWLJKEhKPMIIt8/wD5feUv LmgeZV1rzBY3TWNtdRtp0olEi2rmRvq8lwsaI0is0TBXXbkpqu4zOzdpZJx4aAvmwEAH0xbXVtdW 8dzbSpPbyqGimjYMjKehVhUEZrWx5F+eL+WPMNmmkx2smpa3Y+q0FzbycEtdlMwZjVHY8UBSmxpU qSK5Om1M8RuPVjKILf5EeTPK+jpNcGOX/FcavHdmdtliEhjb0FWilOcfFyakMKeFbNVrZ5djtFEY gPUdZ17R9FtvrOp3SW0ZqEDVLuQKkIi1Zz7KMwmb52/May0bzv5mXVrHTLjT4Kr+k9UZqtLDFFzM vobqvC3i5LRtxxrSozP02vniHDziwlAF7x5J8j+X/J+kfo7RUcRSN6s88jl5JZKAc2Oy9B0UAZj5 9RLLK5MgKZBlCXYqlnmHQYtbso7Z7mezeCeK6gubb0jIksLckYCaOaI0PZkOKsA8xaL5g0y7n8ve UmupJbzQ71VlN0I3W+uZw31+Z2eNmlryHNAWDMo2XcFU3vvI2o62srz3M2lW2t29sNbto3ha6jns 94XSRkuYSWFElXpsOLdeSqB1DQdZ8va3aQ6PqF9qc2sy6ldXMNxNbQD1pYI41k/dpaqUibi1AGcC pUHfFU28paD5kmTQL/zL+6utF08Wsdu0guJpLt41iuLqeUFlJYRngAx2YljU0UKzPFXYq7FXYq7F XYq7FXYq7FXYq7FXYq7FXYq7FWCzQPceX9V8wRJ6lxFqkupQDqXisGFqVHtLbQMB/rYVSzXvKGi2 0kczMsOh3hQ2Opqqutm7nkkUytRXtGJ/dk7xMaKy1BChJY9a120jXVr+UNZ2UU03l+0vmleC5YIR 9YtrlgzmSi1hhuAzcW2beoVThb3y7psujWOrp9SibT20fWbTUB6dY0j9WOblUpIgdHXmjEAv1ril Ipf0vpizraXFzp2jXXI6eJFeJ9X+NeMFEBe2uHWqiRVV5R8RWtTihNIbnQ7C0t7bUwmlarc39sl9 YygRra2dq7XUcMPaSJjBQyIT6jtvuQuKqOq394dWuxYh9Gd5/wBJaHd3EDteyPcII5oba0G7rLJG WcS0py5cehCq2x00390Vv4jL5slkMA0i7c3Bh4UZru9kIAlhUMHSOMLEdgAW3CrL9P8AL+nx+YF0 oA3Fvp+myNeSSbvNPqctHllbu7LbPXwDeFMUpp5LuJm0KOzuGL3elu+n3LHqzWzcEc/8ZI+L/TgV PcVdirsVdirsVdirsVdirsVdirsVdirzrzW3nxvzEs7bRri9tNFu4rGG7u4YVmhiFNSeZl9ZJYUf kluGbjX7IPUDFWKWOufnKmlXl5Kl8NQvJdPlljntnEdrbyacGma2SKzu35fXFMbxiORlFKqOXqYV RWs+Zvzgs7e9uY7e8uL2KT/Q7W2s/wDQpBBZ2srqK2c90wmmnmVeXpD4Kc0YUKqYyXPnu8Gg3U95 q9vci+lTX7CC0CW8Jks5vSiiYWwea39cIvq83C8uRZSAVVSfTtf/ADjTybFJ6V+fMMWmym5sprLk kfp2h+qzRSyQlp7qWYL6sTSNQk/CtN1WT6BrH5gf4n0+z1QXVxp4m1C1mK2ggBjt7idLe8uZWg9F lliSLikUsbVPIK6H4Qr0bFXYq7FUFreorpmjX2otuLO3lnp4+mhan00xVT8vaZ+jvL+n6dJ8T29t HFMTvycIA5P+s1ScVYjo8FzrET+Unr+hdDme11Ofl/vRFG5NtaKR29Hh63/A9zhVAebriTRrW18o 3sTanp2pXVuNKUMrTtBDMkj2bhyKkcQiSMacW3IIqVUD5r8r+YLXy7cXutzfX9Ptomls7Bp2J06R W5xMZH4/WwKBGL/EBXhXFCCtLbzR5xie90ZGn8vwCRNLGqzM0kMtFBmgkRhK0yMG9My/ZH7VSQFU LazXGqea00i/juLjzbaxzrFdXVwLcxyO8bwzL6bMkYhjVqxw8uRP7S1IVT3V21byZqll5g14yavI krQzazGQgeCaMgW5tyeMBWUKylfhbfkQcUski8tau0a+ZYrhG81Sj1SA9bV7dgCtjtt6QX7Ljfn8 femKoryRfpqsutavwaKW4vfq7W0lPUhFrCkRikAJowk9Q/TgVXt/9x/nW5h6Qa3bLcx+H1m0pFL9 LQvF/wACcVZDirsVeYat5787aZcX0dzcaVEljr1lpkrPaT1+oagkZiuF/wBL+Jw0vE9vhY9qYqzL SNV1V9Q1Aag8Uti18bPSDa2s4YLHFzkNzJzmQUkV4+ZCLVfFwuKo3X9f0vQNKm1XVJHisLehmlji lnKhiFBKQrI9Knc02xVLtU822CabqrWdytrfWKiKNtQgmih+tTRerBGBL9X9UtyX4Y3rvTY4qs8i +ddP80aRbyxyqdTS0tLjUrdY5YlRruL1FMfqj442KsFZWYbHeoxVkmKuxV2KuxV2KuxV2KuxV2Ku xV2KuxV2KuxVIfOP76xs9N76nfW1uR4xo/1iYfTDA+KrvMGqXZni0TSWA1a9Us09AwtbcGj3DDx/ ZjU9W9gcVYob220LzD+ivLDQCO4tha3t5cljb29za85jLLJsJpmid2deQaqgsRhVLn06z1zX0tNI iGtS/VJZNR1zUCyxO87COGWIoAzJGElEaw0UHowIJxQsttMi1LzGNI806j+kYdMd2bU7hikNz9XR W+rRxmkMbxhq3BBZmApWnKirei6Vf6nqWqweVboaPBFI8w1NQ4+t212WeBBAQo4o4kVJ/tBQOFR0 VXaTpPlfU7fWtJ1aGLR74TlrIljIENhHxkuIriQL6xWcytJy+Kh+PFUPpema75itrqO4mFybe3dL HR715lcQXMQRb+2uJBzPMhvTEquVHwl1JOKpl5f8wJY6Wl1pkq21xDE76p5YvCYozJb1Fy1g5BC8 XRqqlU8QhrilEeV0MWj22raHV9ctol/xHo7AxSXDSEyuCj8aSK7sYZKUcbVodlU88w6jZ3WhWHmi wf1ItLnS9L0IYQbw3asDuCsTvVT0Zd8CsqBBAINQdwRirsVYX5r/ACzs/MWr3d7PeNBBe6e1lPbI lSZ1WZLe6D8hRoUupaCm5I3+HFUs8wWmr6Dd+Sbe2n1C7eG4vDrF5axXP1eUzW0jSTXccHOBQ95I jr6v2d6GgY4qpeU9a8w+ZY7PS/MFpeyaXqmgRwahNLZmGJ78qwvH5+mgVWV1Cfsv1QFRXFWtS8r+ Yre68taBHqN5PcTR6jLquvJDcC1ll+qiCFbyCGVYyDyUorv/ALr2IqcKodvL3mXyMbXU11dL6Gb9 C6RcW9pp0gmNpp5cOyr6t27s8cshZUTl047jcKzDyHqmvahpt4dct7i3vYr67VFuIRDS3ad2tlQh VWThAUBcV37luWKslxV2KuxV2KuxV2KuxV2KuxV2KuxV2KuxVhfnXzHZaRrlhcXRBXT7a4uYoSwX nczcYIF5HZRw9Ysx+yoJwqxjR9Q1/X/rMWjRSzLeSc9W1irWoncfCIkmILQ28Q+FVjDSkdfTJLFQ nms+SFtfLwvLtxey6OY7220yJBFYRrbuJJY47f4ubPGHXnIWap7YpQOo+cLePVNVXTpz9e1SaGzs 5YF9WRLS2iFZIoxXnI00sqxL0JBY/CrUUJrD5COr6TFaaypsdPt1rpukwOGMMlDS4uJd/WnqxJH2 K1ryrXFKHOvy6V5osm1vjDfJBLY38qjjHcQAGa1u4x2HON0Za/Cz06EHFVDRNKj82aPZaXeR10a2 /wBL1eQVDT6hcMZ3gjcUZVheUmRlNeXw/wA2KovX9O13TEhuGle8SwJbTtaVC91bCnxRXsaCtxbu NnZBy/a41HLFWPrrGl69ZaxojQxs2q3dnLDbtxlEUl3KsN2YnGzKvAzq69Q9R7KGUax5L1SN47rS Lx5prUEWn1h6XUK/yRXZDl08Y7hXVvEYpYr/AIwfRtUnh12ye0GoApq9myenFchhwNzAgLqs1DSR VdkkH2WL0GKGe+QtRW88tW8frC4ksS1k86kMJBAeMctR19WLhJ/ssCWQ4q7FXEBgQRUHYg9KYqp2 1tb2tvFbW0SQW0CLHDDGoRERBxVVVaBVUCgAxVUxVTntba4CC4iSYROssYkUNxkQ1V1r0ZT0OKqm KuxV2KuxV2KseT8wfJ0kmmImpITrECXOnHhLxeGRWdWZuPGOqxuaOQfhPgcVQ97+Z3km0ktlOqQy w3M0UH1yN0NtH68VxLHI85ZY+B+pSrVWNGFDiqOt/Ofl6W+vrN7uO2exuYbP1J5I445pri3juYxA xb4+STr71xVFf4h0QNqSveRxDSHSLUnlPpJC0kSTLyd+K7xyqag038cVcPMvlwxGUarZmIFQZBcR cavGZl35U+KJS4/yd+mKr7XX9Cu5YYbTUbW4muYhcW8cU0btJCSQJECsSyVU/ENtsVR2KuxVIPPn meTyz5VvdYhgFzdRenDaW7Hir3FxKsEIY9l9SQcvbFXmvnfyb5k0yJPPfmS8tPMx0qNW1bSDbNbw /VVcs4tmWQ8jFzZlEqty7+GFXstsIBbxfV1VYOC+kqDioSnw0G1BTArFPPWu69FqGjeWvL5hg1XX muCb+6QyxW1raKjTyCKo9R/3qqik033xVivlbyzP+XfnHTbG5eDVbDzGZLW21MxGK6trqKJp/SI5 yKYpUjb7NPiG+2FXrOBXmHm/S9R/MHzRfeWbeaHTtK8t+g17qDRGa5ku7qITJFD8UYSNYmVnNTUk Cm2Kpv8Al5e6rp9/qXknVRBJcaFFbzWV9bIYkuLS6MgRpIyz8JVeJg++/XFWcYq8W0Dyjrvm28P5 haLd2egNNJKdBtEtjMrwq7xie6JkC8pt3oibbb8t8KvR/IPmi48yeXI768gW11KCaey1K2jJZEur WVoZeBO/EsnJa9jgVZ+Y97pun+S9Uv8AUNPj1SK2i5RWUoBV5mIjiFT9n43HxDpirA7Tyx56/Lfy vd6/a6la3sVty1HWtAWB0hMKLWYW1w8ksqvHGvwlqhqb4VeuWlzFdWsN1CaxTossZO1VcBh+BwKp apqlhpVhNqGoTC3tLcBpZWBNKkKAAoLMWYgAAVJ2GKpDN+ZPlJPL1xrkd2ZLeBp4Ut3jkguJbm2Q vJbxwzrFI0o4kFeO1DXoaKrYfPmnWS3dnrrums6ZDDcapb2dnezRxpdycIBEyxP6+7cKx1qVY0Wh CqonU/NAstes7USW9xYTTpp96I3BuLS8mQy2/qqCfglWgAIDVZTuDsqyHFXYq7FXYq7FXYq84/5U xpwMxN6J1k1CW8jhuYBLEto9tdQpYEB0ZoUfUJH+1Uj4duuKqcX5PX0c1tdnX/Wv7aS1ljkubZp4 62i36InGW4aRowup8VDyMwEYqzE1xVCD8g9NQwiPU3kgiWOH6vKLhUeBbSztJEc2tzaSMzfUAwq3 AcqFGoMVZHfeR9fk1HzDPp+uRWVr5gkguHT6o73EMtvBDAOE63MY4MtvvSMOKni6mhxVg8P5GNca odMnvbtdMsbYyHUPjjE+oSXc1xDJH++ec+hBcvDI3qAsCPiqDQqzzyd+XsHlvUn1BbhZZZbJLOVE WfiZBd3N3LMHubi7mrK938QaQ7itd6AKy/FXYqlfmjy7YeZNAvNEvi6214gUyRnjJG6sHjkQ9mjd VZfcYqw+68geedbgTRvNHmaC98uI6G5itbP6vdXqRsGEdzL6joqkqOfpqOXTbFXogAAoNgMVY35y 8oTa6dPvtOv20rXtHkeXTNQVBKo9VeMsUsRKh4pVADCoOwIOKpfo3kzzHP5htPMHm/VoNRu9NSRd KsbGBre1geVeEk55vK8krIeIqaKK064qzTFWG675N8wJ5hn8x+UtVh07Ub6KOHVLS9ga4tLn0QRD IQjxOkiKSvIHcdsVRnk/yfc6Nc6jquq6gdW8wau0Zvr30xDGscClYYIIgW4Rx8j3JJNScVZNirz6 LyH530U3On+U/MdvY6BdSyTQ217Zm6msTMxkkFq4ljVl5sSqyKaYqyryn5ZsfLWg22j2bySxwc2k uJjylmmlcySyyHuzuxY4qr+YdCsNf0S90bUFY2d9E0UvA8XFejKezKaMPfFWFXH5f+fdWsBoHmHz TFeeW/hW6MFmYdQu4VI/czzeq0ahgKOyJVt+lcVehxxxxRrHGoSNAFRRsAAKAAYqk3mvy9da3bWK W14tlNY3sN8jyQ/WEYw1orR84+7clPLZgDv0xVgDfl9q318eVH1C9azbS9SuZNWWOdLaa/vr0TRN Ogd7dpYW5v2JDClAMKsv/wAN6jq+p6Pr2oH9GXltGYdW0xOM8dx6UyzQcZQRxEc8fqIeNSrFWAPQ KkPljS/MWr29mL+C7trSPWrrVrj9ISTNLGkM8i2VpCJyZOPERylvsAfCu5PEq9JwK7FXYq7FXYq7 FXYq7FXYq7FXYq7FXYq7FXYq7FXYq7FXYq7FXYq7FXYq7FXYq7FXYq7FXYq7FXYq7FXYq7FXYq7F XYq7FX//2Q== + + + + + + uuid:B356723E9A14E81188E88C6EC4CB6E13 + uuid:5da937a6-fc9b-4248-a315-698c69e6ff3a + + uuid:B256723E9A14E81188E88C6EC4CB6E13 + uuid:B156723E9A14E81188E88C6EC4CB6E13 + + + + Print + + + 1 + False + False + + 460.000000 + 180.000000 + Points + + + + + MyriadPro-Regular + Myriad Pro + Regular + Open Type + Version 2.007;PS 002.000;Core 1.0.38;makeotf.lib1.7.9032 + False + MyriadPro-Regular.otf + + + TimesNewRomanPS-ItalicMT + Times New Roman + Italic + Open Type + Version 5.05 + False + Times New Roman Italic.ttf + + + TimesNewRomanPSMT + Times New Roman + Regular + Open Type + Version 5.05 + False + Times New Roman.ttf + + + Symbol + Symbol + Regular + TrueType + 11.1d1e1 + False + Symbol.ttf + + + + + + Cyan + Magenta + Black + + + + + + Default Swatch Group + 0 + + + + White + CMYK + PROCESS + 0.000000 + 0.000000 + 0.000000 + 0.000000 + + + Black + CMYK + PROCESS + 0.000000 + 0.000000 + 0.000000 + 100.000000 + + + CMYK Red + CMYK + PROCESS + 0.000000 + 100.000000 + 100.000000 + 0.000000 + + + CMYK Yellow + CMYK + PROCESS + 0.000000 + 0.000000 + 100.000000 + 0.000000 + + + CMYK Green + CMYK + PROCESS + 100.000000 + 0.000000 + 100.000000 + 0.000000 + + + CMYK Cyan + CMYK + PROCESS + 100.000000 + 0.000000 + 0.000000 + 0.000000 + + + CMYK Blue + CMYK + PROCESS + 100.000000 + 100.000000 + 0.000000 + 0.000000 + + + CMYK Magenta + CMYK + PROCESS + 0.000000 + 100.000000 + 0.000000 + 0.000000 + + + C=15 M=100 Y=90 K=10 + CMYK + PROCESS + 14.999998 + 100.000000 + 90.000000 + 10.000002 + + + C=0 M=90 Y=85 K=0 + CMYK + PROCESS + 0.000000 + 90.000000 + 85.000000 + 0.000000 + + + C=0 M=80 Y=95 K=0 + CMYK + PROCESS + 0.000000 + 80.000000 + 95.000000 + 0.000000 + + + C=0 M=50 Y=100 K=0 + CMYK + PROCESS + 0.000000 + 50.000000 + 100.000000 + 0.000000 + + + C=0 M=35 Y=85 K=0 + CMYK + PROCESS + 0.000000 + 35.000004 + 85.000000 + 0.000000 + + + C=5 M=0 Y=90 K=0 + CMYK + PROCESS + 5.000001 + 0.000000 + 90.000000 + 0.000000 + + + C=20 M=0 Y=100 K=0 + CMYK + PROCESS + 19.999998 + 0.000000 + 100.000000 + 0.000000 + + + C=50 M=0 Y=100 K=0 + CMYK + PROCESS + 50.000000 + 0.000000 + 100.000000 + 0.000000 + + + C=75 M=0 Y=100 K=0 + CMYK + PROCESS + 75.000000 + 0.000000 + 100.000000 + 0.000000 + + + C=85 M=10 Y=100 K=10 + CMYK + PROCESS + 85.000000 + 10.000002 + 100.000000 + 10.000002 + + + C=90 M=30 Y=95 K=30 + CMYK + PROCESS + 90.000000 + 30.000002 + 95.000000 + 30.000002 + + + C=75 M=0 Y=75 K=0 + CMYK + PROCESS + 75.000000 + 0.000000 + 75.000000 + 0.000000 + + + C=80 M=10 Y=45 K=0 + CMYK + PROCESS + 80.000000 + 10.000002 + 45.000000 + 0.000000 + + + C=70 M=15 Y=0 K=0 + CMYK + PROCESS + 70.000000 + 14.999998 + 0.000000 + 0.000000 + + + C=85 M=50 Y=0 K=0 + CMYK + PROCESS + 85.000000 + 50.000000 + 0.000000 + 0.000000 + + + C=100 M=95 Y=5 K=0 + CMYK + PROCESS + 100.000000 + 95.000000 + 5.000001 + 0.000000 + + + C=100 M=100 Y=25 K=25 + CMYK + PROCESS + 100.000000 + 100.000000 + 25.000000 + 25.000000 + + + C=75 M=100 Y=0 K=0 + CMYK + PROCESS + 75.000000 + 100.000000 + 0.000000 + 0.000000 + + + C=50 M=100 Y=0 K=0 + CMYK + PROCESS + 50.000000 + 100.000000 + 0.000000 + 0.000000 + + + C=35 M=100 Y=35 K=10 + CMYK + PROCESS + 35.000004 + 100.000000 + 35.000004 + 10.000002 + + + C=10 M=100 Y=50 K=0 + CMYK + PROCESS + 10.000002 + 100.000000 + 50.000000 + 0.000000 + + + C=0 M=95 Y=20 K=0 + CMYK + PROCESS + 0.000000 + 95.000000 + 19.999998 + 0.000000 + + + C=25 M=25 Y=40 K=0 + CMYK + PROCESS + 25.000000 + 25.000000 + 39.999996 + 0.000000 + + + C=40 M=45 Y=50 K=5 + CMYK + PROCESS + 39.999996 + 45.000000 + 50.000000 + 5.000001 + + + C=50 M=50 Y=60 K=25 + CMYK + PROCESS + 50.000000 + 50.000000 + 60.000004 + 25.000000 + + + C=55 M=60 Y=65 K=40 + CMYK + PROCESS + 55.000000 + 60.000004 + 65.000000 + 39.999996 + + + C=25 M=40 Y=65 K=0 + CMYK + PROCESS + 25.000000 + 39.999996 + 65.000000 + 0.000000 + + + C=30 M=50 Y=75 K=10 + CMYK + PROCESS + 30.000002 + 50.000000 + 75.000000 + 10.000002 + + + C=35 M=60 Y=80 K=25 + CMYK + PROCESS + 35.000004 + 60.000004 + 80.000000 + 25.000000 + + + C=40 M=65 Y=90 K=35 + CMYK + PROCESS + 39.999996 + 65.000000 + 90.000000 + 35.000004 + + + C=40 M=70 Y=100 K=50 + CMYK + PROCESS + 39.999996 + 70.000000 + 100.000000 + 50.000000 + + + C=50 M=70 Y=80 K=70 + CMYK + PROCESS + 50.000000 + 70.000000 + 80.000000 + 70.000000 + + + + + + Print Color Group + 1 + + + + C=0 M=30 Y=70 K=0 + CMYK + PROCESS + 0.000000 + 30.000002 + 70.000000 + 0.000000 + + + C=5 M=70 Y=90 K=0 + CMYK + PROCESS + 5.000001 + 70.000000 + 90.000000 + 0.000000 + + + C=5 M=90 Y=75 K=0 + CMYK + PROCESS + 5.000001 + 90.000000 + 75.000000 + 0.000000 + + + C=30 M=0 Y=95 K=0 + CMYK + PROCESS + 30.000002 + 0.000000 + 95.000000 + 0.000000 + + + C=60 M=5 Y=95 K=0 + CMYK + PROCESS + 60.000004 + 5.000001 + 95.000000 + 0.000000 + + + C=30 M=0 Y=10 K=0 + CMYK + PROCESS + 30.000002 + 0.000000 + 10.000002 + 0.000000 + + + C=60 M=10 Y=5 K=0 + CMYK + PROCESS + 60.000004 + 10.000002 + 5.000001 + 0.000000 + + + C=80 M=5 Y=10 K=0 + CMYK + PROCESS + 80.000000 + 5.000001 + 10.000002 + 0.000000 + + + + + + Grayscale + 1 + + + + K=100 + GRAY + PROCESS + 255 + + + K=90 + GRAY + PROCESS + 229 + + + K=80 + GRAY + PROCESS + 203 + + + K=70 + GRAY + PROCESS + 178 + + + K=60 + GRAY + PROCESS + 152 + + + K=50 + GRAY + PROCESS + 127 + + + K=40 + GRAY + PROCESS + 101 + + + K=30 + GRAY + PROCESS + 76 + + + K=20 + GRAY + PROCESS + 50 + + + K=10 + GRAY + PROCESS + 25 + + + K=5 + GRAY + PROCESS + 12 + + + + + + + + + Adobe PDF library 8.00 + + + + + + + + + + + + + + + + + + + + + + + + + endstream endobj 2 0 obj <> endobj 22 0 obj <> endobj 23 0 obj [/View/Design] endobj 24 0 obj <>>> endobj 21 0 obj [22 0 R] endobj 5 0 obj <>/ArtBox[8.69482 21.6172 456.34 158.043]/MediaBox[0.0 0.0 460.0 180.0]/Thumb 46 0 R/TrimBox[0.0 0.0 460.0 180.0]/Resources<>/ProcSet[/PDF/Text]/Properties<>/ExtGState<>>>/Type/Page/LastModified(D:20180216105558-05'00')>> endobj 37 0 obj <>stream +H‰ÌVÉnÛ0½ó+æ(¤p_€¢‡8A× "4‡¶’Ø®Ý8.² Eÿ¾CR¶ekUÔˆ`â†3óÞ<’c}ÁÑÙˆÂñÉÈ¡À,©)d~ð8!—°DküÜ‘£7¾?‘`Á€šk'-H‘;£5ÜÞÿ›I41ÿ ÜA¶úÞ’Œ‰8ã<ÎÚ‡4Ìb¶ÒýBXŒ©úE"¦ä¼$Éàð$œÁÐXÎÈòÜâL½<©éð(ÕtÒ~ñ”’NC ´«“ êåIM‡G©¦“ÀâM‡G)êT6ûÜb'åØI·º¨‹Ý Ÿ×4Ë`±£uG7}# ïÁXo¦[P4‹!Šï,‘ÐàÏ©nñG©šü¹ì©ç*w”Ê&¥dΤÆ(‘Kgú`µ)Ƶ‹;´iíióâwÅ h½az£´ ×ѲÅ“í´Ts)1-ŠuÅ´(Ö ÓeK±ãb}«Â«_Á®ü¨˜’ÐaüUl¹d¹Ç‹{ò%y›b}É|<ž,áãuj“?)¾O!ýV¼÷ ýNÐïsÊt2šß,& ~§a%ã»ê1´´mÓg&7NbKSÈ¿|ÁÖçßÙÙ²U½g¡¯ðâòà >Á¢~Ûi?Õûa©*–Ú*ðÐ(-™nõTKýsEAi®H°nõþOE‡A©V´sÀ‹ò|sŒøÆÄ.dd€<¹N‹„b²Œæ\A1&ÉÜ›0‹ Ðn°G¬f¿&¿Êyºš·BìžnGg,·Öa9¬t¸ ó#~µN_áöŠR~üz'¿2ºJoVK°ÅobÆs‹º<§VóeTÊIY+vÖ˜Ý*¹É^Soéë•5KÓÐÛý­Q0 ŽçÚh{û»ÐÛ—©LžS½:UÉõ-ÚDò<ÿ¹|Z·îÓ³àû+Àºp2÷ endstream endobj 46 0 obj <>stream +8;Yhp_%?bm#ip8bpdkAMOg@0PcqbY3 +0=YAsHN9fb@po~> endstream endobj 25 0 obj <> endobj 28 0 obj <> endobj 30 0 obj <> endobj 26 0 obj <> endobj 32 0 obj <> endobj 33 0 obj [34 0 R] endobj 43 0 obj <>stream +H‰\ÁjÃ0 †ï~ +ÛCqš]C`Mä°v,Û8¶’Ù(Î!o?Å L`ƒüÿŸø-Ý´×–|ýÆÁv˜`ðäç°°Eèqô¤Î%8oÓÞåÛN&*-p·Î §–† ª +ô»ˆsâÏ.ôxTúÎÙӇϦ;‚î–¿qBJP@]ƒÃA½šx3‚ÎØ©u¢û´ž„ùs|¬¡Ìýù7Œ çh,²¡UUHÕP½HÕ +ÉýÓwªì—áÍ]^Ä]<]šìÞß7N¾Pva–ÖC¡¶£~°±oÈ endstream endobj 34 0 obj <> endobj 35 0 obj <> endobj 36 0 obj <> endobj 42 0 obj <>stream +H‰„V Tçþþ™ÝAv eEf‹¼|QQ‚š“ìEÞ"EñEŒ•Š¬QQÑâ“øŒ+*¢1"ÖÍiRi<Ç#›xlRêI[k9=èöî솠'mþÿ¿ÿ¿óÝûß{gÁ Dx¤¾’üúôfÓW…$éزגÃ" ã×vÜ|’¥fä¥åß^ÊKÄ_TkV‰ =´q€fÉb²óçývt[#඙x,^Z˜›{~90€xž“•–y·kÊ€íÁøÔ°ˆ§÷axNÞªu;*’4€§à‡.]ž‘áèŒt¾"/m]¾ök7 ¿LûÅeiyY•²eñ÷õùË V•Þôtß"~Wþʬü¿‰Þ×€@û8|uÇ0Œ3Å¿:Z¨H»]Ñ°ŒŒ¼|’9 G³š"£RhkcZŠ•A$^…`Ú×¢PÛú¨Í.êÏô6?ŒÅä µhD3>G×’e¢}WÑ@s+Ó|­OsDÑœS4õÏiv+š³Š¦ú9ÍEs^ÑÔ<§9ªhŽ)šãÏiN*šSŠæcEãÀ_©ðª‘¨ÏêBÕä¢x`B)QîdÍâ ÇE\Â'àUU¬TÜeÎJÑ{Ϲ²vŠâ¿Ñï)þ2‘¥~ïëäó W€è»Pû´–n²ŽËT¢ÿÓ#*³ãžx>No…Š_CÄ~2©Vö¸Ãþ +ª`Œ„Sñb6P4N’Çgpÿb»¸KüD¾Š?ÀåñWùüø¯EY £D“xRÒKƒ¥¡’,“ª¤óR› ™—ÙMö”½doÙWö“‡É¢pÖ…ýŸöoíßØÿd¿oï²wÚ;ìWíŸÙ[ìõÐÙ7SÀžmŸöìàÓOŸ–½z{·ÐzÂáZߥï2u>éòé<ѹ£ó@çÞÎE@§;б—Æ6+:h.íBs>…Å1íÚ¯9,´ßþ]ûÃv[{p¯ëîò»†»ƒ” %kg¸2¢ÿû)íG¿Ûo­£,ù¹g;*û\ãǹÆ5Ыíw®ª:‡F×êì"†ÓÍÎÄ,ÌÆxºÙT¤!Ld!‹©;ŒÆäâm„a)–!Ð&|RLÇ4ÊÚØ+ÊQ†­”Åûp*oBC¾„æˆCµÚU ÃŒs|¢p”KA®jMÒ)+ Ó¹9çVƒ'ª ˆà2]o=òø‡\ŠÆ„¶šR¯÷1¸tJ¨t®Wp›5²;ì–ÄýWY/û‚%Ž¿lð%OZ„F¸Jš(9U¼SŸ@új³s”‘M¾ßêÈÿDöˆNEPÚî$ù>$6âjøPB…ï¨0fщ¾†¸a}v ·…ù¨%„Ô©Š®‘o`QD•×™ƒ¯¡<óÆ毎†¤ +áRÔedË&Læîpk)ùŸp%Â9Ö‰mB=ÅÃ›Õ +%\:sÆd’„­¤ñ¡ÈH´.  Àc! —X(§£»9Þ›ÂC΃‹Wg¢[xÄzX˜:HÝÀzp%haãÔ“ÑË¢ÔWX´z :‡Þ›NÜ´Þ¾±˜Ê£È±©äô|¬¥Ê-#·WÓ%´SàüŸESMt‚0,V£T<Ç`u6.hf¦-v®YüE +5¾ÀŠ:hC¢Í³Pl²ÛÍ*ƒ`± Cm|Ö¦ + +¼ÿß”÷C³ÍbÓÄ™\fãRM$L6éàHLò8Sh4ÆÙMP'šÏ0¶ÝÒÄ웚`ò¿HŸT~Q +©µFQŒ[b²±TbÜŒ$%ånãéUñIæ@‹h­33­b¼˜“–IÀ”•YVKAL6/¡yžY²ÅZ }d–Åò2Ùà°£RìX-d!×e!W±@žÒ&ãlÑÆ'šçšmE&ƒ-Öd1H’gkN4ÛšMÉb¡]ž}HiÝ°d° ó@Âì9Š/§ŠA¬Á‹Õêä%[‘Õj°’.¾ Í/^ĺMP,òAqM¬(QQJ‡€~ H„Óâ²Îû8B*YB©_f?«Ve ‡é÷Œ¾±n*¨™VàT»uïVtm·Ún…¿¤—ôAô­ÏV¡·€7ô>xV­Øó÷•ê‘Î[ýlžj-_CíàÍØ1ˆPû©4^n"¯óà?ÒùûDjjFŒ#G‰²ŸÞÇËO«öòQµ0EŒÕµµv·u·µê½£ûþ£é½­ßF>ŽiíŽÔûF{ûF‡3ŸAQ‘ÞãôQz1~2‹ÆþÃ~ÙWqUüœýzï%i€$|¤x€0ò€Pž|ÂG¾J0 1F¤Ù¼·¼·Í¾ÝÇî> ØHCÚ2X ˆ©¢CQ#ƒe€b¨u¦LgDÛÒÒqÔ§êS;ˆL[žçÞÝÄ€üåè¾³ûÛsÏ=çÜ»ûîÙ;¶H™>mfi>²³2¶¨ÀST.XX¹@ SlŽÖn=zi欲ò2Dµ5‘ĹÏmoÛ:^è3kîì 2€Ñ1¬¹ƒ¹Ÿ´olÙÐ4)XX`wÿäô¹þÎdaiÏŸ—ÌšSU_ðZc-æ-Á[~=ab¼©ž{tÔ’¿ã˜ _ã¿wyÏìúÆ“©Åwú?é ¾¬°ïÅ/Ntö¾(ƒ-wúÿñÛ`Ï]ß”ì7(]¦…‹ýn{ELè‡ßI½­ôC-4ít¿‰¤]ŠCµ·’ì¢Ù´=XLrŒd©ÏóIÊHšI¢$$õBŽ‘tQ棕]ÅýÐM>›™P¬Äwà;xŽ±+-‰]â)8Evå³bqý~­RoÂNÒ×*‡(n¦ ‘lµ26“îm)ž}ü]'ýé”Ñýšõ£> +Ò (°Å÷•â×+X¸ÆÁ8I~"G˜ÈÝB£Ü ÏȃÙ[r D¤+P)½³™ÈƒP&@óÑQâådW(¡ë ”ʳa ][ÈÇ\É!€ê3U~ +&*©0 B±²‘VÂé‹äîìG”ÇöŒð6ô’߃ŠÉT’Ë$=4¾×$ú¨å7ƒ › 6î{…✡¹YÅ„1ÍO·¼ +jý¶$Ç©ÿ~åXOÏc=›/å–Šð’/=LȾšbÔÒûðuW%{|6{‹|,¥\Rd!Ûý¾Ì`¸†ËÅ›xŽæ¥Í»rÞ+t¤<¦Óp|FÚ îóY¢Lû,“þ>+O» ƒ0–>m={y"×ÿÅgÒ£—³@úBësÆá,Î"³Çe>3û&ÎéÇ`Òç}þuq–I_€û|‘Ísœ©ØB1žò9E8È9@úÑøŸ™ÞË-Húñ˜õ9…Â8Î!–§0Ûg¦÷æö–§Ðà3å)ÄùÊeù]>‡ (ìã{ÆBÒç}fz–§H1ƒ'<ïsBÂib‰æh»Èy +c‘Í[1LeöâŸÉØH=Q’ȯ"¶qVˆóÅvÎâ<ÑäâúœsÙÛ%îáœÇmŽpÍõÇ9!ÎÏr.àþ_â\Èm^á\Äû¾Åyçßs.æ±®s.áúÛŒƒŽÂÑ8 °‹p,ŽÃñ8‹ñAœˆ“°'ã ãTœ†ÓñS8gb)ÎÂÙøi,ÃÏ`9ÎÁ¹Xœ‡óqVâB|a?‹‹q .Åe¸Æ¸«p®Æ5¸«±k±ëñ\‡ ¸?‡Ø„͸[ðóØŠ_À6ü"nÄ/á&|ÛQÅŒa5ÜŒ L¢Ža'˜B-Lã´ÑA3¸·ávìÂøe|»ñ+¸ŸÀÜ…½ø$>…Oãn܃{ñ«¹Ú–Œ¾U543¦…TÛ¶¶Å­m¦G†¶ÙÍáÄßâÇLÚkì°Üd@3´”fºy:­p¶£ÅØÿV¦Í2s˜ŠÖjÃMc‡+¯0ÒIUZ©¹ª¼VM¥ÔÀê´£–)µ‘J\Mêæ$‘TcÖ©é´ªÔ«©Ž¸*<’Öe„V=ÐÒc¶e +ëu±1iÉMz"¥ŠÍj&°Ás%®OêbÉzG—U¯ƒÅK°xr\3Èý)]MgÑ:y4Ëff„ízÀòB‰6ÅqX…Ÿ#¢Ká2~8ªFbŒ„ne+¥%Ô@ÜrSª›TRºI‹Bób™q9©©¶+ÅŒLG ®«)¦rÒj\ËKÛVÚ²ÙÔ©FÈVã¬iÛƒN&–t“ª›Óíò^¦±QUQ?Ó×étÚδvA1DÙ«HAÃ" +UŠ¢DT@d¦}&voÞÀ´ˆ°¸¢ j\¢Ü5\1u)Æ —(òÁ•¸ï+îÆ…ú?ÿ÷¦ÓúÙ0ÉüÞ9çnçÞsÞ»÷6$L+‘N…]±%pâV¢-ˆ~cvkˆ@Za:aÛlBYµ$[U%[¦J°5éxf7öÑ„†¿WÑ e&AQVK[½µ4BÀ2[á¼nîEè×Ë¢D2¦3Š´6=1iû#­±„ÄêÛ¤mš-–Ó7zsÅÌ v¦µ4µ# ͦ£ž8V_-ê„=Ž9V?5ꑇË1ëH¢ØÌÄS|ÁÙ $ÉÆö£*¢¬¤¨ý0'j +Ëš|„·&`º±.ŒixÐQ±±ÍVקœÅJ«ìùÓG‰:Å+Ì8œqlìßÅê¨é¶.é•[⹂¨ lfÜ +^—¡œ‚ê}Š¢N¡—þ5Å fî 1å¥I;ÞnfrŽš™¾ ‰j9OÍL¿e„ÚÝÅRª)d›1]L›Zói4["v³[˜«‹3HIoU(¡\Mh…³šìH3Îl…s=!Š·Ä‘9á¸Ä°¶æ×¥ídU*™ˆ7¦S˜],‰‰%"m -Ñâ;câ„Þs¬÷¬ñžã +ÜÄÆç?PíëÂÍ¡T‚Õ¾nßa2OfÉ$… F¤´ºÛWŽMv¯löRçY*°™õ¯S‰ ÚŠ-©Ì³TÉø~éÆ^RÁŽVõvT‚»Öf쎋e´g É&ì—ó±“e-a¹ä?–Rìoý‡/ÃnÝ×"]¾Ø@ÃÕûŠ†Ô­Šœ^ۓ׋ÞA\ÁòIn¾<Ê>R”ÒC$ÿ!ÿ&ÿ"ÿ$ÿ '#%!&"$?ß“ß‘ß’ßðZû5å¯äKlüxª¦²!_PþœüŒü”ü„ü˜üˆü€|ŸÜO¾G¾K¾C¾M¾E¾)—âøàÇóXhûhÛ9_Þ ¼×£Z^§üù*ùŠì_¦ü¹‡|‘|¾?Où9òYò²›|š|Š|’|‚ì‘düêrµž½_-²uC~<Pë@É.y G%?K plj1äQ>"kŨíAªmANúñ\ëì÷Gx¾î”ѧÚvÈvYMmªéHÛ½´Äµ‡¤òýlsy/yû¼[&ƒwQ¾“ö;8ú6Z¶zým•©Ô´?µ8¢i»Û¥¼ 11plÓÖ·r´[(ßLÞ¤yŒãÝDðFZn ¯g›Yz¹‰cndéµ¢—×kXgK¯&¯bL¯d+ÈËe¤²ç|»Œ–õd§—;x1 ÊŽœk9Ÿu¬±Žq[Ëù¬ÁúíGÉ–¬‘Ý(éÕÕÏÔ^—} +lZb€º"jÑ·U[^L®”vÖZéj§µÚhÉ+ÈåœuºÖ}¯SøØÔ¡eÊÕØR-†,‹Ñ_Æ Ÿ¾Åè[ô-‰ëÆpØ’Œ~’¾©Åf¶YJÆH“l /"“’‹È…²€ïúBj èç”Ï'Ï#ç“ç’óÈs©¹”Ï&çg‘g’³Ézrfχðô ö?ƒ–é|£O§|YKNãl¦R>•<…œBžLN&'¡µœSyy"9žGÖÀfÈXÊ'cÈãÉãd&XMy49Š)#0†OÕF0C‡SFÅUÈ€ÿƒÀcñ.>j‚52pÒúGÓ2˜òQä Z7Ž$àØYëprYå± «S‰ ×0øQI[._õÐ*8b¹WVηQem’Ú¶”oW˜–YB“Ed,$rX€¯«!Ü£„{”dß­<öì£ÿK0û%°v ÷øoÁ'þ»ñˆÏ7»sƒ/U}H~rh†ù?ƒÿ`þ6È endstream endobj 27 0 obj <> endobj 39 0 obj <>stream +H‰|R}TUŸaÙY¿š’iw<;c”@ +‹Y˜d~‚Š¬T®Gbõƒow—MbDS–å@¸ˆa‚z P”t?*‰DÁö¤A"²XVbt3bNhÈSiÜ&#§ã ú ›¤j}*g2èRõ™:C:—½žûŸ›f-Ò¬Þ’£çæp©úõ†K ±` ‹Ä°·¼±e¶ÃVâ˜Æ Ób%µEV`ù˜;Œ â+ñ øß^Y^—½@6A/«’ z‡y›¼o—vˆ;pɾÔ!+õ‹]qîb¬èK̓åð€@ºHƒ'pW*ÜDâ¿>ÌCRLˆ´ÇCžˆ$!¨Î ½· 9¿5 ýNÙ-—šþK= ½Ðz.ujFýЬ QmÙ¥öÐþã¸Ô +2£Ô!Ž>}ʼnÔ*Ž†äåÒû‹ÍNÐ QvñUè nC+VmÝœ¢:°J^ßt®¦yd[ÂnW •ÿ4^¹°cÑhogSË•• ¶U®5 +­mJõ9š² Û‹‹TÉ))±L¢þÈ?^û«À’=|ŸÞ,FõAzžcN8)»&ÓÔ¶$A+¤©n”½|@n%bN\ÉécüD³`Qÿ†&1Qš ¬_„ÃîhºÆZ]¶Gu}ÿ•Ö«ÌÏs^A²Õógż×Öµ•ö6›’\Íß±>¼eD&Ü¥yCÚ‡k˜ zÁ{ìêoä ëØ7äÕÆu{ßePèËCA(ð—˜~íBmõA–ìâo¦šÅy½½-#Ð é\Zµ ŠS¹kš£4! EXXp?Æ÷Õz鈪LêHy¤¼Ÿ(n,±ŸQŠwaîrº ¤°¸PEÙ—~ÇÌ\ûÃ蟗~ì9¿N³—µò–m¼’ø^x£ü»ñ–»`’‰É`¦-Ue5û•#:`Q(šh:š>A·/ÖÖSmë–¤éùX&ìM;LXK7ÝUyêÎóèP¸¦„%Q}YÌÏûpŸ¯rR ÂWÌsç+¨±³®H:ÝdÌȬ7k®?t´„=f>”¦"Ñiü´y¸Ã)sø:]Ú0‚ì”ÊÌuðÒ¿9wR·%ÎûÊö•ÛTö}í­_2ƒ Ë¢XtÜ m +˜ëzÉëšÄuF•@Pí»äQ„5Å’¼FI~Î÷ˆz¤?s÷epLÜA#å´™H‚ÇfÀ 0žûC`ðbYa)ý¸Õÿ5„Ǿ±bàŸÚ},½èmIÊìA>ï3ÏQ]ßa°IJjʪU=ÕgÏ}ÍTW•U,µwPAu•íØiÙÎĽ“¼,—ÝMP Å¿ÈçÖD‹6^IrOµ\¼ Ê!Ù©ƒt^“½à"Šž®¡o“Ú"³±‡²öŸVžh8ÚÙÞüA᧪úoäŸä$Õhÿؘ¨U­k‡3ØkÆÃ&2%3=æýuûk7K½›Ï;®ãîɈ}tèü%ß¡ZE|UÒáÍ*c]cA+Ót¤¢²–ÝÝ#ßýa`d´[›®²P>בŒSéÈÇ욘‡W‰keUb> ‹Ýñòr¾lç.å®â»‹Ty‰™›Œù›s³ +rwÖGûÕ—ŸÙw¦åj癡n˜ ~0êBÌ@' 7n__”5E v4¶ðǧÁ)?’·¹ÔÕ(ѳlv›ÛTIˆË*Pl…;ÛÖ^ÑP¡PÕ%-,ªù{Òøëœ哞ϋ¥ô? >íà endstream endobj 31 0 obj <> endobj 41 0 obj <>stream +H‰”—{TULJ÷™}çœ;<¼""A$jÄ*(X1&¾«‰¶iQ[Vk­ˆ(( Y©Ú¬Dc|5­(/ª]Ñ¥F.7Öc•%Ī1_ˆÏø$jkmì>s&§´]}ä»÷oæì™ÙóÍœ3sÁ€P˜ㆎ=pËð´Fª©x¢ð…‘Ý{4ž3–êÆeæeÔžšBå +€ø±™/Å}“Xw`p €ufbÁ¤¼ýÃï˜I“rgL,ü°s@Z< 21;+cÂÙÔ);©«ÅÔ_¯lªð݉œNåýTnŸW4}Û¡î¯Sù@JZîÔÌ Ø=à6@I*•ÇäeL/ˆèl–ZAñqùyYoUe¡ò»Þg ¦N+ºsv‚{‹¢ +³ +R¯#“rö\ƒ¡ÔØ<£0c<ÄeÎ(Ì…Q“ +³¦ÀØì¬ñ…097£(¦S”1:mpÄ<~ µ /4ƒæb—¨NPoÁ ! |Ax(Ê~®2€¹šIŒ©ß£†ÅQç¹G?ã-23ó +`–²ó”]¤l‰²¥Ênš›3 ¶MÌÉÏ€ÊVääçA¥²s¦MÍ…£ÊÖP`Ô*{‰HæÂueoçeMÈFeÒc”5)MCÍÔ±¨#kÏÿSÍ?”N»à&ÖÛÄ5±¡M¬PÖéG6±!Ú6ƒöÉÐÃc`äB¼¯Á"Xkà-0 ;v"l>Ç›/«U6D,• òí)CòArp¢öýmK> ò3‚jÊñá1Žo xò-•FcQÿQ[œ~¢*ô,|ôk«Æ#2Œ²þ½}¬>ªäóyŠhcÌôÌóÌS L iàé"Zˆ™â%ˆ¯ˆÙÐJÌs FÌ ¡µX*Ö@œ¸# «·“· $y½‰ì½å½)r¿¬„^²JVAïðÒ2¤%ôƒO¶w”Ñ–%ã÷=é}` <ið3“¡f«Dn)¼ ë``T@¨ƒ+ôö7Â#ÃcHk µÕ*³ÞSÞoíT¾Üz_ù€µ‹|©?)_fíVÞoíQ¾ÜªP>`}Œü^*ù)úCåˬ}Êû­ýÊ—[•Ê¬(ÚoUQ©œ¢?V¾Ì: ¼ßú³òåÖAåÖ!Š.·S)@ÑŸ(_fQÞo}ª|¹uTù€õEþ…H6Ð;?ëÿ"R­f¾Õ:¦É×dNh25šÌIg«uJó9­¹œÑ\Îj.µšÈ9M¤N9¯‰\ÐD.*"—4‘ËšÈMäsMäª&rM¹®‰ÜÐDnj"·4‘zMä‹ÿA¤Ja#øÿ#‘ÛšÈMä®&ò¥&Ò ‰ÜSD5‘¯ôŽ¹¯É|­É<Ðdþ¢vÌCÍ篚Ï#ÍåošË7šÈc‡}Va8Dsˆ´‰CDp‡ˆ0"ÂrˆáAßH%†ãPKDnB<4˜,‚"ÂëÒ!"B""Ô!"Âl""Ü!"š9D„Ï!"š;DD„CD´°‰ˆH‡ˆhéQÎŽ­2"Ú!#ž°wŒˆqøˆÖšO¬æÓFsyÊž©ˆÓ\žÔ\Új.í4—ö—ïL¤Þ%ÒA‰×D:j"4‘ΚHE$A骉tÓDºk"‰šH’"ÒC驉$k")šH/Mä{ŠHoM$U飉<­wL_MæµcžÕdúi2ý5™ú²vÞêZF'„|:1Љ×`ø!¤ËjúÒ?ö,“Ç´*–Ç•J£ºZËRCTÜI­Šå)¥ì¸ÓZÓ8ÑtBvƒÞ´#à'0Ž¾êEð2,gܑκ#Õº#sGªsG:ïŽtÁéâ·#É›¤†ŠATwK«bY¯ÔªûB«ÿ–Ñ%7£ËnFWÜŒ>w3ºêftÍÍèº›Ñ 7£ÛnFwÜŒîº}éfDï¾ÑÍèF˜h kÇÚ©³˜nn!=Õé_D«6 Zý{ΰŠî~Ø Õ´Ð–F¤gt6’gaÆ‹öÍÍ»[©Ô~WU~«Ø'¤ÞTꈫ>uÕQW}¦”};”¬ÚÖì2Ùõì˜uÜU'”BšE(D°ÕÂÎd1³³X¡bN6‰‰dvN%ì#@Š,a§ÜžN»êŒ«ÎºªÖUç\Uçªó®º §õ =ß:2:ŸÙZ‹ÎgVJþcŠXË-eÝv—ô¼-ö[Jk´Žm¤øMl+3?óC °w œý‘mÛÁvQÿ¨n£ô^Ð_ݵÂíÛ)µ]OÞfoSŸÛ)Ù¶‡îg´Úl91¶š­VkO_zu— ú¶±Ul´fk؈¥>>€6Þ oôó¾ïÝ«ú÷©›ª-¨”í¶Žl>­VÓ7œ"p(E|­Ô0W=çªç]5\)›—¢èÿE{Õ²Aµº§Z4ªè¯Tä}û è…=s:GÌùæi·3Òa3ÞÄ8ìˆ Ø {`/|_Ãy¸ƒoàR\Ž%¸ +Kñ¸·àVôc·áÜ…¸«ð Áj¬Á3X‡—ð*õU·ñ.6ðŽ…OåÓø¯ø¯ù+|6ŸÃçòù|!/æKøoy1_ÁÇWòÕ|-ÿ=_Ï7ðͼŒ¿Ãßåïñ]|7ßË+ù~˜åÕü8?Ékù~…_çõü.oäø#L)Li†›>3ÂŒ2£ÍX³ù”ov2»˜]Íîf’ÙÓL1SͧÍ~æsù 3ÃÌ2§y·y·{wH&M,C¥OFÊh+Ûʲ£L‰2Yö–}e9X“#ää(™.ÇÈqr‚Ì–“e.Þ„í+GlCkñÀ° v¡5èŠ]i­“0 8¦` +˜8gƒ…spœ‹s!çã|Æ…¸¼¸ƒÄ%¸B°˜V/Wà +Õ¸Âq-®…f¸׃7àhŽ›q3Düý²Îñ>ãøõAþ|’OR>Ãg(€Ïñ9 +ä‹|‘‚ø_¢`¾ÂW¨#_çëveÞä›Âwù.uæû|Ÿºð~@¡üR?âGnW@(EØUI‘HD"uE’( +)H¡h¤!bÐý¨2AÝ1)™È¤8ä ‡z ¹|äSOŒÁê…BRo” „PŽrêƒI6Ñ$¢ÕÔ5¨¡$ÌÀ JÆl̦ÌÃUà!ÒDÔ¡Ž*ñOhžâ)U¡õ4Ù°ašbÜ;U—qÑTãiÒ‡ÖK²$ÓI—tú¹ ´QË`zO†Èª•á2œÞ—‘2’6ÉXK›¥HŠh‹”J)ýB*¤‚>J©¤­R%U6:í3íKÄaͱ\ÇËøm~—×ó{¼™·ò>À‡ù¨¥íŸâ³|ž?åË|•oðí~ÆuC8/C†"£P€b”a"&c^ÆLÌÁ&|€_âWøµ]QŽ#8Ž8‰3|ÞÎq Wp7qð_âk|kÆÃ4ãÛÈ2ÞdÚ›*<[šSŽë®}öá..ñ”Vâ#í%@:JWé&=¤·ô•Té/ƒ$[†É-ãd¼L)ö·Ö¨ÙHÍæP§9Õi¬NsSwA­eÔWîê«&ê+õUSõ•K½$ê¥fê¥æê%OõR õ’—z©¥z©•z©µzÉ[½ÔF½ä£^j«^òU/µS/µW#ù©‘üÕHÔHj›@µMÚ&XmÓQmÓIm¢¶é¬¶é¢¶ UÛ„©mÂÕ6j›HµMWõ@”z Z=£è¦è®ˆUÄ©z¨zªz©z«Ô}Ô‰ê¾ê$õ@²z E=ªHS¤«ú©ú«2ÔÔÕƒÔ™êÁê,õ@¶z G=0D=0Ô>!Ð0%:WY®,ç)¿#”ß|åw¤ò;J™­ÌŽQfÇ*³Êì8e¶P™-Rf‹•ÙñÊl‰rZªœ–)§”Órå´B9¨œV*§“”Ó*åt²r:E9­VN§*§Ó”ÓéÊf]¯·hs8ÇpÁËy¯å \Ë[xïåƒü>Æ¿ç?òiþ3_à¿òßùÎwìÓÌ?-›_X6#,›Ù†q ¨ÄLÇ+˜…¹ØŒ­ØŽØƒCˆÀoñ;üŸàO|ÁÎÁßðþ[¸‡á1¾Â7xnœÆ˜¦¦9ßA¶iÃÁÆÏL6ñ–Ë"Sj*\ÅMšˆH i-mÅO¥“DIw‰—I’4ÉLÉ‘\É—1R(%R.ÕöWNý?•ÿ3• <Æ*qÊc¼òØSyì¥<öV”Ç>Êc¢òØWyLR“•Çå1UyLSÓ•Ç~Êcå1Cy <T)™Êã`å1KyÌVs”Ç!ÊãPåq˜’˜«$Wó”ÄÊ`¾28R¥ ŽVÇ(ƒc•Áepœ2X¨ )ƒÅÊàxe°D,UË”Á Ê`¹2X¡ NT+•ÁIÊ`•28Yœ¢ V+ƒSÉá°9Ö¦šbÚD;h?£Oè}Fwè1=³YêûÌCá69ÅS"i×îB~bÇEüµ—ðS;®0Kȉ¾f¦“Íl;¦š9vLÿ/¾ÒÿÖßh‡gÚá-í0K;¼¦æj›Á̼†#´šßX-h¬^o¬6Vo4V‹«7_Tò¸±ªÓÊiÉ»Ê׈ð-ž“ÓÞŸmJµ÷hVì}º)5±÷×r›A”iCœØ×pµp¶üÚ3ùÞ‹Ê®Û×uÆîÕÙ¬vEkÎó-íö³ïf¾§y°!C¦‡=óê‹wþB\¼tù—+W¯Á›ç-W.ËÇË…†Ê¥2ù'c±Ÿæ×Ó/Ýdèjek˜lˆ‘;eŠá¼á¥1ÀØU©¤˜eUÌW^ÈÚ(•¤*uA8³æHó0ó1ós±eŠe™e¥åWËk5P5«íÕÎjµ§Ú[í«NRwªÇÕ˪C}ª¾P‹´ršM³k µ&Zs­…ÖRk§…k£´)Úm—ö̪X+Y«XmV»µµ‹µ»5Ü:úКfó““r¶Š¶@[ ›j«k«oëhûÊ6(È/¨|UÝOÊÿòze½š^KÖCô&z }¸¥ÏУõ8=QOÑ7ëÛõLý€~\ÏÖ/è7ôööVö6öûû`û0û¨!ãTMµ¦Æyü<¡žž–žÖžvžmž‡žboÿ +Ÿz‹‚Š¼Å^ß'ÿµd™@V¿^~†`C˜aŒa¢¬7q2#×H-ñÊXÖ¦ÔPæ)9Ê3™äBr³Užˆ"X˜$·áæãæ" X¢$·dËsÔjª¦vTÃÞrë§F©»Õ,õªzK}®¾Ô@«(¹ÕÓiÍJ¹EJn Zòÿpëlífí-¹%”r« ¹U·YÞr‹° ,ᦽƒ[X)·=YßXÊíŒäv]rk^Êm=Rr‹-¹UMö Çìi*¹µò´õtð\öx½á…-K¸iEQ>nÅwej>•vIVÂ÷¥Õõ%]ÑPŸ7æÈV=^À›ã=«Èúá½ôßÔ| + åƒÌs÷ÙK{ sÜè®ä®à.çp›ÜÂíï.ãfn’Û°Áíç†|ßw‚¼™%~FÞëÛiyã +ä³òö4(è‘“7éöØÜÈÜ y™îì»!yñîE¹i¹I®$×jW,€k½ofnU×hW?Ùkèjåjì +vvp¶w¶p6s†:;:ë:mΚÎÊNtòS€4’KAJ¤$J¦–”B©Ø]J£tÊ LÊ¢lÊ¡\jE­© µ¥g¨1‘ÚSu ŽÔ‰:S>u¡®ÔºSõH=¦žÔK ¦¥&mkE½©õ¥~ÔŸ +h ¤A4˜ +©ˆ†ÐPFõ£µn Ãi¤Qô,=G£i ¥q4ž&ÐDšD“éyz¦¨2šJÓh:Í i&½D/Ó+ô*Í¢Ù4ÞU°Þ…5°Ö™µ¶ÞSåð>l€ðl‚Íð!lh®z¢*T¥ª‚{ê¯j«Ú¦¶«ÕßÔ'jæ¨jx¥üuþ&‹/åËÍz\Í×ñõ|#ÿЬ­ücþ‰Ù’Ÿñ|7ÿŠÃ÷ñü0?ùI~†ŸãùU~“ÿÆïóRþÁc(ƒrxP UØŸÁvêïêSõ™zªBªFÕª:UÕðBPµPõÜâŒçÜ÷8b¦b{ÌÃŽØÙ<Ý»aì…}°ãPɽ8ÇãdœŠ/â+8‡'â<\ˆ‹M-Á·q®À•¸ +ßŵ¦”Þà finæ)¸ÿ„Áí¸ÿ»ðK܃_ã·ø½Ù8G±Oò4\DŠhËïŠ&"Îì!Ÿ)û€h!‚"I´©"]dòV"[䊶f+u4ÕßEtãRt=DOÑKô}D_ÑOôb€(‰Á¢P‰!b¨&†‹b¤¹3JíT_üÿûp›+îüïûˆgÅX1Q¼ ¦¨:ÌA‡íD:1N'Îñ;®“è$;)N†“ã´qÚ;œ®NO§¯3À)t†9£œ1ÎçygŠ3M]Ô±º1»È.±Ëì +»Ê®Q½mÙÌ›Ûma‡ÙÒ&Û¶•íØÚn`‡Ûv¤eGÛ ÙuvÃSá©ôTyª=O=!U¬JÔ uRR§ÕêŒúIU?«sð+üwà®:`í´>‡
ÙÝÖ—Ö÷ì¦õ…µËÚ¯ZoXû¬¥¼À,ÐÁfK Rg­ì¶JâCøP>Œ煼Ȯ³ë•e=a·SÀÚ(®<ðg¡µW™Ÿ® +S*Ò9î;%Î k£ußúµÕZË:Y«Y>›ÃÖ°µl›ký‹ÍW¤›ë8¯½Ú§ý: t íê NÔIº—î­ûè¾:Y·Ô):U§ét¡3õ@=HÖ…ºHgél=D÷Óýu f‰Ùb.T_©=êkµW}£þ­¾Uß©}p~€ÃpŽÂ18ÅP'à$œ‚Óp .ø +×à:Ü€›p nÎ;® ±‡p/÷q?º'àDœdˆ€qá} ŽÅqƾØûj÷ã4#:c¤{çäZîxORFZF¢+ѵÌàZUä™àèêd¼Õs5u5t…9[:[8ßp6v¾ælè¬çŒp†:þNtÜsxw·™E^ŽÃŽ}Ž½Ž]†vȱұÉÑÂÑÌñ¦#ÌêqßL.²qí)’þ9þ÷ý=þwüoûß²ì5Âk‘q>_ÀÉ¿ÿ¥Qu4ñ¾ê}ÖšeS@#kqh4£æ9mÈËTÈíŒevùw{¯&F‹ôjããµ_ZyŸð1Ã_\>‰>Û yÎ'ó}Îb™ù*Ÿ»ì¢æsïå[Ö_ÙÿÁó€ÏÁ}ÿ+l6ú-þ²ÛêwÿWóõòuûÉÿ9šJîå“M!nȇ8X k`øB¬‘Í&Áxa$ÁTDpÂ}Xká1<‚'° ÖÃQ8  3tè +Ç¡cpNÀI8w ;œ…3¡܃Yð3œƒóÐ~Lƒh轡/ô~ýa €A0†ÂÃá.Œ€Q0FÃX» ÆÃ8˜! ²a7&b*4¡-ù8“q.„(DBF¸ã\Š)˜Š¥Ð 5Zq.‡xŠ+p%®ÂÕ˜†kp-®Ãõ¸7â&ÜŒ[p+nƒ\¸€±‡ÛqîÄ]ø-zciÜ{Ð}ÑË€®cYôÇïð{,‡åq:þ€{qîÇñ`Ø›±"VŸð be Â`<„‡á<‡L¸6´c†â<ŠÇð8žÀ“x +Oc Ãp¬Šg0Ïâ9“Ï¥ƒ|!_JGé$Qæyò•t–.ÒUºIwé!=¥—DKoé#}¥Ÿô—2PÉ`"Ce˜ —2RFÉh#ceœŒ— 2Q¾–I#“eŠ|#SešÄJœL׺†®©kéÚºŽ®«ëé»ú¥=:[×× tC™!ñ2Sd–Ì–92WæI¢$É|I–²PÉRoåmò6C0ÞÇøøãÌÁ§˜‹Ïð9ÖÄ<ÌÇ,ÄZøBB¥”I™•E‘b%ª”òÂÚJ+«òV¥•òU~ªŒ*«ü±Ž*§Êc]¬§TUQURª² +RÁʦìjº +Q¡X¨*ØP…©pUUUSÕU„ª¡jJŠ¤Êo4×Ut×àfŸÿöü{ÿ…àÜ–"-ÒâîîîîîîîîîR „Üݵ”RŠ[)m‘à.7kÝuŸgÍË<Í·ŠWóãµ¼ŽÃ9‚×s$Gq4oàÞȱ¼‰7óÞÊÛx;ï་wóÞËûx?àƒ|ˆó>ÊÇ$¯ä“«rM®Ëcy"qò”SVÊFÙ)å¤\”›òP^ÊGù©Ÿ ‚Tˆ +S*JŨ8• ’TŠJS*Kå¨5 †ÔˆSjJͨ9µ –ÔŠZSjKí¨=u ŽÔ‰:SêJݨ;õ žÔ‹zSêKý¨?  4ˆOÒ`BCi §4’FÑhCciŸ¢ñ4Oó>Ëçø<ÿÂøW¾È¿ñ%þ/ó|…¯ò5¾Î7ø&ßâÛ|‡ïò=þ“ïó_ü€ÿæù!?âÇü„ãø)?ãçA)~Á/ù¿æ7ü–ßñ{šH“l"Êø£Ml“ð'þl“Úd6¹MÁ_ðB@6¥M À|4þ]c¿×Èw6 |„HdÓÚt6½ý¡HŒ$HŠdHŽH‰Tø¾Á·Hïi‘éñ}`‘‘ ™ñ#~Bdµ?Ø È†ìȜȅÜȃ¼È‡ü(€‚(„Â6£Í„"(Šb(Ž(‰R(2(‹r(/Ïä9* bà‚ü@P •QUQ ÕQ5Q µQuQõÑ  ‚àë !¡1š )š¡9Z %Z¡5Ú -Ú¡=: #:É tFtE7tGôD/ôFôE?ô§i4fÐLšE³iÍ¥y4_^ÒZH‹h1-¡¥´Œ–Ó + WòZÞÈ[z$k%LÖI¸DÈz‰”(›GÞÑŠ3£Ì3ÎL0“Ì43ÃÌ1óÌ"³,^~a&ÂDšhcbÍV³Óì5ÍQsÒœ¥§æ‚¹d®˜掹oþ1MœyFÏè9½ —ôŠ^Ózk ÚB¶°Dˉ‘÷òA>Ê'ù,_轧ô‘>ÑgúäB cà#>ásÈ›!·Bn‡Ü ¹‹/þW~ˆO¾ñ=ßúÎOà³ß÷ÅW?ð¿öú‰üP?±ŸÄOê­ñ¼µÞ:/Ü‹ðÖ{‘^”ímðb¼^¬·ÉÛìmñ¶zÛ¼íÞo§·ËÛííñözû¼ýÞï wÈ;ìñŽzǼãÞ ï¤wÊ;íñÎzçþËt•@×x¦áçý–{Õïï±”8­‘DDBB"c)¥hŠK$b›¶–*¦vª¥´iµœ£Zã(­N-EµÂ% ¢š‰ÄZµî?Oœ™sæ>ç¿ç¿÷_¾ç{¿÷}žï5ù¦Àšæ¤)2§L±9mΘsÖülΙ_L©)3åæ¼¹`.š +Si.™*Sm.›+檹f®›æWsÓüfn™Û掹+¥ÂÜ3ÿ2÷Íïæßæ“çä;N¡sÂ9é9§œbç´sÆ)QÕ겺¢®:9øÛÕ2_Ž´ÃìÄ?¥ßà[ta!öc ( z€¨“u’äœFŽ¼/ËÃ:UÑCõ0¢×Ôöà®T9â(éèhǨl3{61Žß©ëÔ«s¼N^ü:X‡ë8€ÍÈ’nX!=d–¬”,ù›ÌÆ.Épžp›¸p›ºÁnˆê6sÃÜæn¸ÛÂp#Ý·Û×Mt[º­Ü(7Úmí¶qcÜX7Éä&»)î`·­禺ýÜþîw o†o¦o¶:äìvö8ß;{œ}N¶³_V?©\uDUÇÔq•§òU*T'ÔIUªÊT¹:¯.¨‹ªBUªKªŠÑ•žbÛT¬Ct¨nFgxÕŽµ¯Ñ-Ú$;ˆ^1ÒŽ²£é‰¶ŸíOÅ?hsì!ªþ1{ÜæÑA¦Ûv&½d²b§ê©[êVô”·l†K?ù ]e ]%“.3_GéhzËJÝZ·Ñ1:V·Õqúº½âŽ½kïÑ7®ÙëöÝ¢.ý¢~͘t‹¦¾t:ÆDß$}U_áqîЃþO¿)·çízHK:I$ÊöòÅúÚÒYÂé*­é%Ïùºøºêöºƒ¾­ïØå€í‡`Mô*<xçyTð¨ôõÚ‰ ¤{庀ü÷±¡9nJ[®r6úâô@V¡7òñ%žÄ9 +ƒ0¼ˆ­—`(ôBXæE ^Â4T¢‘HD©Ôç{zb +¡“w™ß‰Xêíæ]µ/°G&I +bxž ¢%Š#¯ð²„Hï¸w†¿>D¥4÷¾FÏ.¡"0+Qé8â=$Óæƒ-’!—ŠQÈ4íÌ2o"žcœ’DžõÇ{托ħ6Id{e^~4‚×ø¦…XJÆÛ‘­Úèxû1BÐ]1£yõ-”Hi«»{Þ Þ:þ»·T”:¤ýä…>‰÷±‘Ñ(F«Æ‘öò¡l# +å†=Cn‰˜‰7±€Ì?᳟c7{Œ¶ì@‚­ ´D*¯­` mg­H¢ —lÙ¯7ÛØ@7ï)¯¡Wåyh…adøk¸w$–÷pÝLÏ0MÍ ÷èmÎp,Ö£…äQʸßÅ}iEœ¯Ù¯{C½­^%¹ÔB0žÅ ŒÀdÌÂlü«zñ›<`ß4Oå›fñM/‹±mÈ} ïNá»3¹JÛ±‹(æ,ëIgñ¬ d/+dµì’)aÊÞáŠþJÕçLk½Î|S#4å¸aŠ \yŒvç»9È•†ÒBZsFÅ|þ{œ‰M¬áRî-W˜‡vq Æ6Î{rq +¿àWJÎiýÏ̺we9±N>—ý’#¹ô¢{5P͈HöpÝØ«õbŸõ.±ŠêT¬ªuýªž¯ôN]b`Œñl‘@ÕØâ;êô'øÇÔ:öðú£V†?* ðtàOÕý*oˆ7‡üÃÑmÈt Y®cn&>c&îÄ!ÃéÇ\o‰ËŒo,ăh®Z7é-}ˆþ2ˆH%†Êb´Œ‘ Ä|Y e‘¼Coøà1Örn›åSÙI|'{ˆSôÜKrEn)&±ÒÌæpÁÞµgÏŽt ûÏdÎxòãnr»Èyj‹úFíVź§ŽÖSÙ/|¡è"ý»Q&ÚĘ.fˆoÑE é›l°íi'p~À÷ wÈ©Ôõ¾/}Õ¾‡~Ÿ?É?ÆŸá/ò{µÂ©V‡9ïøÿOŒ/_¦Û§ÌªŒuÑXO±K$•ó©Áz’^®OØqrS‡ÈYY¦ÓôDo“î¥îëÉ2Dí“fì:ëqxžl£ÜQU¦¡ V—%Ò¬”ïÔd¯|5ƒØ“¦¡Yd«uÕ\ÉV9ìŠy? ³Ý evƒ*Dˆ)W Pƪ^¢Öð¡<•¦21Ì´³Ƹjß`¼ŸWK¥•.2P©ÃÔm¹)«©Ç¥¯i®^Qd÷‘4Åu™Š)òºË÷Üûì‚ÈV½Eú©:\­¯”+8®C¥H×ÆðŽÒB5”$uS¥ê½¾Ý^„*qoŠ–XæÎÿ>¼Î +X¥"¨i=©&'%±†z'°·F±í›É<Û¨£‘ŒX¼¬Ž¢3k£’†ÅˆÃæàRĪµÈðÈXê~ê§âN 1âP-ƒÈm>ý¢‘jF-ÉQïSÿPõåfK++‘¦æÊ{¦'•iõ7“‹—ùk=²|;ìI ” À„60ËÏázÎEŽÿ4ºßl4ÑdBežÊ'ÖÐXŒ£¢0—œŸg'™*ïj/3L£Gõ£'æ"Í[ƒx®]²·ÈËÄHo£÷Æ#ÅÛJýåmG,±ÃÕeÚQcså ýègÉüãÕÜÆQÅwï¤û+;'ɲelŸ|‘,[RtJlײÕèƶ„'ÁN\8™¸Èÿ2Ο&†LJ’Ž?&¹„!´¥¤Liù@pÈ8Ù)( ÓúC3 0úÀ _h™ t fú! K¼;Ù® 3n¥Ý÷~ïí¾·oßîíÁ¹=€þçQ{Ñß¡üüßm¿tÛáìL—/—ï"Ä£"4oÑ{èiôOˆÛ¹Œv•ö…r–œ…7ÔŸÑpy¡Ü„94S>'ï/Ñ5ÚgÏŸz«üòe$}D“FZ”sãýþB Ò<³T¯Jõ[%±hApVÂZ¨Þ¶F8ª6Ó2‹²ÔMjðÀF\±é‘ü¤ƒ!MJà‰&ÃœºÌjº é“] OC/c +ÖãˆÁöåu¡pÁìo؃‚,é#XyåíÈøB…‘IšY²‘h _§HÄhk3„îƒw[|G,:W$ yV ð¡!ˆíx®;ÁÌå½TTÑ0Æü°Vá%4!."5ÉDÞ”,¯KxPÕ¤Œž_‹íàÈ®"ïÚ­Q¸"€€¶ Dê RïÀ¨fð³³ræH~¶øh¸û4R$rŠIk(ÈßC#›Œæ0Dz)+ÿ§Š4 l!XÊB~ Rç¸@àÿìT,dö²šO»­ÍÉèŽlå{¶ð[Üsè$8l ƒ#£ºÎm‘eá°Òõ¬,eõ¼>^,ÏOÈ’ ë·HÔôÙL~}ù‹å7.‰För&1ƒ»!µ Ô[ñ…á‚Š/Õn IF´E}ùÞ\a;È´[Bª…&j‚&#™ ¼ñ`W,Œ¥/ÞRš·¤6 °øÉ"FƬcM‰ +&¬c`¶ +¦Z˜ù˜'E߈¶9¬•‹!ûáââ·#D•z_À·‰7á*Lo-"»­H¼y“Dm¯cTÏPö·@N ·"ÃO!oDxZMíî§ö­¦PháT %à 8ƒPa¿ =’ÈåGª} eð|¤´‡8¢nÔ­Ê/9œÄsŽ‹N‚»Ê:ÑUøÄBˆc¯W7Q˜š¯yÊ42¶²šJ `a%½’€+ÞPKˆèÐcŠ"<5uqæ{ÓW^Á;œ~uÀ·çlédpïáï`ý÷¸—O´õXzéÎ~¦/||Ø>|Áò!©noµµ1vŒ;Á 7ܤX(…R)’š÷h?ú_'𘻣¶®ÖåÝÑÙéêhoÙAì¸:ýíWJ¿û×é×öêÏاÚ?_úúÝÒoJøD0ó|ìÎ]Cÿ±éX~Ÿ>kß‹’dc0—èç,îj Õ8‹ä_Ôm®$Ü¢ý¬"ÚxÁ3(O»ê’é´°ú.<Ë8žPDµž¥ªhñ4Ç)T’vU{ÝIüÅbyy‰aÛ¡7[?´ê@t²ñ=lΦ± ,¢"L”;Âî°¯Ul ·$:©¤¯]ùÕOòâ¥Ñ“ã4‡æÓ”‘ÄjŠ>ÎÏøfÄc»ælsÔ=ÇâO;NûN‰gý§¤¯ÅÏÙ.3ºÿBü‚r1ñ<ý2ÿ‚ûï˾«â‹áïÆ_T®37Øü ßuñ'þ ñ%z‰ùWôÝT~¥`þÆàÅ×}ü„„|øß*¯:†ï:HÇBwO‘8º8¦{#Hã{õÂ=¯°zÿþØW"‘Èê_!”tÊj„ÔyûŽÈùê³oCëÝD0kDõYáíHÄ‹„,,vM Õ) +-¤ræ–Y{à‰`(8üÕšp¸¶*ÝhV’U“×ŧ-…Hå°§™‚Ï<øj…lúÜðÈÍìyìni UÊ&Ãå×Bè³½5‘ÒsáÒoKïl/=sÔdzðoGWóï‡%¯Ê]_ïn%„í]í1lÃD´¡6ô¸}o0Ô.ó“Ûää£Ø?[ +ƒJ³üì*Mœÿê—v†ÜU.†è?tWkp×¾»«ÇîZ²v%ÙÞ]Yò®´¶°…ü”lɼۜwÏÑÞ’_£b`bš)]ŸZ’Ûµe¹]9ËÔfÒ×mÍfk·µß¬ønhx …‡C{y°éNÓ ÝšÄRò„Ä}ÞPPž”8)/ЗX–'Л¾êóùu¿­9©¼FcFgwfRã¶hܘ6®Mjǵ[š[;“8óS&šv‘j`vrÀ_qKS–ÄÀ–„¨>ÊM€ëíÕUaO"ÞÍt†þÍkFëä2¯›V­jjêYõ¬ÞÖ·Øßß‘¼1£vE%v¦ =MM«­s$"=ÃÜέ4õ€=†¨¥„¢ð*1ù‹Ëµ›DŒ 'ü6U0^¸~Õt/ÝD{[‚yV³-Ñ£M“¥ªŒ¸…qájú,Ã!ÚD5èáY„ç<&sL:O!õà?Ȫ•–¹‚2W • Ü*oÓ>¯ úLéƒ3Ù8½«;«²a=l$¤¸l©fÐÖLÝ4º¥¼ÜÌkY½Û7Hò 6¨o0¾,,½n¼9ÿyCü™tB?Rí-ñ¬tN>§Më¿2.Dfãïkwä;Ú=#}LââŒ)cfSmek,ÛuëÊ6™,ÛD¢lU•YÇÑk3ø>2ÁMðãî}æA÷!õå¸Ô-f䌖¼ã™µ>0¼/È/jÏëBWp½Æ‡´p,D"fŒe5œ)=‡bÍÐMM×[%9Œš-b¶$½·Ë%ò1. rP˜¡Wh3\Ô >)sŠlcW>'ÿAvËÏH‘îKŽâxZŽ‹çÅwEA|FÒ¿a\@Õg ù‚‰æ­G™=Óž¥fÚ—%Ò¬ÄK3ÜoÏ)qî@¼|7ЋÚsPÆzóE-¥+©Ôž‰ùQÊNcAûDWæGµy£Hí„Vd\fEW‘SkòÏ7³âêA•õ?Š¬rµ'µô£û;ˆ?Bê¬l¢zgJŸNÃJv×µ3¡¼L‹ 9”M|L,„ÕWÛ·‡–ë¦PuÓRyµTp5ÐÒJå~Y›l¬zÿJXÏp©L8Q»x¡qñ|õŠ:µ]xµ¾ÁL´.zx.Z)*êë]jlíý›‚»³E‘D¨™¾Â¡¿ozûÉ@ÏmQ ýø0ú6µ¿;}gòÞÓ ?TˆX‰S ýéÀk-’Çrïé»TXä៫'O¿_éKd 'ÉGÂ9ä"¤øšç$Y}h#·‡<Š¶Ç€fÄ_q}ŸÔ£ÿ×q¾ö>Oć€[ÀJ`+`OÛ€G€}À£è{ +8LçX†ðy»ƒìt_&Š{„Ä!ø ׇ¤Éµ—Xð×Ós\¯Cˆ’&øq´5z£è{¹ô1mG¿8ë7‚q{É´¯Áyô¾D"° „¸yÞ 9Ãn.Òµ–þ +ÿ›Ècü{°k‘ëì#ˆo†¿ðcLŸ/}¾ +5î +ß bÜ?éô÷#Ç]h㜧}q]?l„öÅœÂU.½F~"\%§]‘0[÷eRI×M×¼¼&š?Íéÿ`-Íïa”óc ¹òrû/ðÿ/ +ìY\ZëQ¾@Æ…ã¥Ïà'˜·ýcÈãYÊÊMÊÊ Æÿq_îÁ]W?÷îÞû ”˜bµ üCÀ„PbcLH!H°B@ +cfÚ Šåáh[‘:´2UjÛÀŒK™Pa”‡”>Œ:”Ñ6ÐC‹ÒJm)N)“l¿gïÝŸ—MBÒú›ù̹û>¿Ý³çœ…=i›ås\€½a îÌݹ´¤ƒ¡H{ž‚¾ú¾ð9²Î|xn¶-¶#ÑžÚý)þŸlS™å Õkë;ȶ‘ylû,E…þyn•±Í{m¤Ö¡Šï#ß #>|?õuÔ÷ŽÏÝH³ ¹Žâh»Ç;Dãäpš*öÀþgà{äØŸ}?’? ãî2¤ÄM4gÉww%ë™X‹ó5Ìׄ½Ì‘Í´FËw°lq`¾£ÒÆi +ÚX2ѶžÖ_ îAo3ÍÅ÷?¼¥d ­æû§Sn5õ¯‚'Á¤|§>©ÎÙ›B©°›³àaYAwzt‡l¢QòF}ï⨟‚¹ d•bœpšh¥˜BüÍxù¶à±–{žbx~Èù ;²m®£-iiìµÉw ÙH}§JÔ_ô½*QÕw²Dµ’îâØÀþYÇÒ¾¹¯±×„]¾@9â\Ä>-;Øg)Æ¥ÚviË0¶$›{Š1ŸåXÃÿ_ûÇZ}Ÿ´ŸCÛ«¦¿-ãi—Û¨Þ×~¸™¦›{ †ƒ8Úß +ýü0ΛcÇ*5ÃTÍãÕ üÏþrÈ3j››«^IÄÔ8…¾,ÃÄRÞ'¯™$âhœ&†þ,ÎñTnB âh??OÒÍÞíÛŠ´¾|ùÀïå"ŽŸW—d=$V ÜK®‡ÜÇm2‰nGásÇÓBÑ ˆç´ªíô€ÈÇÆXìÙÍžK¼JªÁÒóqH®cý} ûd_P2ÎÊøe>{ÿ%ƒ\ï4ã?ǽFý_ãÚ×S6qsÅò)Mº”ö‰ë1ó/èý€ŒìE›ËyNÿKÚfSô˜êRR•0ÞF*Æúq½V5Ý™TB9^­:­óŠ4š öR¡¨¦[ð¡í~9bTâe5â#Ç@;l35(ëX­¥º¨ãýÏûx4UçÜæÓ@?ngdÚfÓ0±ó< »º„ï-JéüàÏxVamÔ óÎ\}_~qoÓ0¾c¬ƒŽ7¬ÏZØÛ~º…cblö°7%#ä<’óÆ"†2§Žß‹ðlX7 Î ÷Õr›;™ŽàÊl%RuœŠ÷h–x ç·•‰éˆß{KÃÇc¯~GÓÄ>|F}XŒÜo!¥Èšƒ§á·mó1®sl@;ó4ƆÜBeâúªhB~ðç4H.‚œ *iŒóSªs/R_Œ˜\ª^Ðó3 Õ—57?džh] éür»NôÕºFõd;Ñçàyõ8ô‘’R°O‡A<í÷¹«h3X羇¾÷ÒcÎ&µû:Ö¢:Z–#o‚ÛåHz ,Å÷PÈ_­A™Ö‚÷Á2Ìýä6OÆ {†D]¨¿1mQxÎê£xýÕîËÊ;k€sVífìþr)c½bY¦v3âCÄà/¡ôØbJ¹¨ˆqVÙë?·ƒ²© Ýét%ð+ŒìcÅÕüÇ«…ï.Ççë5ßÕ‚ó]fjNÃk¢œƒê0d­sq{|)@yÊýÌ~šsBý÷u½u~°â=·ëí²}®Ý•Ým4+Š±ƒ„=¬¦rFŽB`—“Þ¦rÆ߃¶=Ëò'Ý09ÊZÖ 6˜Û±ìO¤\Æ͆®<w$ÊûáW÷Õã“/ß]ÆÝŽX í#áóAd_‹y_ÅÚ Ýœ9û| _…ܦ#ŸÝG…÷C~ÑÈèµmÚ®3¾¤³>ÖÝ(ìjÎÿ'pwÞ{Á¯ÿ×k9[©@稥TåDÎYKˆ©m¿%jM‡ì‡¸€›×Š¸Úö.¾ùø~ uõ+ ájZÛQ¯GdƒÌ@þN´`ŽöùÁضóàÑ`Ž¶Ÿ}ü§…ÁøÖgη ™Yëv° l•cæyåG ßBy\0W+¾ÛŽ‚å ü(­ßÜÞ kü‘ó‘NÞ¡×Uvõþ¸ZiÞFvxCôD–^•¼ì­aο;iÞH½¡þ~DŸ+¾qŒ„ýôŠ‚\:‹sJΣ9—õ?sþ˜ün«Ö²_8‘)9wæüÕœ9xçåGÞƒU&nD}«s–@*èÊ:ô¹ˆ·Î>øžøÔsø/1(ßÀq ¨ýøNA¬{û@6£œ yÎÄ4ã[;øØnbÚõ.÷4F^CL- +™eÑU½á !w3v,î)ÝÅîkŽå]Äèhœþ´eç ½Ê©ˆ‰U¨ÝŒ—vȺ)w—çö´lç=.[y‰)Ûth·mÏä3”‘Àºw=…ßrÇ'¹¿ÑÁ¾Ç‰û–±GUQànCÌÊà/ÿ«L€7®Zº'’Z©(ée~÷ª`'êþ9‡Û _tVÁ¹Wm(åTÙ¬ûN ™Ó=ÛvËù¹Î±gÚ>ËúS(ià0Ïœ5¿=±öI÷DüΕÓÕ9¹X9`·r$=^F9eøjõ±/ë)~ù‡¡$øùñ |öXöõþ2ݧm•bîÅ1*.M‘ Ô<öé ÍÏ£d7¦Ú៳PŒ¾}‹†‰£ô9שÇÃXU›‹ùW ŒÀ¼¤.ÊXw=$26Ñ­nIŒMÇ:Êr˟ñGi=ú .úåÉI”F±®`Ú†‚!âú¼˜±˜ßYKëÝQ´Þi§ÞÐïxoèØk2eÅVQ%’¨¼ØÌóu*K:®N 7;á—PŸ0^é¸Ê1Ñ|Ç2Õß±7w‡±ŒBYfþ³°~7È-Q_‰®kÆÅÖ –~ƒcNDcyW¹Û¨`®õA¬WírZØU#b®‘V¬Ç>?‰}žÃ{ª÷v)Õˆ<š¬c:ÇjŽÙ‡{lëbÖ‚MžºB.¤sô—²gV .°¡<†ÏŠmIÛÓ*ÄHI÷ˆ4TÈíT!¾KUøŸ…‰>ë  ö}‰uäƒaûrs)Y neŒ|‘Êp†½Bú +µ.—`7¬[oPI5rž^ç?Ÿ@}xÏÀM\'áÓìëÈo©ç!ñÞéýã=CsÅ›Áù'ëµþM’÷N\8p˜ÚéÌðnU‹½TÈÿWÿGäT8Ó%Ð÷¬˜ìîë×Q¥ÿ&8€=Yÿ¿…Ò½á”îßKårüçÇA&ê!]MAŽS®þàü’q bîÖl’N­tOFúxì§c€|ŠrAÜϸΠ´Ÿ‹ÃïÌàu%´SÎ6F@?õ‘¸ç5 kOÆüÛ ã$|c‘ +»°À˜C8/¿‰íFN…ºœ16˲Àõ,ã6a}† êYŽ¶AýèNôèª_WztUŸcƒúœë GWófÙ >ë +úÕØ ¾¦ztµÏÙ6¨Ï¾‚lP?ÁÖþéoàu¼KOC>¨Sü¶Eti?o¼/ÔÜ°|$ì÷_öË>¶«êŒãÏ}ÿµô +T^¤-BŠ­Â¥u–ò6JKA…i!ò"2(Äl2ˆ/A¶éŒvdaL™2cfè$qÃ`6‡dÉ6£aº1@øÝ}ŸçœûëíÒÆà[rÚ|~ÏsÎ=o÷Üsžó=ëZà¿p¸GÕ QwÞð4À]$¬k!y|_Õ‰ú ׂyZ+Uu“¿V}ËøtŸR7ëÁ´t>اú“¾yüûaûf]æUÝïjÜÉgaUå¯|¨ÞQê½ÑBè€zw%ÎV!îVÑ@>Ã.–3ñ6côKè“{Qï.*^§Þ"*Er_#JÔ` »©4LyÞtCÆ.œßCÓÛ”s“¼‘oÓÑ;ùR…ûKÙÄ›Ð;8oü^T‚8=6ãšàc옳¯¦úÖZËÞMÈßèu.|æóxY£9oÂ.g½^ô²‘_H…Ï`Œ§Ú*vþN…þ×q~l¤,ÿ(öóe*KŒ¦b*•áùröè>Y8kPn6ʇþ^r}ÌC€9l¤ÌȲވæ€û@ŸeÞbÊs’¢YúñØR6j£tMgŒuL×5‘ŽŠi +ÖI“¢>¢÷‹ó3zÿ˜m­7&Ð(§‰z{;OXG¥[=&ÿ{³ñý´žõ—€°ˆæ{?¥wÎò­ÔTCÓº”Åú g¬ôÇg´÷ tþ1Ê·aMþ¶Ú#á^0Ì+ÿ'°[ÇŽz•/{yWšuþ`X¨žó³pò¯œUí˳Uªü•=ðŸ$²lÖ£ŠäßáÓ o\§bny}|Þ†Méz~ÿŽlºþ¼¦ÅÆ©éa¥';°¨þCéYÑ©‘ŽneÑÏ ÑvbÃSÚ~¬ósy­q¬H·-ºúZöšúUi`½ÏRû-]_§Ù”¾nÛξJ·²¸ÓétºnoÇæéy‹»EoÖ ‘¹¬•c¶k«ûSºåo22 S:k ãÉw‡ÐíÁëŽñ·Û@ëû«ð.#†‚ ²5¸3tj'&“(j¹áÏ5¡æ8ƒJŒç´|û6ˆÞ'¸ )W8çÃKí!c-iïíá#0Á皦ÖDóÍc4/Ñ{GãúÚ½Þïx½ßåËzïöÆ{òð¾¶ùL[ãæ5èwÿÍÂûy€&kæx|¦9&pÜÂsçwX'°îbu®ZIº[ˆ¾ ö¢h$Dò`ú|’ës,”x¸²Íù9‚ñÝ + èü¡¨ó©ÜX{}àžUç:žÄ1è nâØ’ ÙãåîAš¯õÞa­ý^Å>/e½„ç¹*Þщ¹ˆö÷£BÜ ÏàÞ¶“ÖkÞÕlÒÚoŠ¦;êí…}1ŽS}V&õ+ÐßÃ`›ÖÛýt$¡òSc;Œ±ô—ì‘ç Ð ÎkTêÇŠ³8?Ð è·Òn¤žnÒOC[)ýQ&{á8ÊNGÉ`ºhŠÛœÕ-{ÛÙAeΞ0 ‰ÜI(Ÿ -X›TœÍæ8É}á]ƺµ”ëL€þâ3 +ýpnò ‹œÙX¯S°.zཙ¥x÷ó +g%XKÖsà"üÈ¿€ù­‡ÿ2x@Ú{ÁÏà‡ývÊ@ÛåH3!¯ö;à» [aUØ÷ÃÖ¢/çl5˜²´¢êYë`·‚eºÜLòì§Àhø…°ƒa_£Éãö¬?êò3ceîm)ãA³gΧqØSã‚YX—Õá~ëUº3© ¾i¶º?$ª{‹è(¾cÔ‚-HÿÞ~™Œe¢Ðîwm=ŸݵTãžÃ½ï'4ÌÝF¹Þ(œ«§©Æ»…úºQ1Ú¹Lb°~þ‰ïVãL§Àzc‰áßEÝ3ÞB …ÊÂþ ÈÚ»¬5]å‰Ïêm—Rd¼Ï"ëç“íŽ,íÔ™Ÿ¡Î#¬ODcãÌ—óu,Ý£5ÜhŒ‹ï¼c½d¢Îx½Çã}Šy]i¸ì±—ß!GÚ½ÂýöT¾+HÝYêN®V÷Ûp"ÚÝìí£ +ÆúW¸‘‰¥÷3_vÚ}÷‡á  +~ÕÕi|Ë¡šVßÕ_O·1î×PŽ™ =ÙÌuÕwî(í×ÒÆî>zµ‘^CÝ‚•¸WrÝ‚ŽÓö+ˆû@ÖÚ€«Óx§±Lê½;Jgã[‚h­¥ÖóµÞŸÂ$îî¤nÃE«ñ·ß`°Ž +££µZ•½ûõ-ªöûR7ľ[ÔÙXɱk.â 4¿n¯Ö=$±¼ ÇôXÛY·Êúœ–KƒN”퇳QîI¬ñµÖ¸}Ž³²ŸÇý[ƒïhˆEžÄf²ŽA»˜p‡Ýyë%u·„­zY?Àn¨Õ1ªïò#¶†Û%¾lÐ1ê”A\´~®×±ªgRÝ ¦é84–YŠÀ@Þ#É Šp9ì9—ÊuœävëQ>ë}Þ·8k*ybN¦u¤•pþÑš âˆÖ b;Ò„±zçÚ*µ>Í=ŠuR€ûk~ïÝݹðÍJø¼ö¶H¬“º‹h/ßw=¶rŽoÃ:FLI¿¸6MÇy6Ú[Ly|naž÷b¶QÁ1[ÑAd2ÎRn[ßÁJa3Y7ð8ô½!'vß‹îqrÏpQµ7Ï2p^n¡±hw¨½É‹*6&·ó:󞧛YË°ÕzánØO`sa?â{/ìðoø9Ê¿òŽ¾ÃIÝ…ötFr½÷6òá®t™zû›ø¾ƒ5ñÝl}‹êÔyŽÁ\~(ÿ•*ÞÓÊ/À%¬ëUÿŸxõ×OâQ¢N#¾89ó ƒÁ`0 ƒÁ`0 ƒÁ`0 ƒÁ`0 ƒÁ`0 ƒÁ`0 ÃÿQö ú”*éy +ȦÎt+Í ò‹ÝOÈCš¨Â~¿ñßò˾E’bߦZ§}‡–ÑFí»±2õ ÷µïÓMVBûµJ´Ÿ r´¤ü Zk]Ð~¶Ýl¯æ‘Êßp¯‡ö-Êõ&kߦÀû†öéݧ}7VÆ£,ïqíûÔÅÛ¬ý€x/i?A=¼BígPßMûÙÖdÿN´l¹úÊ +¶‹Ï3Ô9Ø+¾/ù¿_f2xGü„ø'ÅÏÐs¨|5‡ÊWs¨|5‡ÊwceÔ*_Í¡òÕ*_Í¡òÕ*_Í!û™±ñw’±?+–Ÿ#þeñ;óت¯®ðóùâw‹•ïÎí$ŠŸËï)uGˆß[ÊŒ¿O¬LaÌï/ågˆ_"þ\ñKÅ_Æ~"6þD¬¯¬X~Vô./R ÅŒ”Óx t?̓LM´¬ oÓRÉ©Aj|þƒü…R¢ Oªé›ø/¢zä-@ý´\Ró`ç¡ôJüÎEÉ<_,¹E4ö!)Õ„¼9h‰Ë/ ÑÒÔIï¿¢ƒÚEiõ+°C¹ïåzœE4#(Ço Dë é>ͲŠƒ»ÑÜB«ŽSײó2¯cXÑbë2S2æ‘ fÏŽ+%á€ÁRá —EX­MuâÛyÝž£Q#nA4× •ªI‡’©*Ù%¬» ?–𭈫69vnì„Šô¶qÍn™}¢ûåµOoÕN=dÎðâ¹âS9’*…©¦6NíšÚ9µ>ÓCŠªµË7;:yÇä`¦«P ΙöjéÝG!£Ÿõów µ-ã£Ê;P±TÍ’ªê®œ*¹STµ&§ËR^u\Õ㥒T–³å9Wª¨¬*çK5¹@ª«2¼H÷v ¹XjJ-¹DjË¥z*Õ•zr™ÞR ô,n¨ª ±ÞDMårUÍ¥…ê tÕ­Uï·‘¶r¥´ÓŒ;èýÜQo–«¥³t‘käZéªwLw­³§êƒ^ªúèýÙWo½ëå=Ûo”›äf å¤UfêùœEYF!9ºvƒ•©<ÃNL†¥â+'ùzÞ¦'ãP&Ãe„Œ”Ûå¹SFÉh#ceœÜ¥ø¹Wî“ûe¼L‰2Iå!yX‘GAy\&ˤÈTý½2 ©ò´L—gäY™!3e–ÌÆIªùæÊó2_ÈBY$‹åyQ^’%²T–ÉrY!+åeyådµ¬Áɪ¡×Ê:y]Þõ²A6Ê&ySÞ’·åyW6ËÙ*ï˲M¶ËÕÞáTùDvÊ.Ù-ŸÊgÚ_ÈÙ+ûd¿ƒÚ±‡åKùJ¾–oä[ù§átœ!Gä¨ü$?Ë/ò«ü&¿Ëò§ü%CPg¢ÎBETBeœsp.ªàÜñ˜€‰˜„ð ÂÃxâ1<ŽÉ˜‚'0ObžÂÓ˜Žgð,f`&fa6žÃÌÅó˜‡ùX€…ºû㼈—°K± ˱+ñ2^Á*¬Æ¼Š×°ëð:ÞÀzlÀFl›x o㼋ÍØ‚­xïãlÃvìÀ‡øãìÄ.ìƧø Ÿã ìÁ^ìÃ~ÀAÂa|‰¯ð5¾Á·øßãüˆ#8ŠŸð3~Á¯ø ¿ãü‰¿ð7… ™ÂTžÄr<™§ðTžÆÓyËóLVàY¬ÈJ¬Ì³yÏežÇª<_žb5^Àê¼Ñb ^Ìš¬ÅKX›—²ëÊc¬ÇËXŸ ˜Æ†lÄÆl¦¼œÍØœ-Ø’élÅÖ¼‚mØ–W²Û³¯bGvâÕìÌ.¼†×²+»±;{°'¯c/öff°/ûñzÞÀþ¼‘7ñfà@ÞÂA 1“aFh3‹ÙÌ¡Ã[9˜¹Ìc”.cBqúL0Ÿ¼…ÊaÎÉÛyïä(ŽæŽå8ÞÅ»yïå}¼Ÿã99‰ðA>ćùåcòçdýÍú§òINãS|šÓù Ÿå Îä,ÎæsœÃ¹|žó8Ÿ ¸‹¸˜/ðE¾Ä%\Êe\Î\É—ù +Wq5×ðU¾Æµ\Ç×ù×s7rßä[|›ïð]nænå{|Ÿp·s?äGü˜Ÿp'wq7?ågüœ_p÷r÷óòóK~ůù ¿åwüž?ðGáÑ8¬m9 +€I_Ķ&6'ÖĶmÛ¶mÛ¶mÛ¶m;ûÎ~EUàKàkà[à{àGàgàWàwàOà/‚!8Á…Ѓ°‡ðˆ€ˆˆ„Ȉ‚¨ˆ†èˆ˜ˆ…؈ƒ¸ˆ‡ ‰IÉ)©iéð/2"2# ²"²#r"rã?äA^äC~@ABaAQCq”@I”Bi”AY”CyT@ETBeTAUTCuÔ@MÔBmÔA]ÔC}4@C4Bc4AS4 ¶ÍÑ-Ñ +­ÑmÑíÑÑ Ñ]Ñ ÝÑ=Ñ ½Ñ}Ñý11ƒ1C1 Ã1#1 +£1c1ã11 “1S1 Óƒ”Ÿ‰Y˜9˜‹y˜XˆEXŒ%XŠeXŽX‰UX5X‹uA#؀؄ÍØ‚­ØöÿìÄ® #ì šÂ>ìÇÄ!ÆÅ1Ç œÄ)œÆœÅ9œÇ\Ä%\Æ\Å5\Ç ÜÄ-ÜÆÜÅ=ÜÇ<Ä#<Æ<Å3<Ç ¼Ä+¼Æ¼Å;¼Ç|Ä'|Æ|Å7|ÇüÄ/üÆüe0g€ )š!’¡ša–áž‘‘™Q•Ñ1“±›q—ñ‚FŸ ˜‰˜˜I˜”ɘœ)˜’©˜ši˜–阞‚Ö’‘™˜™Y˜•Ù˜9˜“¹˜;è1y˜—ù˜ŸX…X˜E‚vSŒÅY"h8¥XšeX–åXžX‘•X™UX•ÕX5X“µX›uX—õXŸ ØؘMØ”ÍØœ-Ø’­ØšmØ–í؞ؙؑ]Ø•ÝØ=Ø“½Ø›}Ø—ýØŸ8ƒ8˜C8”Ã8œ#8’£8šc8–ã8ž8‘“8™S8•Ó838“³8›s8—ó8Ÿ ¸‹¸˜K¸”˸œ+¸’«¸šk¸–븞¸‘›¸™[¸•Û¸;¸“»¸›{¸—û¸Ÿx‡x˜Gx”Çxœ'x’§xšgx–çxžx‘—x™Wx•×x7x“·x›wx—÷xŸøø˜Oø”Ïøœ/ø’¯øšoø–ïøžø‘Ÿø™_ø•ßø?ø“¿ø›øWÁ\A”d…PH…Rh…QX…SxEPDERdEQTEStÅPLÅRlÅQ\ÅÓ?Š¯J¨DJ¬$JªdJ®J©TJ­4J«tJ¯ úW•I™•EY•MÙ•C9•K¹õŸò(¯ò)¿ +¨  +©°Š¨¨Š©¸J¨¤J©´Ê¨¬Ê©¼*¨¢*©²ª¨ªª©ºj¨¦j©¶ê¨®ê©¾¨¡©±š¨©š©¹Z¨¥Z©µÚ¨­Ú©½:¨£:©³º¨«º©»z¨§z©·ú¨¯ú©¿h i°†h¨†i¸Fh¤Fi´Æh¬Æi¼&h¢&i²¦hª¦iºfh¦fi¶æh®æi¾h¡i±–h©–i¹Vh¥ViµÖh­Öi½6h£6i³¶h«¶i»vh§vi·öh¯öi¿è 鰎討é¸Nè¤Né´Îè¬Îé¼.è¢.鲮誮éºnè¦né¶îè®îé¾è¡鱞詞é¹^è¥^éµÞè­Þé½>è£>鳾諾é»~è§~é·þ诃9¸†iÙáåÐã°çðŽàˆŽäȎ⨎æèŽá˜Žå؎㸎çß œÐ‰œØIœÔÉœÜ)œÒ©œÚiœÖéœÞü¯3:“3;‹³:›³;‡s:—sû?çq^çs~pAraqQsq—pI—ri—qY—syWpEWreWqUWsu×pM×rm×q]×s}7pC7rc7qS7ss·pK·rk·q[·s{wpGwrgwqWwsw÷pO÷ro÷q_÷sð@ò`ñPópðHòhñXóxOðDOòdOñTOótÏðLÏòlÏñ\Ïó|/ðB/òb/ñR/ór¯ðJ¯òj¯ñZ¯ózoðFoòfoñVoóvïðNïònïñ^ïó~ðAòañQóqŸðIŸòiŸñYŸóy_ðE_òe_ñU_óußðMßòmßñ]ßó}?ðC?òc?ñS?ós¿ðK¿òk¿ñ[¿ó{ðGògñWówÿðOÿòoÿñÿ‚À €ù‘mÛ6–mÛ¶moÙ¶mÛ¶mÛÆÝ?Cp„@H„Bh„@FX„CxD@DDBdDATDCtÄ@LÄBlÄA\ÄC|$@B$Bb$AR$Cr¤@J¤Bj¤AZ¤Czd@FdBfdAVdCvä@NäBnäA^äC~@ +¢ +£Š¢Š£J¢J£ ʢʣ*¢*£ +ª¢ª£j¢j£ê¢ê£¢£ š¢š£Z¢Z£ Ú¢Ú£:¢:£ º¢º£z¢z£ú¢úcbc†b†cF"A…у±‡ñ˜€‰˜„ɘ‚©˜†é˜™˜…Ù˜ƒ¹˜‡ùX€…X„ÅX‚¥X†åX•X…ÕXƒµX‡õ؀؄ÍØ‚­Ø†íØØ…Ý؃½Ø‡ý8€ƒ8„Ã8‚£8†ã8“8…Ó8ƒ³8‡ó¸€‹¸„˸‚«¸†ë¸›¸…Û¸ƒ»¸‡ûx€‡x„Çx‚§x†çx—x…×xƒ·x‡÷ø€ø„Ïø‚¯ø†ïøŸø…ßøƒ¿øÇ` Î ÉP Í0IÑ Ëp ÏŒÈHŒÌ(ŒÊhŒÎŒÉXŒÍ8ŒËxŒÏLÈDLÌ$LÊdLÎLÉTLÍ4LËtLÏ ÌÈLÌÌ,ÌÊlÌÎÌÉ\ÌÍ<ÌË|ÌÏ `AbaaQcq–`I–bi–aY–cyV`EVbeVaUVcuÖ`MÖbmÖa]Öc}6`C6bc6aS6cs¶`K¶bk¶a[¶c{v`GvbgvaWvcwö`Oöboöa_öcà@â`áPãpŽàH2ˆ£8šc8–ã8ž8‘“8™S8•Ó838“³8›s8—ó8Ÿ ¸‹¸˜K¸”˸œ+¸’«¸šk¸–븞¸‘›¸™[¸•Û¸;¸“»¸›{¸—û¸Ÿx‡x˜Gx”Çxœ'x’§xšgx–çxžx‘—x™Wx•×x7x“·x›wx—÷xŸøø˜Oø”Ïøœ/ø’¯øšoø–ïøžø‘Ÿø™_ø•ßø?ø“¿ø›ø—ÿLÁB!J¡F%YaNáAI‘EQMÑC1K±GqOñ•@ •H‰•DI•LÉ•B)•J©•Fi•Né•A•I™•EY•MÙ•C9•K¹•Gy•OùU@*¨B*¬"*ªb*®*©R*­2*«r*¯ +ª¨Jª¬*ªªjª®ª©Zª­:ª«zª¯j¨Fj¬&jªfj®j©Vj­6j«vj¯ê¨Nê¬.êªnê®ê©^ê­>ê«~ꯨA¬!ªa®©@i”FkŒÆjœÆk‚&j’&kŠ¦jš¦k†fj–fkŽæjžækj‘k‰–j™–k…Vj•VkÖjÖkƒ6j“6k‹¶j›¶k‡vj—vköjŸöë€ê눎꘎ë„Nê”NëŒÎêœÎë‚.ê’.늮ꚮë†nê–nëŽîêžîëê‘뉞Ꙟë…^ê•^ëÞêÞëƒ>ê“>닾꛾ë‡~ê—~ëþꟃ9¸C8¤C9´Ã¦e;¬Ã9¼#8¢#9²£8ª£9ºc8¦c9¶ã8®ã9¾8¡9±“8©“9¹S8¥S9µÓ8­Ó9½38£39³³8«³9»s8§s9·ó8¯ó9¿ 8À]È…]ÄE]ÌÅ]Â%]Ê¥]Æe]Îå]Á]É•]ÅU]ÍÕ]Ã5]˵]Çu]ÏõÝÀ ÝÈÝÄMÝÌÍÝÂ-ÝÊ­ÝÆmÝÎíÝÁÝÉÝÅ]ÝÍÝÝÃ=Ý˽ÝÇ}ÝÏý=À=ȃ=ÄC=ÌÃ=Â#è òhñXóxOðDOòdOñTOótÏðLÏòlÏñ\Ïó|/ðB/òb/ñR/ór¯ðJ¯òj¯ñZ¯ózoðFoòfoñVoóvïðNïònïñ^ïó~ðAòañQóqŸðIŸòiŸñYŸóy_ðE_òe_ñU_óußðMßòmßñ]ßó}?ðC?òc?ñS?ós¿ðK¿òk¿ñ[¿ó{ðGògñWówÿðOÿòoÿñßÿ\”u|YÇ6 íï7Ü¡‡w‡Û%)IqHK Å­ç¶É6YºÙ  +œ»»»»»»»»»»+·É{ýÜ?ï;óvÞüæÉNåVt@GtBÊáƒtF*‚.Ø;¢+º¡;z 'za'ôFôE?ôÇ Ä΄]°+vÃ` ÁP ÃpŒÀHŒÂhŒÁXìŽ B£ +ãAÕ ˜ˆI˜Œ)˜Ši˜ŽjP‹¨C=fb0s0ó0 °‹°K°X†&,Ç +¬Ä*¬Æ¬Å:ì͈#-H" ëÑ +i8p‘Ád‘ClD6a3¶`Oì…½±öÅ~Øà@„ƒqÅa8GàH…£q ŽÅq8'àDœ„“q +NÅi8gàLœ…³qÎÅy8àB\„‹q .Åe¸WàJ\…«q ®Åu¸7àFÜ„›q nÅm¸wàNÜ…»qîÅ}¸àA<„‡ñÅcxOàI<…§ñ žÅsx/àE¼„—ñ +^ÅkxoàM¼…·ñÞÅ{xàC|„ñ >Ågø_àK|…¯ñ ¾Åwø?àGü„Ÿñ ~ÅoøàOü…¿ñ¶²;²ËXNý °3+XIváÜ‘]ÙÝÙƒ=Ù‹;±7û°/û±?p wæ îÂ]¹s‡r‡sGrGs Çrwb˜UÇ£¬æxNàDNâdNáTNãtÆXÃZÎ`ë9“³ØÀٜùœÇù\À…\ÄÅ\Â¥lä26q9Wp%Wq5×p-×qlfœ šla’)Z\ÏVÚLÓ¡Ë 70Ëó,p#Û¸‰›¹…{r/îÍ}¸/÷ãþ<€ò ÌCx(ãá<‚Gò(Ícx,ãñ<'ò$žÌSx*Oãé<ƒgò,žÍsx.Ïãù¼€ò"^ÌKx)/ã异Wò*^Íkx-¯ãõ¼7ò&ÞÌ[x+oãí¼ƒwò.ÞÍ{x/ïãý|€ò!>ÌGø(ãã|‚Oò)>Ígø,Ÿãó|/ò%¾ÌWø*_ãë|ƒoò-¾Íwø.ßãûü€ò#~ÌOø)?ãçü‚_ò+~Íoø-¿ã÷ü?ò'þÌ_ø+ãïüƒò/þ͸U:HGé$eR.>ñK@:K…T +„"ÒEv¥«t“îÒCzJ/ÙIzKé+ý¤¿ ²³ ’]dWÙMË*Ãd¸Œ‘2JFË+»KPB–*'‰JµŒ— 2Q&Éd™"SešL—˜ÔH­Ì:©—™2Kd¶Ì‘¹2OæËY(‹d±,‘¥Ò(ˤI–Ë +Y)«dµ¬‘µ².Pp¬`°&R–±BE; +GK £åu…¬[têk‚!ÍpiA¸v‚^øÚU%†B!åÇêõ|µ?Öœ57š~£„@ÌMºŽÙ0ý3âFÖuü‰tŽ` NG™:ªNE™ +sx!oúm…¹jÒ.¡|nÂÍ—ÛÞàŸ¯>; +óU”ÓN'Ô ù« ¬Âb›UX¢&s%¨Ñ°f•© +Ê—àkÌZNÒW(Ž*¢ZkT‡ü+Už-íòT×hÖúfé´á[š2óFÙ”å3ìLÊð%L;oÌLβ½ì9+™6ÊòF¡,“²¤à$Ìl.îfÍD³]anŠÛFÚ³J—¬®÷gÌœ—¬ÂÈfÝ6ÛlÉJV!SYbÖJ¦òêcÂms”ÕìæS:,áp›Õœ3+Ýl>U¼ æåä=e3ž·\憂µÑ°M'núRn!gŠw϶›´â†í¸ùÊbp2kØùÌ6³9_*2 j†5kCz>4Ns¼¦þÖóaíWi¿Jû½>ÒÔù#Uš:>ÑŒjVkj½È͘fæ:34ë4ë£Z?ªõ£Z?ªõ£Z?ªõ£Z?ªõ£Z?ªõ£Z?ªõ£ž~¦Ùvã­ïnŠô)ÏnQÌ浟ϥŒ„é+Dk‰-–m{oÆÝT©¬¬w“|Ö2’…ŒbVû GÑnñÇ­lÜ6+,gcsÁ[”/Zz.—¶¼÷eÄMïalÜæ”å +Ž¿ÅL{O£¼8ørO¼ëfR¦?m9¥¿ÑŒ{KUã +ªc …ôñÆô±Çô±Åô±ÕÖWºÓQõbi#^l†"j‹»)¤[lsâÛmõƒcš5:½ÎTôשöfªö¦XT³3ÛeMþ?kX7‡pHS·•pfµ[—Ún—Ïj6²å)oð7ä-;aú­ º*KWÕ ª²TU ºZŠ˜Ý.ûúÿWU¥«¯ªfk2kšŽm8 +®'uÇR¿Z04.¢-«s’e¦×÷èR\]ÊUŠÛî€"Z"R%íJɵÛèÒâFóÅ6ªÔFuö‚ÎÞ¨²TöF½Ñ‚ÞhS»ìmíìíìÍÛmŸ×e“¹ÊX±—Ú´×+ÿ3±:EÃ, r¶áýP%ÛÝnWÅÇk›–áoòšW[›B“êÿm*ª)a™Y3gåþåºìzãÆ­0 $v'd ìï†ä!%a¯F_À¢h³h»@Û»‰3‰;ö¤ö8Ùä×WçfÞCÎÑyHQ/)]|-ѳÿäÄo‹¼þßýTgžÕ´Õæ'}†³Ãþv{±»;M =YäbšånÞɯV§«i­r|÷êãþK‰WwS‰›òÇåq:f¶——»›ÓOËb:ÎOÇ›ãÝë2œ¥õr3ŸCK¸ò1´4^ §Ü{ñë‘hõîz?/Inün’_¾»Þ}ÊIÙOéç á¬Îªó~9îàœýw> 'ÎÙŒ8ÿëöóçítô_¿ÿ°}ò·û'¿òïýs˜O~Û?ýÇÕñüŸË ù¯íýsøO»«ýÓßîö«_ ÷ –öËíÃ$wv’»2É}™äÏ÷/]:ÏÞσÿ4ŸàgßçAÏÍÓ|”ŸÍžÿ± ý‡~sÿäÏýdŽeÜOo¯ŽÏ–c}Úñå4â)›®ø~€à³ÿ +û¯ø~€/ð¾ÀøŸý[¾Àø_à |/ð¾Àp#Ü7Âp9 $Âp#Ü7Âp#¼¯a¾M>û×®¼äú±]£õh@h¹®Bk´A[´C{t@ó¸b·ƒÛÁíàöüßóÏÿ}ùŸqõŒ«g\=ãêWϸú ÊøzÆ×3¾žñõŒo€?Ààðøüþ€?ÀàÔ©;Rw¤î˜_×›Í+æ7ßâͼ^&®‰»ùþ¿yˆ»²c›ý>émÑíѳ® |çfŸNPA#šÐ…à¸n€'pŽP_¨/ÔÍþ›”úB}¡¾P_¨/ÌK +'‚¬ù”Z¯{t@óº'öíÄ~šØ7ûebÿHø=á÷„ß~Oø=Eòñ}Â÷)ÑŸèO¥? ‚F4¡Z£ ºAWê_|:|û|UUR*õñ‚Žê—Ü3ÂÉ—Ô®¾¡¥l_1Áù¦”Ϊu ‚f]‚R¼)Å›R¼‘WÜì·Ûét–Q5¥xÃ=¨†úöãþf«ýÍþ¤ë݇}¾®^‡H b ¸ÍuZ¡¦äÔx(95æ’9½©Wè’ž;7Í +55¦ÆC©ñPc3Ò9ÚôѦ&½ëV¨IïãÒ`2¾LÏ—éùT‚ªu X :¬KàJàKPn\(•C©JåP*³tùDš4ï,]~ò'uh©.¥h>{ý:ïY“¶(žõxÖãåD;áñÄu‰ëRÉÃu,cŒeö±Ì>•Ù3¼qìJG^*sNeΩÔK¥^*õªR¯*w³*ó­ÊÝ,­«R¹*•«R¹*•«R¹.•‹KëâÒº¸´..­ëR¹.•‹EëbѺX´.­óKÃسY÷a,ȳÃþpØ^-ýCÍfæý +56ðÎ`;…N±‘Îh;q¢·NôÖ‰Þ:ÑWtV¶öôµ­QÛµ­q}cklèÜØck´t¶6½µé­MÇÍÞºyj˜ôΦ÷tö6½·é½MgƒôvƒôvƒôƒMg'ò£™X¯PÛéèt¶“§ ا ð„`†05†05†x>‚Øt±ébÓyrB´éѦG›Î3ì3ì3ì3x¦BeÓ+›^Ùtž¶`Ÿ¶`Ÿ¶`Ÿ¶ÀÓì1ì1ì1xƒ}ƒ}ƒ}…gÒ§ÆCúÔ0é¬x›îmº·é¬ªØU»ªbWUXU±«*vUÅ®ª°ªbWU쪊]UaUÅ®ªØU»ªÂªŠÝ)ZœÒ¦SXP±§»°lb7 aqÄ.ŽØÅ‘¼8?1Ûe¯Üß:ækMGø1#ü˜̈?fT?fäYçWA·öN#¯QÐH4Š%*jš…µFÊÊÊÊÊÊÊÊÊeˆ2D¢ Q†(C”!Êeˆ2¢2¢2¢2¢2¢2¢2¢2¢2¢2¢2’2’2’2’2’2’2’2’2’2’2*eTʨ”Q)£RF¥ŒJ•2*eTʨ•Q+£VF­ŒZµ2jeÔʨ•Q+£QF£ŒF2e4Êh”Ñ(£QFSn½ÖÈiä5 +‰FQ£¤Q¥Q­‘2œ2œ2œ2œ2œ2œ2œ2œ2œ2œ2¼2ÔçN}îÔçN}îÔçN}îÔçN}î|sþîêx{s~\~_~ïçß’¡¾vêk§¾vêk§¾v¡y}u<þ±}ü²»<^¿/ýêm§Þvêm§Þv“·OûÇÇ×*G½í¤ÉŸ›öÅñfwºÚß~¸8}=.Á]IRc;5¶Sc;5¶Sc;5¶Sc;5¶Sc;5¶Sc;5¶Sc;5¶Sc;5¶›}»åFm/ïO¦]ùÕ‡ãéýîpüú¨[1êm§Þvêm§Þvêm§Þvêm§Þvêm§Þvêm§Þvu5ß{7lÖ¨C=PA#šÐr}6èmÑíѳ¶ð[ø-ü~ ¿…ßÂoá·ð[ø-ü~ ¿…ßÂoáwð;øü~¿ƒßÁïàwð;øü~¿ƒßÁïà÷ð{ø=ü~¿‡ßÃïá÷ð{ø=ü~¿‡ßÃïáðøüþ€?Ààðøüþ€?ÀàðGø#¼Þo„7ÂÞÚåû:逎Yó}Ô¡ ¨ Mh…Ö(œ|_'mÑîͬa:jÃZÞ^n?ã,ßóO7ÿcÚë¹½TÆ£Mh…nÐíÐn<¼í§ñù´¾6úyG|;íóÓ!µÿþji\Ýîv÷Ÿ_küáøõæi_v·§süñx›k¹xþu{wØ^|Ý~ùvµ½þ¾}vw½=¾¼¹Üß^Þ_<ìþ|»½Ý¾Ïé1 ‚æ5OÕ]îÐØFGÛ¿F—Wô_è ô†G½‘Þø¨7Ñ›õVôVzkzëG½|ÝTö£´²¥•ý(­øî©ìwÏÔ0éö£´jélmzkÓ[›ÞÑÙÙôΦw6ýÿl—ÝŽÝÈ „o¬ È{#²Õ?ÈUwKz<@0˜]dØÃëE°oŸñœO³%â\ zØ¢X<Ý$Kµ±¹©û¦î›ºlê®…T´êÃ;Vq1þt1Ä}asQ÷EÝu76MÝMÝMÝMWwWwW÷ÄfR÷¤îIÝW6Wu_Õ}U÷ÂfQ÷¢îEÝ+›·Âc³½gÕj¬ý/¬çÆÀm\ܨ“:5"ÕP7ÝdxÔ]7‚´=¼g½ìÒ³uægÛ¥g=Û.=Û»é²K'·K'7:¹]:¹ÑÉíÒɳk—³k´fÓÖlÚšM[³ÑšM[³ik6mÍÆ‘7mͦ­Ù´5—Ñ´5›¶fÓÖl\SÛÕ]þ‹!î\`»\`§a»6l׆íÚ°)ÜéĮص»vb§»vb×NìÚ‰NìÚ‰];±k'v:±k'víÄ®Ø3›Yݳºgu§œú¥œ:ÄÐ/ÄÐ)§®åÔµœº–S§œº–S×rêZNrêZN]Ë©k9uÊ©k9u-§®åÔ)§®åÔµœº–Ó p†ÎÐÂZ8ƒI?tÒ¿⮓~P_Cëkh} ­¯A} ­¯¡õ5´¾õ5´¾†Ö×ÐúÔ×ÐúZ_CëkP_Cëkh} ­¯1 %†¡Ä0”Ä0ªºWu¯êΧÊÐO•¡Ÿ*C?U<ºŒþA™Ž¡›ãÐbZŒC‹qlïYµ__C¿.†~] ýº˜ÔÜÔš›ZQSëeºä;á•yá•ÉÕÎU]¹À™u“kšE7¹Œ©,=9òÙt³ß/ðœìÔ“Õ¼Õ¤=§¶çÔöœÚž“i?/Ó~ã·‡7¼cÜTãËZpZÞ³^^…÷ ‡opøvápô[F·åíŒ\YQ)趌nËÛdå»aÛYoj'ïh‚Ú ±ƒ»£!vðÏ_†JʧJÚÁßÁßÁ=è‚CÇø‹ñçY:ÆnñÐ1~è?tŒtΡshçÚ9Ç™Ô~»¦“¾¦ÏyžÈܜַÿòÛåí¿úö_ûðúú"xþöüôôü÷Óþ‘{7ÜecÝúúøýñ×ÿûôó/ÿ`ïøÛë{)½ +³O¿=¾lü&=d+âÁ†xäè‘£G‰%zÔèQß<¸¦ôñ÷o?ÿŽQ^ Iy‰¿aù³É—úÀº°«³&Ö•5³ÖÊÚX;ë`¬ë­•–n·×Àkà5ðx ¼^¯×ÀkàµïÖºKç÷uð:x¼^¯ƒ×Áëàuð:x¼Þàw ðx¼Þo€7Àà ðx¼Þ8ñø}“sàOð'øü þ‚?ÁŸàOð'ø|FãÂh\¶ãµƒý¡¼ü¥O_oûŒÊ…Q¹0*Få¨\• £raT.ŒÊ…Q¹0*F岓#sÙÉgç<ððððððððð¾=<°.¬Æꬉueͬ…µ²žñ:ë`¬ëÎzû=¶€¿€¿€¿€¿€¿€¿€¿”wÿ|zþòåññéé—ÿ~÷›xÉB& ™,d²ÉB&F&FFFFFÆ '`œ€kƒ +òK™‘‡‘‡‘‡‘‡“‡s"N>N>+ÏWž¯<_Ïçä»’ïJ¾+ù®ä»’çÊù¬äµ’×J^+yep3¸Ü n7ƒ›ÁÍàfp3ç”ÁÏàgð3øü ~¿€_À/àð øü~¿€_À/àð øü>Ìb0‹Á,³Ìb0‹Á,³Ìb0‹Á,³Ìb0‹Á,VÁ‡a †±>Lc0Á4ÓLc0Á4ÓLc0Á4ÓLcü>Ìc0Á<óÌc0Á<óÌc0Á<óØæ1˜Ç`ƒy æ1˜Ç`ƒy æ1˜Ç`ƒylœxü^˜Ç`ƒy æ1˜Ç`ƒy ¦1Æ`ƒal‚3Á™àlàlàlàlàðÑo8|üÿcóC e0”ÁPC e0”ÁPC e0”ÁPC eûþxüå`—`×`7±ó5Ø9Ø%Ø5Øí'±åY ±Ëç%Ø5Ø[}kˆ]CìzÇ¿[cë³b·»…ØíÎûû<ï|ÞA°s°K°k°ÏØ9ĺg—`×`kn%¼_Bn%Ä+!^ ¹©o ±kˆ]ïø×`kl}ÖBìb·»Ýy_cŸgRÂ}vv v ö»„XùŽƒ]‚]ƒ­±õ݈uïy ¶þîÞ¯áýr«!^ ¹é³b·»…ØíÎûûÌ»†û:íììì3v ±ò;»»[cë»%Ä.wž—`×`klõ¿#bßó×3máýÞoáýrk!·3v ÷uÚ9Ø%Ø5Øgìbå;vv v ¶ÆÖwKˆ]îõ¨‡TÀ/àð øÔK*àðËöáéoŸ>þôô¯çï_¿ý‡}ò)äSɧ’O%ŸJ>•|*ùTò©äQÉ£’G%J•s¨àVpqqññþ ÿNž<;ïwÞïäÕ9§N¼N¼N~ü:ùuòÄÄÄÄøÓŸüùMò›ä79ÇÉ9NâOâOâOâOòä;Éw’ï‚?÷¿þûó_µs‚¦í\É€‰‘˜‰‰‘˜‰‰‘˜‰ ‘˜‰ ‘˜‰ ‘˜‰ ‘˜‰ ‘˜‰ ‘˜‰ ‘˜‰‰˜‰‰˜‰‰þÏv,KŽQÞ1 ž¤Ã–-Ë +V¶³ò=f€Ž˜a&6¼=ÕíϺ ÁâÆQéÚúÓUÖÉ#Ž°r„•#¬aå+GX9š¾Ñ¼¿ÑÊê4ÓBºÒJ7ÚèNŸõzÒ‹}Ñ»þÚq;nǵªP;nǵ3ªQ;nÇí¸×ΩvNµsªS\;¨ÚAÕªvPµƒªTí jU;¨ÚAõÀ=p¯õk’{'¶÷ßôé/ßÿúÍçù›ÏËÿù¼óyþ’î¾ù|ýÏçõ¹þ w~Ý¿ó[gZèBWZéFÝi§=éEƒ¾hÞ:ãÏø3þŒ?ãÏø3þŒ?ãÏø3þŒ?ãÏø3þŒ_ð ~Á/ø¿àü‚_ð ~Á/ø¿àü‚¿à/ø þ‚¿à/ø þ‚¿à/ø þ‚¿à/ø þ‚¿â¯ø+þŠ¿â¯ø+þŠ¿â¯ø+þŠ¿â¯ø+þŠ_ñ+~ůø¿âWüŠ_ñ+~ůø¿âWüŠ¿áoøþ†¿áoøþ†¿áoøþ†¿áoøþ†ßð~Ãoø ¿á7ü†ßð~Ãoø ¿á7ü†¿[o·Þn½Ýz»õvëíÖÛ­·[o·^÷<ÝótÏÓ=O÷<ÝótüŽßñ;~Ç¿ý{N~•ü*ùUò«äWɯ’_%¿J~•ü*ùUNÏú ú¢÷ó%¿J~•ü*ùUò«äWÉŸ’?%Jþ”ü)ùSò£äGÉ’%?J~”ü(ùQò£äGÉ’%?Êòð<?J~”ü(ùQò£äGÉ’%?J~”ü(ùQò£äGÉ’%?J~”ü(ùQò£äGÉ’ÿ$ÿIþ“ü'ùOòŸä7Éo’ß$¿I~“ü&ùMò›ä3É_²>몟Ÿ$?I~’ü$ùIò“ä'ÉO’Ÿ$?I~’ü$ùIò“ä'ÉO’Ÿ$?I~’ü$ùIò“ä'ÉO’Ÿ$?I~’ü$ùIò“Üñv¼oÇÛñv<~“ü&ùMò›ä7Éo’ßäþð~Óñ;~Çïøþàøþàøþàøþàø'þ‰âŸø'þ‰{➸'î‰{➸'î‰{â^¸î…{á^¸î…{á^¸î…{á^¸î…¸¸xxx88/œÎ ç…óò|/¿ë ÷…ûÂ}á¾p_¸/ÏùÂá¿ð/ñ/q'q'q'q'¿rÊt¿·oi¡•nt§}ÖùZoÙnÿ)õþýR®N¹:åê·.ßýùû~þôÃ/ÿúéÇß~û29•»y”éTÔý2O¥Ð»é¼µÒûðU¢ŠCTyÑœ¨ûï/ó­ µ^Z/­w©oÝi§=éEƒª#oþr7™·Î´Ð…®´Ò6ôE­«¾ÅÂPC! …0ÂPC! …0Ó³N§=éEƒ¾èý’†0ÂPC! …0ÂP8¼…Ã[G!…pÂQG!…Ã[8¼…°ÂRK!,…°ÂRK!,…°ÂRK!,…°ÂRK!,…°ÂRK!,…°ÂRK!,…°ÂRK!,…°ÂRK!,…°ÂRK!,…°ÂR8¼…Ã[O!<…ðÂSO!<…Ã[8¼…0ÂTS!L…0ÂTS!L…Ã[UáðÂU8¼…o!l…°ÂV[!l…°ÂV[!l…°ÂVÜakÚî}õÖ™ºÐ zïÏmvßì¾Ù}³ûîý0möûf¿ouÿýß~ú÷¯/½»ãõúŽ~úëç|o2Mæ×ÉqO»§3¿¹6sLþü㟿L¾³ÁL ]èJ+Ýh£;í7s™æiŒæ1ÒB†Þ-aZÖqÅ6FmŒö1úú+¼;p¡×}ÿ¤+LZÕdýû×xëJµ®Iëšt•I ›´°ÛåÞz»ùÂå.¿h±qûE7XtƒE7Xžÿë‹n°è‹n´êFëÓ vŠ[q«zî·þ]Ç£º.³è2«®µê +õnáS½£ÜT¯J7êmóÜÛì-VǦŽM›:6Ýh«¯ç—jåw_fZþ4¦ÆÜêÝàVÇ+ÒÆÝÆ«tŸ ÞW•1ó1ZÆhuÕÇÌx™öñ2ݧ‰÷U£š>^Ô>ª°ÉZ•öë?æžê§êûSý<=Õ·1SÆh£§ú}ÌlcôqçSýX}~V?ÆÌSû9fžµ¯13ÖÛlžŸµã™)ÓÍcô¬þ3ëÕ1z¾™3㹆EÌ‹Ú·ÁYÆ7³ŒoF´kÛ¨aÕ/£z[°mc…u¬ºŽ;WÕocõuT¿ŽêWÕoÿÕ¯£úúT?îïç<ÞÁ¹>ÕšëÇhT_ŸêÇÌÇwòñÜÛSý¨æ£Â*¶§úQé7×¹öT?Þ›6ªo£úöT?Þ¥gÇnã]jOíã½ÙGíû¨}¯†½iØ›†½iØMÃnvs_Ó°›†Ý4ì¦a7 »iØMÃnCcÙ ¿á7ü†¿ãïøOÝ;þŽ¿ãïø;þŽ¿ãïø;þŽ¿ãïø;~Çïø¿ãwüŽßñ;~Çïø¿ãwüŽßñ;þàøþàøþàøþàøþ‰âŸø'þ‰âŸø'þ‰âŸø'þ‰âŸø'þ…á_ø—(Qj»ÈÑEŽ.rt×u‘ã4š?Ÿù;Š¼æö‡{‡”½<­²Œ¦VöuŒêmcÔÆh£g‹—ÑØÊhl¥—1ŒÑÚÊhm¥FŒÑÜÊhnË4Ñóä{ókóõüöü¼~6ÌF㬆…YK²¼Ìò¼éØ„è»ì;‚£®q—ášw-¸ò û¼}Á¾h_²çyíµvÝÞnï²ØOØGíãöüQû¨Cér ¸N¸˜ÇåuÕºtW»+;èxäëÁe°Fjœ„ÝËž’¸I”â)Ø ˜ çÐ^„æ¡çFœqnXÝ°ºa%`áiº€ËkÏyVƈø%á¶À[kj»^- ===Q Ê'X¡ìÚ&m‹v xÅWkù»»ô/ɘŸ.Æ*ŸèÝ[櫨QEÇ«èhÕCáH@/&µ¤/Y™œ°õký¾þÊþ [›Öæk«l›°…µ°/\ž°Õh5¾šÊš ׸Wò ÛHëlëÖ›­¶dkëp+«Ç­K›Õµ©å>¡WÌ õîÈ£Ê,.' îŒpp ú›2 æÊ ¬3°Î6 äaÄŒx¼€¹åö1é-áW>ãg¸ðisG°-Ò‚Gn枆ZFg[³Òn€¥½ÍŠ—v^Ãð€ë”¹N¿N<ü;IòÈM¶/‡ýbf0€YÀÆ:ñßÏö+3øO+Ó̯«uk9Y·ŽR¸Æé‰x”ÕØ*}Kò+’OKK®Ð ZÔ{-êÏZÔ[Ô-h(•$Ç9Éeº+¢^Ž¨mµ*¢b¶‡HQ•µ’í‚é_$?.Ù¯•©—©•©”©¯•©GËÔ/–‰q›pvU¥H²K0=/¹EòfÝÅÕ_pu?Wë¹Qé%Šì¤Qr©äbÁôÃËä_£’(f¢f¨Šg"….›¡ä?fhäßfèä_fè,¿N?¦ò•Fï™wyd-ý;m¶‰þG–~@›Ét z:IBÔ}Ó ñ?Âø è¿AÊ"þuÒ.ÇÑfiÍ÷CÓ߃¬Mÿבõñˬ/›þ»°ž5ý§!ß7ý‡!#¦O,ðÚÊ#kèAR¡ˆØ^âSÄJZ­Œ_Â̇¡»²ƒc¦_ŒŠŠÚdju-b•×©FÚe:njò"Kˆ&§ØD4¹èbâ“Z@Ýrñ*)—ê4µ“˜Å~Ùw—ÿ3tM\8ùu›—ø¯ãúö¡ûÚlNñßΉr™ü¦?C}Wùo´küçºÏäóþŒŽþŒB¯ðŠl V¡Wù¬ÿ ŸÑ¤wBƒ·z,´_Ô:ù«>ôM~Ò],ƒÁïƒ;ጷ†¦øN_†Â­‡L_Åwh_åÛanÈÐæô¯«Èˆ¥ÔbŽ©«|+2nÖäR¾\ÿŽòyâ Ïë~ÇsŽÇ>ÇŽGAÇ6‡×QâØä(r:=Îçjç*§ÓiwÚœŠ“8‹2Ë‹z5Á),²{„Øm‚m²íQƒÄS_¡NgÇø‹+ñ=Ô(Œ“øÞF£¾:žq,ï6ªã†³ý+)J¿—@ÏP¾“¡do6¨0*6 +›:æ¥5§Î ¾§Ãx»$aDc¹$7víñ>Ù1§UúcÑ9e@H¢cŽ¾ ívúB4‘ #åÊÂHHˆK“rFÊiZ†µÊ0lÓòX4U^ž z—6‹ lŸweÐÁì\H¹Ú… L)%r® +¥T„a?d'sz²Õ„ºådîÕDN¶I¥|>„ø}"$UïC@ÊW/ÝS÷Ýš/»œñÉ<>šy(½S™Á.°b'bªÿŸ¿¾Æÿ!˜¦»oèõi±.-Öt/f½q¢ÇëM¸%^ƒmîêé}FhwŸqKë‹´¨7ÕÝûw¯pwkÑéííHõê}Q³[ïŽiÝÑDzr¸)þ™\§s¹š†0Ù°˜¬I䚌?ÀîI‘+.rÅE®I}RæŠïn¤ñöŽ”“4&šžÌê©,c–†(Žß5­C«¶*H S¯ãAðЦM\Aº4E°"*\nm?@?@· "]:¸¹tpsusqpsòÿ’*ø¸ß{¹÷.¹p÷^rÓL!zè–+ac»t}Ç~Å”gY†ßVA†ñjµ¯ +9žãQÕI¡u¸‹Ë=د”gü~*Á½Qm0Éì çÿ5­uD#¡#c'¾E[iÇG§v¬bÄõ®rÚ³”f»^š«…ÊôÕPÔXMTΘî͹XˆÌ¹è‹¡‰±˜ˆ +œµëj,>…eM<‚~2§E£nd4 äÓI#›mO°Kœv9NæÛUP-cÏЯà|,»¾w`J˱œÀîù4c(é£c[îtg×Ý{‚½¸Jm«“Úà$µÊsm؇ÃZÞ+âàÍÙ ú¼ð r–k¹ÉÃMšµ¡fZr¼>C'"¥eÄ%.8-w¤¥d%8vC%ÿŸ÷ŒkðØ J¼šn3d…? ±¸ endstream endobj 29 0 obj <> endobj 40 0 obj <>stream +H‰”— XWÇ«_÷ƒ% Ý3Ðã… ˆ(„ â#ââD.adDT4‰q³nÖ(1jŒ×bL-’EŒ—'|»ZŽá8Çù®Ò²%–£Õ8ÐnqQqvÄ?X÷-ŽWº²ÊÜ~"ÆZò2ú+S³\9»÷'Çñ—»xzi«ÞM UþXî>wIŽ{|Q¹8àóhsõt¨){vI„ä•äbÞœI%Ð¥(«t*$„>ÐŒ9 %<¡ rž¸B0R¶Ë !ìÁŽ‘“‹ +ò ]´Yj>Àî“þãŒL!à€Ž ƒÀKЮ$3_>Ò­WG´ÞÇÁK³f·>6{@ÐüÍ×øø!€V·æ"0Rq +™R&ù“ûä>–ÔDš@­b0šOË@¢å´‚h}¬ô]º ÚÒ-t „ÒítØèz¯ ¯6ÐÙ”f]M£L£ ÖtÛtº›ÓÍ#!Îœa΀ž Ð)8;ÁAbÅþÒhå³caL˜óa!>‡•P›a'ìÁ¹ÓàËpWßx$H‚Eº¢T+ÕI—˜­—.3Û ]aÖ#}Ž¶ÕÌÖI_2[/}Ålƒt•Yô5´×pTÞß0['ýÙzéÌ6H×™õHߢw½tG èý³uÒ?™­—¾g¶AºÉ¬Gú½¤[8ò ÷mfë¤;ÌÖKÿb¶AºË¬Gº‡ÞžŸÉ7”á;øÿ¹Ïf^+=àdr2œÌ#Næߘ§Vjâ|šu.t.TйP¢¡¢N„J:Ju"ÔK'B½5"Ô¤¡fõщÐ:jщЖê«¡~:ꯡ­t"4@'B[ÿ"+`l„êÿF„êDhN„Zu"4X'BÛèDh[ щЧôCC9'ÆÉÈÚŠ¡ +çcç|œK8çÁ‰¨œˆ“ilj´çD:0"9‘NœH$'Ò™‰âD¢‘8 õp‰\‡[ðP ‚í‰tåDb8‘nœH,'Ò‰ãDzp"=9‘xN$yšIäDžáD’øŠéÅÉ$s2½ÙŠIá|úp>}9Ÿ~œËsÚLiÎeç2sy–s¤sùÕDnDs"©œÈN$ʉ cD†s"#8‘tNd$'2ŠÉ`DFs"c8‘±œÈ8NäyNd<#2™È‰dr"Y|ÅLâd²ÙŠ™ÌÉäp2¹œLžNwsA«[Ûõ…¥¸û[`*nûÚw%œÐyõ…!0Úœ»{%ZÍ“¹ +6ç0U†çr¹ +6ç¡*d~ù\› ˜Òü¦pŒyB ¢p¯OÁ¯Ë(ÈÄ]½Êa¡¹ÐÈTddr™¦™ŠLn#Ó4#SÉãLæP¹hž›ËU°ùE¦ +ñ\9W¿TÑt£¢R£¢FE3Šf•Í6*šcT4Ϩ¨Â¨è%£¢ùFE¸· +QB~iCH$œ„kÏK¨€¶?¯VÁ;¸~?„Z\¹÷qÍZ« …X!I(ÌÔºÓY ä-¦Îªö±"§P­dê´¡ÎꬡÎ1E°7°íî•ä+<®`×ê ¯zC50%âªò…@âawÂãkä0—3ŸOŸð±’-ùDô\AþfD:o¨Ï uÁP uÉP— uÅPŸ3Eñ‰â*w@{r 3­Å\ÇÑ®C{=Ö’Oð¸Ž|aÜ÷%Ÿ·7YL*ñ©T‘è¿™lRMªÁì$ïaw´‹ì†²‡ìÅø"ëÝñMÒº2­£òÖz>¼w=^ØJ¶bÌÝè/’ävaØ_‘eÈÈj²šu[Þƒ²¾¿dY6²†¬0ŒqdS´)z™ŽšN³øZçfbÕ~ƒ‘¶i]©Â(øe…SwSìÏú9÷CìµC$öœÉü¹ž`uœ4žÏUT«™úÚP×+q‡æý‹óÅB¥Ö±ÒUø‰Tf‹UKJ©DÍ"6d>Ø;—õa}(þÒSû(¨”æGtqS¡›·]؇=xå¶"Óp_Æ>ÞèŒÑúšôûðÉj6­æÁ´ „럟ü Z[C½¼MfŸ––¾~þ­ZYƒÛ´ y*Ô&+vGx„êl×¾CÇN‘£¢»téÛ=®GÏø„§ŸIê•Ü;¥Oß~ý |vÐsƒS‡¤ 6|DúÈQ£ÇŒ÷üø 3³`RöäœÜ¼ü‚)…E®©Åîi%ÓKgÌœU6{Î s_,ŸWñÒü—üæ•ß.üÝ«¿_ôÚâ%•¯/}cÙòo®|k¬Y»îUëßÞðλ7mÞòÇ­â¶íÕ;v¾÷þ®ÝìùÓ‡Þ»oÿþ寇×||äè'ÇŽŸ8yêô™³p®¶®¾Áó)œÿìÂÅK—¯€d¹†µ +Ú +´B¹ÐL2†lÃÅ4±Xœ!–‹¯Š‹Äõâiñ®ÔRJ£­©¥×éÛ[¡í°í¸­9l^Øê°ua?„=ƒd›ÜON•3ä1ò8y¼ü¢¼[®‘ëä ò÷ò¹IñSŠSé¢Ä* J¢’¤ôU&(ó”eÊÊM;µ·¶[í»ÓebO·O°/°/·ov‡—ÃÏàr„8dGG'Ç@G–#·=ÿp» +*Q-ª¿¨¶QCÕ5RUÕ"µB] .T©o¨ëÕmêûê^u¿Z£žTϨçÕkÎDg²3Å™éÌvæ: Å‘®ÈYQÁ›ì›5’ƸÆÄƤÆÞ}Ӛ›±ÿÙ¨ÂÄNÆ’j1B*–ŠsÄÈe±¸A<+Þ“|¥¡4„¾NÏÒ›6@.E¶[S„U —ª°[2ÈmdE(å\&Êòùˆì‘/É·ä» +(È%Z‰QâŸàR©Tý„Kª}Ä„— tGÇç™ç™'óZƒŠ $vBjßwj×V[|­¯hK-EQû’Eˆ}IH¥vj‚ľ« Ö{›ÅújÈb‰D":‰~ÎwzN÷œ™3sÏÌ{çþÏ}þóûWèktYðN—"F—RžëÒßs`ž.åÿE—ïtYàµÚkó;]Î]n]¿Óe÷£Kÿ#. «lVC£Kˬ6YísÊçêòæ®)™âæß“›ú˜Ö=·hrç•NºU3¥)Dvlöù¤~B$-Ira¯3£÷‚̨‰ÝÈn`׳}íºvÛÇ®m×´kØÕíªvÛÛöÊ}OB@^¸1aœý*!Ò>kv>H˜09qLüøñ QvßÄ ólïøñ!q!qkâ‚…ˆ[Ÿ»/Þ=ndÜWfæ×2Î7®’³½³³©³‘³¾Ó×éã¬êôt–vº9Á‘ìHrØŽûŽ;¹»§1Žh‡9¹ã¤cc»££µ£•£’ÃÓQÁá‘“·¦sn¯¢MÙ¬p w sYî²,Ÿ«LËó7ã±æ·77+¼MûÚäV¡~Ta¦OÂ$]³ë­Åè<ÏqÉyÛÌÜG¼çѺ©éûè~ÿø¼×{wöËmúË¿g}Þ·þÿvÖ57ÄÿÆÿrBs®†¦o­{¿ý¦<(Ö + ¦~ Å<,"Ä&C'®b¶ÍO,©"MÌ¡†ZŽZI1ì½Y<3¬ý\¬[Å“Ò¶>`¸| 8'‰Óâ¬Émç —_Å7Ⲹ$b í|+’ÅBÃêWÄUCóD’˜exh°ø^ CÅp±ZŒ# å£ÅÃIcÅ8aæŸ Æ‹‰b²˜$"ÅÏbªá§ibºÉ?ÅPCOh’¡Y&!.ƒåáâµÈÐâ ¬€X «`5ü äƒü†¶ÖÀZ‘.^Â/°ÖÃØ›`3l­° ~…í°vÂ.Ø-2Ä5˜ Á°öÂ>ˆ„ýP +Áˆ‚Âà +E ¨H‰P Üà ‚âƱçÀaˆ†8Gá¸CI±]ì€Rð‡PÊ@Yð€“pJdŠW⎸ å ùUO8 gà,œƒóp.ÂoP*xÃ%ˆ…Ëp®Âï" +*C¨ +ÕLÞ¹×pîÄ]¸÷à^܇‘¸`ÄCx£q#Æà±â1<Ž'ð$žÂÓxÏâ9<ð¢L–)æ¹dÜò2^Á«ø;^Ãëxoâ-¼-Se:Љ`Æc&⼋÷ð>>À‡hã#ü“ð1>ÁdLÁTLçø Ÿã LÇi8g Ÿñã Ä™„³ÌlÜy®ñçùø30³0_c¾!A@’ˆ1¹&‹òQ~*`î·BT˜\©¥bä×áÜ„[pº!Õ§ÔQcjBM©5§ÔŠZSjKí¨=u Žº‘n¬›žz j[?Òê$ýX?ÑÉ:Åò³ü­+КiY³¬ÙV°5Çškͳæ[ ¨}D© u¥nÔzPOܤ›êf´ž6ÐFÚD›i m¥mô+m§´“vÑnÚC{iEÒ~:@QtP\ñtHܤÃM1t„ŽÒ1:N'è$¢Ót†ÎÒ9qKÜ'nˆ?è<] ‹ô]¢Xºœ›^sS¥I7è&Ý¢ÛäÐÍu ÝR·25ÝF·Õít{ÝAwÔôGº³î¢»ênXÝÕw2Ç$!&Å UÃd†®F¨ÔH5J6ž7FUãÔOj¼š &ªIj²š¢¦ªijºš¡ü”¿ +PXJw×=tOÝK¬?ÑŸêÞ:U§é§ú™~®?ÓŸë/ÔnµGíUûT¤Ú¯¨(uPR‡U´ŠQGÔQuL´Z‹¬ÅÂR ÒÀ Oá<‡tx  ¯ :dA6¼†¨aGH|%I%YºH--™jÊü²€,( ÉÂÒU‘Ee1éµdqYjƒt—%e)“ºJË2²¬ôå +Í1Tà u ®¬¾²’ô’Þ²²¬"«Êj²ºqé¾úŽ¾«ïé:]¿Ôꉬ!kÊZ²¶ô‘ud]é+?”õd}Ù@%«9AN”“äd9EN•Óät9CúI Re œ©ÒÔSõL=W/Tºz©2T¦z¥²T¶z­rÔ ,™X1³ k¶8çç\ ±+á¢\ŒÝ¸8—`w.iÅr)þ€Ks.Ë\ŽËË 9 —àR®Àž‚¡\‘+á2\ŽaÎ^ìÍ•¹ +®À®ÊÕ¸:×àš\‹k³×ẸWájÉkp-ûò‡\ëã/¸×ãnÀ ¹7æ&Ü”›qsnÁ-¹·æ6Ü–Ûq{îÀ¹“µ„;sîÊݸ;÷àžÜ˸Ê&þ˜?áO¹7ÆŸó܇ûr?þÉ_qÜŒ[ø¿ü5à<ˆ¿áoù;ÌCø{ª3õ+Æí¥Vˆj-³–[a<‚à‘<ŠGó<†Çò8þ‰ÇóžÈ“x2O±Â­V„µ’§ò4žÎ3Øý9€y&ñ,žÍÁ<‡çò<ž¯³x/äE¼˜—ðRáP^ÆË9ŒÃy…ɯóä|¹@.”‹äâÜÔ&Ct¶ 5ùq¹Éiá&¥E˜ü¸Š#ôk£ßèûÖ_ì×upWWð÷Î9÷Þs€mhq—xBàîîNwO°(-Å-¸k‚KB hqŠ †Ii×ØÍZ»¿0;]Úm‡öŸ¶eîÜ;÷½wÎ?oÞ÷ýÄÈf;ÅNµÓìt;Ã~`\›q7ÆÓxoãc|MYãgÊ™ò¦‚©hüM%`MeSÅTµÚœYtFàDgVŸÄ·ø6ßád¾Ë÷ø>§p*§q:gÀSxÏ!“¯[±VÌÄLÛßÚg%X'íÇÖ^+Þ:Å7¬)Öqk§Y§í9ö\¾Éeؕݬ—övg;=Ù ’¬Döföå” – µ–[/œíh+Ê®eͳëØáö;Ê^h±Øì'Ûe‡ì”]²[öH¬ÄÉ^‰—}’ ûå¶Ü‘d¹+ä ’Ã’(G䨓4I— y %IŽË#¹'÷%ERqîÄÝpÆiÛ³|ŽÏó¾È—ø2_Ïà,œƒóp.Â%¸ Wàs¸ +×àºEyr;sYŽWÊÉá+òûºyÍ5ç_iæçZÆómÖLVϬAYaYã³"²&ý„lÄÊ•­›ï‹æ5Ë|“-˜Ÿé·Y0Y!Y³†¿z‘?¦™ìúJ4¯)æ?~ÉÞgþ³¼½jq™êé2Þ%Ü%Ìeô ÆyNr9O‹sê-zs=¤ÓM€óS9”Ý£TÎËÊõÕ›Ìà˜ìXã\GI?x’ðÆÞ¤ìé8ñï«£oª­s‰cÙwûE?Q3ÅáÔ—#Æqè÷ãÿ«¿ïôòN/ÿ·z¡¼”òS*H…¨0¡¢TŒŠS *I¥¨4•!Wr#wò Oò"oò!_*K~TŽÊSªHþTéïjKí¨=u ŽÔ‰:SêJݨ;õ  +¦ +¥žÔ‹zSêKý¨?  4ˆÓJÃh8 ‘4ŠFS…ÓKãh¦Oh*}JÓh:Í ™4‹fÓšKóh>- (ZH‹h1-¡¥´Œ–Ó +ZI«h5­¡µ´ŽÖÓÚH›(ú·ä45MMW3ÔL5KÍVsÔ\5OÍw&…(µP-R‹ÕµT-SËÕ +µR­R«ÕµV­SëÕµQmRÑ*FmV[ÔVµMmW;ÔNµKíV{T¬ŠûµúOqfʳêœ:¯.¨‹ê’º¬®8sÕUuM]W7ÔMuKÝVwT²º«î©û*E¥ª4•®2ÔõP=RÕê‰zªž9si¦úò—i*C¨ +Õ :Ô€šP jC¨ õ >4€†ÐCh +Í 9´€–Ð +ZCh í =t€ŽÐ :Cè +Ý ;ô€ †…žÐ zCè ý ? €0à +Ã`8Œ€‘0 +FC„à ãÞ‰ö7)Úî܃ƒ8˜C8”{r/îÍ}¸/÷ãþ<€ò ÌCx(ãá<‚Gò(ÍaÎcx,ãñ<#x"Gò$žÌñþ˜?á©ü)Oãé<ƒgò,žÍsx.Ïãù¼€£x!/âż„—ò2^Î+x%¯âÕ¼†×ò:^Ïx#oâhŽáͼ…·ò6ÞÎ;x'ïâݼ‡c9Ž÷r<ïãÞÏø âÃœÈGø(ã$>Î'ø$ŸâÓ|†‹%¶€ (Ñb„E$‡ä”\âÜòž¼/.’G>%¯ä“üR@ +J!),E¤¨“âRBJJ))-eÄUÜÄ]<ÄS¼Ä[|ÄWÊŠŸ”“òRA*Š¿T’ ”ÊREªJ5©.5¤¦Ô’ÚRGêJ=©/ ¤¡4’ÆÒDšJ3i.-¤¥´’ÖÒFÚJ;i/¤£t’µ²NÖËÙ(›$Zbd³ŽÖ1z³3ÕoÕÛôv½CïÔ»ôn½GÇê8½WÇë}:Aï×ôAgæ?¬õ}TÓIú¸>¡OêSú´>£?Ógõ9}^_Ðõ%}Y_ÑŸë«úš¾®oè›ú–¾­ïèd}WßÓ÷uŠNÕi:]gèú¡~¤ë/ôýT?ÓÏu¦þR¿Ð_é¯õïôïõôõŸôKýgý¥ÿªÿ¦ÿ®ÿ¡ÿ©¿ÑßËØ 2Êhc 19LN“Ë8Lnóžy߸˜<æó¡Ékò™ü¦€)h +™Â¦ˆ)jŠ™â¦„)iJ™Ò¦Œb§Úivºa?0®Æ͸ãi¼Œ·ñ1¾¦¬ñ3åLySÁT4þ¦’ 0¦²©bª¢z¢z£úbYôÃrX+`EôÇJ€X«`U¬†Õ±ÖÄZXë`]¬‡õ±6ÄFØ›`Sl†Í±¶ÄVØÛ`[l‡í±vÄNØ»`Wì†Ý±a0†`(öÄ^Øû`_ì‡ýqÄA8‡àP†ÃqŽÄQ8Ã0ÇàXg?´áxœ€8#qßâÛ|‡“ù.ßãûœÂ©œÆéœOá<‡L¾nÅZq03mkŸ•`´[{­xëß°¦XÇ­iÙîá4ë´=ÇžË7¹ »²›õÒ~Âîìa²'{A‚•ÈÞìþÀ$Á"¡Örë…SAÑV”]Ëš÷/®«ªŠë ÿÿ½3¼7¢ˆëCÜæ9€Ê"*&à†(<\­ iÖ“>T1ÍÑSð`ëÒV}·D5ëÉi4Ae@Q41Rc"âB 5âRƒI4£QSO€7ýÞ3Çhçž™sgæÎÜùþåã±¼ˆ7ò&~™S —Øãµrm§¶KÛ­Uh¦V©Ui{´½Zµ¶OÛ¯Óš´óÚ­F; Ô>Ð>Ôii‡µ/´fíŠö¥ö•V«ýCûZ»¨]Òþ­]–;å.Y!>µ³×ÙÛëí'ì'í§ì§í ☨ÇE½8!NŠSâ´hÿgÄg¢Q]G¤>C}qö’ó)ŒÈº‚óºÿôM²ZÕ¹dø +¬/e2àæŸÏ‡G$í u –µ‚\4ü¯¼le½Œ¢;|ŽÆ“‚U‹i ¥€]uçi4 r/¥¿A»í@b-Ð÷)ôìÍ“Á[o+ Ô¡mË}g|ÍVwë.u„¬ÏR½Ž»¾yH»ž÷ìÖ®y½Ð¢¼¡:,²VÁk:ÅÓXX«˜^q_|69€xè÷^@‡R.¾ûª”_)ßX©ÖëcØüÙÈÀˆ¢lxu)¼¶¾ÛI»35TGßÒú,YÈ¥\ÉÕü£è*v‰³J+úº;ÖëV+ÃÚ‘Kƒ1`Áñ4²¼H[á©ãtŠ.ßþÆ;œÿÈ«ØËëx3—!KýRƒ¸Œ¾å=ð×:ð¾¡J!:ïæ ,[ž¯Ì·ÕÊ€v]ðïaÀM2l˜,þ˜x v¬¢ýt²ýH-°KhÁ#9›‹¸„—óz~‹Ï£,ô÷m’eoiÈþr58a98Æp€R_”/×D~Ü´FBn7ÆïhvY‚Q +;TЇðÖ1 öÐ|ŸZ°›€Ÿƒ¹;¹?»0ž…×Ýü<çq/åw¸Œô6úë0Ñ]Å+âÄä7Èg/Ëmrl”>ÅRƒÕ¡j.ô-Wï¡_Zcg›nÛn?ÙÝV×vÙ×Þ×Í×ß7Å÷gß–ÛZd-¶Þ¶¶[»¬ +«6©Øí |éýi"'ƒž¡ç!ÿ\šLzimÄØYˆ;CŸs}q®Ã³7:ݧVèƺüNäivq;x züÎBòÇb/2ô)Q¥Ü!œËK$‰8%Ϥkè£#9YqOÄñK4QHØ0}ßx Ûd#Š€ÃµQéÆÛ‰Ä_¹ù¦øÉÀšÕt"­&êD­¹TÍDT¾µ±°Œ3¸14üá[¥U†¡Wäyàæb— i"hMoRw^¦bÖÅ šJ[äjxÚI=(ZyAEç{²ŠÞebØkDß!ïMUÆ3)ÍÈûÑÔ—oÒnÈV/Å®V‚ømþ5dè%íÀÇ1Š¯S¾\È +øÊJš*cùž2DJš ?­¤\¾ÉvÚÉeâ;i3/ƒöWù¦¸J èðË6¹^pãî"†ÇÉÁäWx:¤‰ Ûªƒí"q\]ïËY¼Õ#ò’’)÷‘Âq¢h•ºHãL™dݢȠ²ƒï¬•Ji²6)ÁmßÃ:óé¼uTÆ)yʤ–ê–áàMò÷ªÛºç[ª®É4K½aMÅ"¢µ¨‚¢ù{»÷Å“°”CÙÐÒ"²¨·¸Ã÷©ˆ×#:" I2GÍæX«¢6AøI”#kfÊ…È3ûè(Ð^‚ÜÞEÌ@)àl¨J lî*s¨˜–Áÿ‡PMË1ë£þÝ—B§‘÷~ƒX¼È¥ˆº b¸â¦)¨¥+¨_ÊØœ”1É£G1<)ñ©a C‡ Ž=p@ÿ¨È£ŸSïÛ§w¯žá=ÂÝ»uíÒ¹ShÇíƒÛiv[ªHÁë2Ò=ºå1•(c„8ÿ½‘‡y=ð˜:¥?¹ÆÔ=eú“+S°rÖÿ­Ly¸2åÑJÕGѨ¸XÝeèæ©4C¯á©YnÌ_J3ruóV`ž˜oÌ;`îtâÝV¦›ìÑ]fú¢¯Ë“†ßU·K5RóÛÅÅRe»`Lƒ13ƼJv$s`"®•‚ì ”n¤¹ÌFš_SFºòfš“³Ü®´žNgn\¬É©3Œé&ãÌŽ1%”ØÆ J5mmô9~m¨T¯Œ­õ®­ ¥éž˜ö3™yÓܦÌËõïÑ)û¦™Ž%_‡ýr‹ŸwNu¯zümOéu…ÍÑý·^ï*Ý|+Ëýø[§ÿš›‹à[™îñ¦cëµ~#†ÅC¿ø~U*•o¸üO<…º©ãŒo¡þ÷š”]ì¬ +O9`5S¸K÷æ¸ §9¦§‘›—Ö«²+y³‹÷ôHÑ{<ù&.¶2´ÓCkV†tüyÒ¾Ãã“üGï³Àrÿ,#û‘9Ù/‘1(0õ:$qP$ÉÉO"ïŒ$,ÑËøÊœ 7Ì1µT7t„ÿ¹ÿ{S ýãÕÛÔuÆϹ÷Ú×q|›Ä×7/;¶oü¼$qÛÁNJnÉJ%Ð$Äk ŒE¨E<¶V턶®PóŒ®ÛÐX·þVÉ ¯›v´A¼J¥®+ª6u•6²)í涪Ñ*ÉÝw®@4i›åßýÎ÷}ç;ç;ßÃ÷Xr¥î"H»”ù×bɦœÄèåï"2$űP_ Ÿ§e9 +‘º`[!‘àã2V/yF¥>—vð. >´®Ìúµs·›dõ€ª ÍÀ¤Gºû²¼ m.GJ­ÜŸ¦†ˆfr^#¬'š‘yÍ‚ùå{î i“oá[ÀÛ‹Û‡ilÿ/êogõ=Rg÷@Ÿ«=5”‹mgï".«_º ËÒÅ­}t9•Qå´®…J\˜L˜>KšñÂרWò•5A)êìêHóC+³Ï~³Ûý©Ú—ÄJ'Ìrn¦òb¾i¿È=KŠ‡ÕÙ;J™é:àg'•ê\©¡Ô&UÙ,¹x)57ÀtjGûÐ|FUíå鎃ýpˆaœ€j¥Ðò1 ïïSðþž¾ î½û{ûÆ)Lµ-ï«]ß„ !E—R R¹‡:1Tú8eÒUå +B#º–Ñ:ÿ¤Š‘.3ÍË0zR¥²2^—Á§^ð²­0‹‹ZÎRø¾‘U©­J120÷idf™û•šŒ†û­âUgò~óq‰Ìßkžm^ÃßiîšmF-0ægàQvº ½ðÀ šqÑ“3Š®š.f.amš¾D¿¯ª(þH)És™ÜQü~!t¿R~<ôJÍïÎËùaUûD-¶–×ì¯ÕQ±Ð*eñ”F-VO b%º8 Zĵâ"ýH[`-JitÂþ‘oÚGcŠaÝ.z}>ÁÂÙýµõ>¯©–48|*}Nå¯òx@p?|6AðÕªÚÇgœE-µ*]£pee|¾ øžKY.âVÄÀ J€[}Á÷º À½#'wÊò ?{/C6á'!I|FÞg­‘­{ø+rI žŸÂ…ñ¢xa¾˜¿±èMÍû¬ü•+uaœD²œÜ…aÙ$êL‡z:Ó hš·QTûEmùµé¥ðéGI'1Í›ho¨oŒ5ŠF£äñE#F6;*jŒ5ÔÛÅFÖÃ)Á¦O‹F|ô¥ËvÚl²p‚ÿQOû±e²l~´}íêǶ½}|÷ÖGº…ª«ÊÊ­§Ú–<=rz9ýÒìÀF.·äñŽ%ÃOËÁºu§ÛêžÛv +oÚÖ«¬ÚUѼ~n|_ÛÚWÿô×õ«!w(FjÏp‰¨ +~c96™1›×7¼YÁøH +œB¥´,RhÀXª*)Ab»õ³€=,v•pØ]†­HK*9«ã¬î*gÜígXnª¬*?Ÿó¬ïTé½J ù>¾ÏR•,f7—üŠHÄUˆƒBá¢:ñé„lO¨’ÛOrä¾ähNÅM缜ÈyÍ*U9–+þ^ROóTf6™”?Oo&Ӓͯ)›_\¤g´(~7#Ïà»Ïææ=$£;I*!¥tŒdB°QŒäñcH‚ß'yØ( )‰AJ$‘¥Ö^:²fûAuî³}?;…Ã/V rpË꾋6.KŽû ‡g»¶¬:¶çÕ¹Kã;ñ9¡Œ+b}ßÜoÜ‹ëOŸø1ľO›¦j v”‚¸Jñ‹ÛQû3‰h‘¬Ee¥´<ÒoÁ¥¶‹Áv÷@ vYУJŸTÊ,lÀR` T:Ý6§Ó]žï ÜN^LÙUÜz¶ omQñŠqz¨@ÅÒ…àSEN¥<ê$áL4EtŠèTÉ“ª"N¥Ò•åŠ‹í‘°óˆ“r–†œ¢3d~öÅEݘmH%ß©T¶8d1;§Ó\¯õwMÍNýºèA=t/ÃgCÏß QOêKžøiƒDr¾8.zŠãdx® .*Ùþí‡VÃ$-Ùv‰ÓÆ\—@3åz¨±8¦ç‡âwC‰¦`°)ÑxÍVÌñ„Ô6ض,)ý¡«²ÜÞn°'BÁ¦¦`(1·{¶u…•·ñÕ=âðŠh×»On¯°Wäû‘þþ7XÞëþªÃÿDAó]S©Iÿ§øÛJõ"¡WǾ~gÆ0{Ðó`>±°î¹vô¸91cøfy:ƒ}øOkÞߌqòv! ˜Ã(ú=ˆ} BÇŒ£è¨1Ž6ãó˜ÝAjTs3{ [Ñ[Òþ ²صSqíM˜ÿ<`Àh(€Ç{Ó€5€&°yPk¼ x‡P_cÑ&æ¶vðaJ®k—a|ðÃutøwaÿIúö†aƒvƒÙ­½eÕ.Âø:蟇yï%k|ëY™Ýèð·˜ÛÁ9¾ù³ SÁn†v ŽŠ£[´Ck ‡ÐRi_P£xì€}ˆÈ¨BÅç~ úÀÁ¦ø_ƒÜãµ°¾DæšaŽh5¬‚u3 ï%r˜»Î#ß*`t×étˆj@ºAûÓ‹l¹sÿŠœ›œyþLºÿYŸþ°.Y[yYÿàoÿŸ>º Pg™¡ÞC¯3µè)Í]0ÚÐöÏ÷Qüs€…Ù‚JY‡öSðq¥á,ŠOð-ÀØÅœÔnÒw:Ùø2:ò•TÔXSßGSF/Úç­†ý ¤N nGõZØ¢Çêdþ®]1Ὤ›sq:IbÃBÕ`ƒ½>?2Ìnœ||ü„øû×Ḃ ïð†¹Ó°N!ÔÞw5p®=Yh·¡†o‚ìQ˜ç„ÞúAnŸ›Ñ›¤öF.?ó¸5=ö£hp p|ñý›öj ®ªºÂëœ}ι—€€bK D””×TH CˆJ¤)EÊ1 ÏRA,PJQ‚@ ò²‰26(µ„—$tl Vƒ£N[T(ƒRŠÒpW¿µÏ9—p(öÇ7kß}÷cí}ÖþÖ·€}À|Ì;Û ýðÃx±Xñ½$^çÜØä‰ ÄûŸÐß[|×g@|KŒ¹ïƘgN¢Í@!°Ø!Úäág£ß‹Ä¬øé­}VbKbÆ·^l4Ë/õ9%®<«ßÞ JÖ>àì[Q‹w'±¯íi¼i±%4HbVÖŒÚÍéòåMD­ç¼OðÆnmOS®ëé¾õï"j—ó.ü7߉£ V/Ä~%Þ@2Ūóà ã¸Ã©4XÞ±UBëÌ%tgè u÷†µÖìA¨Î˜‚õöâ>«­ZZ »Æª3;Zu†m—ói묱×.7H»© Â+VÐø¿[íÿo`µËiڷ똭:zArDèŒÑèà[ôW ûÂ÷kÂFe(‡Z#n.…V¥Ùˆ9äJë;š¿ÑŸãâÁ lëW4ÃØkܦrŒ$§œžT9x£ØËì´h¹±–æÛ&±äY?^öp¾ð®oõÛ¯zvôµ¿¹Zrƒð³äáh¯ür4._@9v5>q×(>_4Ó¨G0.ƒVr‹ð»äì?ûoÀZ[åüšÁq‘ÂsxóCýñA_fì?üVóp-åùïw~ +ÿ ñx,¤±N.Q½i¨æ£4Î>BtòrªUÁÛ5—á=ù¹TçÑ:.ŠæÑ>ïòÔ|³ß’÷©ó&ò§½Ñhcÿâ5¯Ì ßëw(oð¥a¯õ*8·ç ¯‡ê îE¿ú’òôR{5ó,^%9QM¥D?äBÕúé¹K8ú„¼½ ¹Â[OµK“ÐÎxÚ§¹ Ob„Zú|,ß>TÀo‡Æñ~g"UÛÃqž ôœ¥VßA%Wë{¹qœ"wÉ+U=G0æ2§€ßÒ÷;j|:7‹¦ÀšN!•èû9‹è‹ðh>)°'Ólç2öÁ^v_ä’L®²3¹Xs«ƒ÷œ³=r[ ê+qšÎ¬Úóa?«*JTóy«Ï[pw]¼þdá}Ñ$¢7DCØ;$÷s¥žótZ e¬dÄeMR[€¥ÔÊÞ-RÉ‹µV¨£{•Åej>ô«OD#äè÷RÈÛì +ÍÍ]´ØCÞ>¾Çpi.¸¤èy~Å2©'håûD< xÈò~ï÷pÀ…qÄcˆ¬i^¤³hç#^'RÏ ¯·è@õ>שgù YÌ Ô(Ú­ó{æÝ´Ç Ãwø²zŸÆ_QµZHûÔ è¦iT£ñIu€O˜Íi°™ÎÔo¨@-âZõ4 SOa½•tP½ÄçÔ +.R%ˆÑ¯i¿z——X}hÕk jãçTjþƒJ!Ôûõ×ë/¤"¬«±q‚y¡}õÑÔç f"µðüÍ»Æ_ñÕ÷Ó÷Ñ÷or™çŸœ[ÖÕó0ÆHƒq‡Ÿ‰®ŒÀ7É^ל•î ƒƒÆR:þoKtå<ðÚÅ[œD{.ðK´üxx +ã.¸ßÚã÷ÃV[šëñL!Æß¾'Ì»ògün‡v´kï5|;è‡ö¿ô7Ør€V˜ƒq,{õðú¾ÂøõÀÇho‚}Ìík؉ömžý° +Xt×ú5 KþöºùèÛÚ@êÌ)·d³¾•½6÷xßÿ›¬—[ò›Xïüs4òç¦9Ï·ˆŸªÆn~^n>>‰Zpªæ5É' ×ò§Â¥ÂgÂ¥Âgö"/ï#–S¼ïÞÖëR¯ª€4Wj5;›R5·£í[pu/Osô ëô•1”¤kºJ +c¿ÏU==hý-R«kËͨ[êún˜Ÿ½o°ä,'‹–Ú¹‘ZÉgºö˜„9“qC±®Ô}Z›Sè‡XÉ™fG¬“Êã\‹<ƒj‘7ç!Olm ãV¬|O×>Æcã8ÿkÀ9¬ÓAU WÜÎ;ÍÕÐ!ßãvÆ:Údö£MF„bÀ›ëbÒø@ø$%Pr˜¸Êy•«,pKøs® q“ªk×xÿûâÊþé·iÑXS£šË;s]ÿà×É÷õç…Vá[Õp__}ÓóbížëÄÜuë…&uÁµõÃÖyè‰{xPT+@Íøê4¥}üdåbÝþ¤ÊóSJŠ‘cªÕQ _1³#5,Äuväîî¢Þ e‚Õzä ô×rº±ÑF¬²©Ü¾?P¥Sx µ uîJMÃú{È‘3ZwÉ9"ë1v·j§ßbŠ?ÇiAÉN>ƒsü9²µJ´DJ­|Ôem >B\d¡®Œük£¶†ú +̇)1ñœCacê|# :¸(vu³q\#…ÞþŠš.HÞ˜efÈÕÍÆd¼£ÍˆÁ1õ¸‹š©F/ô_ô°­€R- ÛEeF|zrÍLh[ì£ZÓê 0~‚ Ôe´Ãša̲FÑê +sÅv ýbƒðúÛ~±™A ?ó:~ÜhÜü¸QRèOúøq£u;þN7ñ/;ôgß‚7ºçÎA ¿óMüx$ô?ôj° : +zO¯ž†…¶¦xØ+°Í€"WÇñwŒ7¸˜ àwd/°ù*xú_CûìµýW°f?€í làCäyôAòƒs¿»¢Ó #£¼y-€´¡W#Ý€ðÈœbw?=_ö¾ëázÂ;ß9´¥®XëúÎБK°ù˜s×Õ3è;‘úb9~;žß8? pÁЖ\Ìø0˜éÝQæöÖ/<xÈå.Å[í¯À‡ðÔŽ£7;h¢X«³æÜöÖ³¤¼‘®Ü ßUêRõ&o·:ÒPµýh—•=ÅW—.Z[1>™^rºCÿ,àÍàížõÈc=Áõ¥4[爎àî]ÔBö°Z7‹)|ºEól*þOåM’äîÔ9¹ Y,tE•‚ßfŸäPôËDÔŸ/r‘•Ë«ÃÉÐ/\í|Ä3íX^Üì9Þºœ]ϹÈW)öF:f]¢iÑü—o5ºú6|z'Z2“µÖÓ½²Ÿ3[rg¤wtoOk¡Îyw9Øo1‘tŽm@=Ñ0D|Ö¹MêSW3q;ø´ yª þDàÏ–¨Vdšš-Ý9¼~™º„;òA{)ø<—V`¯4wO¾€<½Rr¤^¿/ü÷4¤³÷ó8¾½Ð¥­íþûÕ\Uu…×yÝÜ›\ÀQŠ\‹cAf`ÂS¡ø/µµ!B(*eH†—”°0°©-´0¥ +”" ÄR*)¥ÒBA¤ÃÝýÖÚûÜÜ„˜ŒÓþ<É|w}g½×Þ{ýøö.g¶Ú¨u©ÚãëÓDŒÍªÊk­6¸C¬#uu¯£šÂhU¿ 3ž£bq~ú㯱utG_ºÏy•šyãèëŽ;¬ß§[ø¶“¿4˜5A¨˜{ëa{Ò%¯Ís3é4ø¼”Þj—÷ ÚÈúÌ[‚1A¯ñóJèHÊÓÔß÷-µÜ|§e$:Q‚¢×¿ú¾^ñmðAw[PÝꆾÒY½·ÄyŸØ§ËÜÞ{þ~°ßéõL¹¨‡½,¾ûŽ…5M?1Ø®a š°/)ÏÕ¢ékÛûÍ\ã<¾Çã¯c›Ôyvê¿»‰&ägJòÅZ[›Ðþž¬×ªµþ3æU´TyBÏú:ºŽ…Ý‹÷Ík[5BÛ¸’¹†y\×úºº¶U_˜gçËôk’ž>VsG[K_׶*’Ð×Ùš»Íb=êñŒ…mñ\¯£Û3uîÐóç-B~°‡Òv¶¡Í„±ñøµ5wÃ= ktyBÛ׶s1Æ‹Ü/®+åø~ûpüM·;Új<ïo0öýÁ2/Ö$Cô}=ð¶P%°*tŠæ»|ËwņZL•À²”wi°+ÉgȽ@#'‰'à”QÃý1Ív%Ùã}ϸhópÝo/t ý½†þ¾ ýß¼>´4tå5Žóý£!„–¢¥t>¥ +íT¡Îu´s]ìq†Ÿ÷D.M^üñùýM´oâþÏßñ¦úOChì»ü¿ÆÝPß“¡ï6ÔßØ°ÑÚ}漡ßÍ †šQ5¾å  š÷'”?¯ae'ÍG ô3ï[€s’.Õ(ÛÆ ûŽypJ• +Ì3¯EFȱ܃6·ÐJ´÷+ kB½ù¹f¹¼7…®Ã¢Ö s'«r¯ÐnÞßfï OSÕØ Æò>Ê{M˜Ô^¬ñ­î>*ªÑ|OÖú:¼_ïw”=©Ÿ±;ÜGÕyÆ{ˆ£ŸÙ€ ¬Òûñn2,î…ÖZ“÷Tø®8Õ&`8ÎCêLÀÙ¶ÃèdÖÒ¬‡[i¿ß¯xª¿÷:Yj‰»T-aÝàä@ƒ¼‰oþ5u€­ÀÀD'…:8[ðÜM}ì~¤¡ ¯…ߣ|ƒ•ØCGÑ7ðf0u®=)ÝI¥Îezn¢7›^Bnã|‡aoXäôTÿ²«Õ'º[¤v;P§+úæÀrŒ´ÌY¯ö8¹”ï¾AùÎ3ôšó4pè/X wÖþ“ºe· ×ì]ôSç*ÍwºATÐ˘ k´À9‚2Ûh+ëgü-誷z¹¯[Ã,è™´ÀΠ‰ö,ÄjI¯¯Xèy`ª=”z@o¼eTßx÷uØOÏ?ö·€‹†oC¾ò8/¾^õ^P¿DÝùÞ^èïr­gC…ÔsáN_šií õÀ?QçG œ/s­YVª*×{œ°/`Ïüî—#ä @€ @€ @€ @€ @€õÂ"Š\¤«ôÍ#lº‹ºÐH"÷´=†\<=å\¯Cü7Q~™[tŸ<1·)J†;4•Þ0Ü¥‡h“áÝK4a¸Î¿æ:ÿšëük®ó¯¹Î¿æ:ÿšëü3Oå1Fr Ç#OOƒ?=2Çp—²#Ë…GyFEîRVä”ð¦ð‡SS w)3µ™ð»8~jOÃ?u ðæâŸn8ûç +OOÊazRï–òïÎå«„ß#þ¿Îþ/„·ä8i†#NZwá­¹|Ús†£|Úám’Úm“Ôn[‰³ÚpŽ£¿Ýç„áçœðì63þhá8Nô1Ã':šy8)ÿá¤ü‡“ÆNW4©|4©|4é»Dýïò3ŠQ7Ì€,ê–K¨v8•R PN/`e³'Oeàü›±”èŒ7ýi2þcô$|ãQ¿œ¦ËS!l!J¿(™‹÷SÄ£oÁΔR¥ðå!ÒPÔËCœbÊ¿£õÞÔժݛFI»ÓMcÔ­gá7F™ˆÍeÊðf:P„zí“¢ëØ5‘§§¨S¾å6Ú›by“”K¶ +P†ß•Ñ$ø¸Õ¯žiŽZ"u½‘x*Æç6†^–KÙBÓr ¼]$BLbOŒÄÐÿRš·Ü¯b)ݹžä&ØéÃLi{<žG ïE’~Û)Ñr ²SˆZzr<Ô¬—š)§Òïz´Z…îÀ˧æ!Fú*Çö“þ2@Ê ,Cd¨ “á2BFÊ(-cd¬Œ“ñò®¼'ïË™(“d²L‘©2M¦ƒ2SfÉÐdŽÌ•y‘²P>”äcùD>•ÏB]ûÉbùR¾’¯åY*Ëd¹|++d¥¬’Õ²F¾“µ²)eƒlD*¤–ͲE¶Ê6Ù.;d§ì’ݲGöÊò£ì“ýr@Ê!9,G䨓ãH#'䤜’Óò‹œQ«èœœ— rQ.Ée¹"W•_®Ë¯Êå7ä¦ÜBZ¤Cz¹-wä®Ü“ûò@Ê_ò·<’Çò‚ ȈLÈŒ,ÈŠlÈŽȉ\x¹‘y‘ùQ¡£ +£ŠâUÃk(Žxo $J!¥QeQåQñ&*¡2ª *ª¡:j &j¡6ê .ê¡>à-4D#4F4E34G ´D+´F´E;´G¼Žè„Îè‚wÐÝÐ=`CìpÀ@8z"&z!NDÁ Ñè 7<ð‡>ˆA_Ä¢úcbc†b†cFbFc ÆbÆã]¼‡÷11 “1S1 Ó131 `6æ`.æa>`!>ÄGøŸàS|†Ïña1–àK|…¯ñ –b–ã[¬ÀJ¬Âj¬ÁwX‹uX Øˆï± ›±[± Û±;± »±{ñ~Ä>ìÇü„ƒ8„Ã8‚£8†ãø'p§p¿à ÎâÎã.â.ã +®â®ãWü†¸‰[øàOÜÆÜÅ=ÜÇ<Ä_øðO(I!LÁ”LÅÔLôLÇôÌÀŒÌÄÌ̬ÌÆìÌÁœÌÅW˜›yd>ó2ó³ Rg!få«,Æ×Xœ%d_ç,ÉR ei–aY–cyV`E¾ÉJ¬Ì*¬Êj¬Î¬ÉZ¬Í:¬Ëz¬Ï|‹ ولMÙŒÍÙ‚-ÙŠ­Ù†mÙŽíÙo³#;±3»ðve7vgÚF;4ΞŒ É^Œ¤“QtÑb4{ÓM½ô±cØ—±ìÇþÀÄÁ¡ÆáÁ‘ÅÑñÇñ|—ïñ}NàDNâdNáTNãtÎÙœÉYü€³9‡s9ó¹€ ù!?âÇü„Ÿò3~Î/¸ˆ‹¹„_ò+~Ío¸”˸œßrWrWs ¿ãZ®ãznàF~ÏMÜÌ-ÜÊmÜÎÜÉ]ÜÍ=ÜËø#÷q?ð'ä!æå1çÏ<Á“<ÅÓü…gx–çxžx‘—x™Wx•×x¿ò7ÞàMÞâïüƒò6ïð®¬ç=Þç>ä_ü›ø˜O4Ñ QÓ´-…–RK¥¥ÖÒhiµtZz-ƒ–QˤeÖ²hYµlZv-Ç?Ö‡ÐêÅ—mÛæòÖ²mÛ¶mÛ-kqi-Û¶mÛÝž{~ÅALÄBlÄA\ÄC|$@B$Bb$AR$Cr¤@J¤Bj¤AZ¤Czd@FdBfdAVdCvä@NäÂoÈ<È‹|È(ˆßñ +áOFE1G ”D)”F”E9”GTD%TFTE5TG ÔD-ÔFÔE=ÔG4D#4F4E34G ´D+´F´E;´GtD'tFtE7tGôD/ôFôE?ôÇ Ä Æ Å0 ÇŒÄ(ŒÆŒÅ8ŒÇLÄ$LÆLÅ4LÇ ÌÄ,ÌÆÌż_žXˆEXŒ%XŠeXŽXùkÇ«°k°ë°þ×”7`#6a3¶ü:óÖ_{Þ†íØØ…Ý؃½Ø‡ý8€ƒ8„Ã8‚£8†ã8“8…Ó8ƒ³8‡ó¸€‹¸„˸‚«¸†ë¸›¸…Û¸ƒ»¸‡ûx€‡x„Çx‚§x†çx—x…×xƒ·x‡÷ø€ø„Ïø‚¯ø†ïøŸ Á EÍ€¡†aŽá‰‘…Qу1‹±ç׸â1>0!11“0)“19S0%S15Ó0-Ó1=30#313³0+³1;s0'sýZZnæa^æc~`ÁÿϭЯÁfe1g –d)–f–e9–gVd%VfVe5Vg Öd-ÖfÖe=Ög6d#6f6e36g ¶d+¶f¶e;¶gvd'vfve7vgöd/öföe?öçä æå0çŽä(ŽæŽå8ŽçNä$NæNå4Nç Îä,ÎæÎå<þÅù\À…\ÄÅ\Â¥\Æå\Á•ü›«¸šk¸–븞ÿp7r7s ÿåVþÇmÜÎÜÉ]ÜÍ=ÜË}ÜÏ<ÈC<Ì#<Êc<Î<ÉS<Í3<Ës<Ï ¼ÈK¼Ì+¼Êk¼Î¼É[¼Í;¼Ë{¼Ï|ÈG|Ì'|Êg|Î|ÉW|Í7|Ëw|ÏüÈOüÌ/üÊoüÎü© +©P‚(É +ZaVá^Q‘YQUÑ]1S±[qWñ_ ”P‰”XI”TÉ”\)”R©”Zi”Vé”^”Q™”YY”UÙ”]9”S¹ô›r+ò*Ÿò«€ +êwý¡BúS…UDEULÅUB%UJ¥UFeUNåUAUI•UEUUMÕUC5UKµUGuUOõÕ@ ÕHÕDMÕLÍÕB-ÕJ­ÕFmÕNíÕAÕIÕE]ÕMÝÕC=ÕK½ÕG}ÕOý5@5Hƒ5DC5LÃ5B#5J£5Fc5Nã5A5I“5ES5MÓ5C35K³5Gs5Oi¾h¡i±–h©–i¹Vh¥þÖ*­Ö­Õ:­×?Ú Ú¤ÍÚ¢µUÿi›¶k‡vj—vköjŸöë€ê눎꘎ë„Nê”NëŒÎêœÎë‚.ê’.늮ꚮë†nê–nëŽîêžîëê‘뉞Ꙟë…^ê•^ëÞêÞëƒ>ê“>닾꛾ë‡~:„C:”aZ¶‡v‡u8‡wGt$GvGu4Gw Çt,ÇvÇu<Çw't"'v'u2'w +§t*§v§u:§wgt&gvgu6gwçt.ÿæÜÎã¼Îçü.à‚þݸÿtaqQsq—pI—ri—qY—syWpEWreWqUWsu×pM×rm×q]×s}7pC7rc7qS7ss·pK·rk·q[·s{wpGwrgwqWwsw÷pO÷ro÷q_÷sð@ò`ñPópðHòhñXóxOðDOòdOñTOótÏðLÏòlÏñ\Ïó_žï^èE^ì%^êe^î^鿽ʫ½Æk½Îëý7x£7y³·ø_oõÞæíÞáÞåÝÞã½Þçý>àƒ>äÃ>â£>æã>á“>åÓ>ã³>çó¾à‹¾ä˾⫾æë¾á›¾å۾㻾çû~à‡~äÇ~â§~æç~á—~å×~ã·~ç÷þàþäÏþâ¯þæïþáŸAˆ d*@À@ƒ „ Âá‚ðA„ b)ˆD ¢Ñ‚èAŒ f+ˆÄ âñ‚øA‚ að?&Ê;¾©*ŠãJ›ü~¿2Ü{€¨ *¤$µ--ÊÈp0”×äµ }MBFÙâÜ(‚  ÙNâ·L'pïæ]ýðÏýžsïïœ_>÷ž¼–h…SЧâ4œŽ6h‹3Ðgâ,œö耎ðÁŠÑ A„P‚Îè‚sÐÝÐçâ<œR”¡=PJôD/ôF\€*\ˆ¾¸ýp1úcb.Á¥¸ —c0†`(†á +\‰á°P0"°QƒZDÃÔÁA=âH ‰‘H! ²hÀ(ŒÆŒÅ8ŒÇ\…‰¸×àZ\‡ë1 “17àFÜ„›q nÅm˜ŠÛq¦áNLÇ Ü…™˜…»qfcîÅ\܇ûñÄ<<„ùX€‡±‹°K°Ë°àQ<†ÇñžÄ +¬ÄSX…§ñ žÅsXç±/àE¼„—ñ +Öb^ÅkxoàM¼…·ñÖc6b6c ÞÅ{xàC|„­Ø†íØØ…Ý؃ñ >ÅgØ‹Ïñ¾ÄW؇ýøßà[|‡ïñ~ÄOø¿àWü†ßñþÄ_øÿð Ì&,`!=ô$Å"6e36g ÂCyç<’GñhÃcyç <‘'ñd¶d+žÂÖ<•§ñt¶a[žÁv<“gñl¶gv¤~XÌN 2ÄvfžÃ®ìÆî<—çñ|–²ŒåìÁ +V²'{±7ûðVñBöåEìÇ‹ÙŸ8ƒx /åe¼œƒ9„C9ŒWðJ§Åj†¡ÍÖ2ÊG°Žëg‚IŽdŠif˜eGq4Çp,Çq<'ð*NäÕ¼†×ò:^ÏIœÌ)¼7ò&ÞÌ[x+oãTÞÎ;8wr:gð.Îä,ÞÍ{8›sx/çò>ÞÏø çñ!Îç>Ì…\ÄÅ\Â¥\Æå|„ò1>Î'ø$Wp%Ÿâ*>Ígø,Ÿãj>Ï5|/ò%¾ÌW¸–ëø*_ãë|ƒoò-¾Íw¸ž¸‘›¸™[ø.ßãûü€ò#nå6nçîä.îæ~ÌOø)?ã^~Î/ø%¿â>îç×ü†ßò;~Ïø#âÏü…¿ò7þÎ?ø'ÿâßüGé`5Q +å‘W%©©š©¹ZèªÃt¸ŽÐ‘:JGë«ãt¼NЉ:I'«¥ZéµÖ©:M§«Úê µÓ™:Kg«½:¨£|ò+ buRP!•¨³ºèuU7u×¹:Oç«Te*WU¨R=ÕK½ÕG¨Jª¯.R?]¬þ ¤Kt©.Óå¬!ªaºBWj¸,U+¬ˆlÕ¨VQÅ4BurT¯¸Jj¤RJ+£¬4J£5Fc5Nã5AWi¢®Ö5ºV×ézMÒdMÑ ºQ7éfÝ¢[u›¦êvÝ¡iºSÓ5Cwi¦fénÝ£Ùš£{5W÷é~= 5Oi¾èa-Ô"-Ö-Õ2-×#zTéq=¡'µB+õ”Véi=£gõœVëy­Ñ zQ/ée½¢µZ§Wõš^×zSoém½£õÚ Ú¤ÍÚ¢wõžÞ×úPi«¶i»vh§vi·öèc}¢Oõ™öês}¡/õ•öi¿¾F6óùÊ‚ɘ¿1ø¡<¾PaE6•hL*Ë|~Ã@¾ PÞÅþÇ€aqž~¿ßÍK+YëŒIFÁoiuÊn°½V(MÔ&âv,—Þa+•ˆ{#y˜&>T•mT®ÊvQe…³Û븨r7< +«"‰L¡“[¼}Ý㸋¾®*~€OÐgè÷öwE)ý]mÊÅw3‡[ +{º¢LžA©X¼Ö“m\]E‰ñ(ñ{»}ÆЧ¤Ì°ÜÓÓª¯·<£vÆ*èy,'µ<ÛÉX°“阓랎ÕÖ[+[ŒÆšgã;•'Rv¤Ú‘=:ìXõ¹(ÿŠ¾’JoÒNçšÉJ¥£»&ƒ|”M噊ÕF3îa$1*îFÕ‰LTF‰7û?ªNÛE‰T&Úø–Ó,Ïäœíp&–ˆ7µGfc –cÇö'šÈ¦íæ¹wvµ±°åÄ™¢FqmÊr2ÉÿÃêLþG}>Àa¹K¿Ù÷w2ìlhÎf?`òb“›–›/+lç£áÿ¤ {kìúÜh6.žt2g^v²Õž¨må,"1«>4­Ï¦Í°Ø-ˆs³Ü¤wŸ&±MãV2‘Τɨí­ÅóÿF;œ+u¿\>÷ý~s½¥æÚK͵•šk+¯,ú—í²kŽÛ¶ÂðŒõYʸéïÈì9 à’ÁFé´Ê4ѸéÝZÞHÛÈZWÒÖq~}IâEct!½8ü8ϱÄ{¸û¸¹OõŸ´Ö×ÓV°Nz¦Ùì?üz·ùýìúyœ^ÐU‹v'‘û6IcÚÞ6i{KXƒÚ³¿.²Þ”Y•ÍAe[ÑêìûÅ}·ÏãÃïß­oÇÇOÛ»÷›ãí,'Tµ¥ª‹TÕ6UuÁvºMzöÃ"ûʪ*ª¯ìùo7›ÍýÝúþýöšƒìXéU[‰©Ñæ ÞßlÆ}ï’Rv”r™JÙ-P ¢¶g?-Jy\Lôçi¢OÓD¯ÒD÷i¢Wdß“ý*eߧìWLtÏDß.²ZŒYŒ??Æ]öæñu;íßó6=î•yxÒƤëͬg—wëñ…šÇ»çñézúñÞm¶ëã·ãf7Öö)ÉÛ´ÿJW½}¿ÝñÝ—'>¿i&5r£*Z¡­Ñµ¨C=Ú¢ЈI“… +Wà +\+pÙ2À¸Wà +\+= +_à³µ¶VÃÖj¾ÂWø +_á+|…¯ð¾ÂWø +_áWð+øü +~¿‚Ϧk*øü +~Ÿý×°ÿš +~¿‚oàø¾oà³ßÀ7ð |ßÀ7ð |ßÀ¯áÖpk¸5Ü.N`j¸5Ün ·†[íáÕð<óõÉûW’{‚”¿îV¨ ŠV¨Ak4ßgQ‡z´CÚ£MuÕn€à¸=ç{Î÷œïóyêê©«§®žºzêê©«oQê멯§¾žúzê‹ð#ü?Âð#ü?Âð#ü ï@Þ¼y‡Ô¯ÚöŒùMφq;­×bì‡éù¿y‡¼C—Þ÷Q[´CÚ£’VÜWñ¡›ÞÓQ+Ô 5Ú …_Á­àVp+xŽcÈoÈoț޿QÉoÈoÈoÈoÈo˜—Éœô ²âSjµêш¦yC8º¼Ý=ÜíæÿWóÿýôÿ››»Ïommò Îƒ&l¸o"~$øŽÄÙ›³ún}÷Ä1Á±ŽEp,‚c‹àXÇ"8Á±ŽEp,‚Ë5”ãG@/(.ÏX<<g.þš"¶EìŠØ/âfÒ"nŠØ±+b¶ˆçl‘Û¾pÞ±+âeîåµ®ÈíŠÜî…ë]/s/Ïù"·/rû"·áþeîü¼›¼Eܱ-bWÄ9wSäz)¶EìŠxY›-î·Em¶Èg‹|¶¨my­+r»"·{ázWÄËÜËs¾Èí‹Ü¾Èí_¸™;?[¬WŽ›"¶EìŠ8ç¶E®æ…¸)b[Ä®ˆ—¹—÷–¬—λ"^ÎÛ÷»â~WÔæŠ|®¨myι}‘Û¹ý ÷/sçº]±^9nŠØ±+âœÛ¹š⦈íÿÙ®›·­$ÃëÌ"ûþ,I6û6Ï ÆA 'ßýÈæ£@ ¼*ÇŸÔoSKU·ór;¿Ï~ï|›=ÿàïóí¼ÜÎï³ß_{¿Ž;ûG¯ÿL×Ûû×Ûû×Ûû×Ûnëm·×ìõv¿^çévžoçåv~Í^o³¦œ§Ûy¾—Ûù}öû{çÛìùŸoçåv~ŸýþÚå6{¹Í^~ðúåv~Ÿýþ·ûgt¿®û.ßßÿ=w„~zi襡—†^×ë7ºÓƒ&=é•£B }4ôÑÐGC }4ôÑÐGC }4ôÑÐGC }4ôÑÐGC_/ }4ôÑÐGC }4ôÑÐGcÄÕ?Cÿ ý3ôÏÐ?Cÿ ý3ôÏüÀüÀüÀüÀüÀüÀoø ¿á7ü†ßð~Ãoø ¿á7ü†ßð~Ã×sCÏ =7ôÜÐsCÏ =7ôÜÐsCÏ =7ôÜÐsCÏ =7ôÜÐcC =5ô„ÐBO=!ô„ÐBO=!Ò^i¯|͵WÚ+í¥W„^zEè¡W„^zEè¡W„^¯çV¯½"ôŠÐ+B¯ˆÿÄ/üÂ/üÂ/ü¯‹ßzÍmݹ;»ßÍýnîws¿›ûÝÜïæ~7÷»¹ßÍýnîws¿›ûݦï¤×uµÆŸñg|߇æûÐfüÆŸñ}_ÚŒ?ãÏùÏ_¿üöéÓoÿõÇ_¿øò?ÿnŸÙ>‹}û,öYì³Øg±ÏbŸÅ‹={,öXì±øÜw5w5w5o5oõúÕë7{nöܼóþÍ^›Ïi3o3o³ßf¿Í~›ývóvóvóvóv¯ß_¯·ßn¿Ã~‡ýŸãás<Ì?Ì?Ì?Ì?ì{Ø÷°ïaßÿÀ?Ο~ùôõó¯Ãä—«åKmÀ1Çh£qŒÆ1Çh¢qˆÆ!‡h¢qˆÆ!‡h¢qˆÆ!‡h¢qˆÆGh¡q„ÆGh¡q„ÆGh¡q„V>Ѻ>ÑÎúc # Úh§éB_ó6ºÓƒ&=éµ_qWÜדÐ= }Å]q=Ý“ÑWÜwÅ]q=9Ý“Ó=9Ý“Ó7\OP÷uOP÷uOP÷uOP÷uOP÷õ wÃ=Ú÷$÷LlÏÿ?üðùí<¼ãçåí<|Kwoçãvn¯×ãçuŸŸ:Бm´Ó‰Ît¡+ÝèNšô¤ué€?àøþ€?àøþ€?àøþ€?àøþˆ?âø#þˆ?âø#þˆ?âø#þˆ?âø#~à~à~à~à~à~à~à~à~Ãoø ¿á7ü†ßð~Ãoø ¿á7ü†ßð~Çïø¿ãwüŽßñ;~Çïø¿ãwüŽßñ;þ„?áOøþ„?áOøþ„?áOøþ„?áOøþŒ?ãÏø3þŒ?ãÏø3þŒ?ãÏø3þŒ?ãÏø3þbÞbÞbÞbÞbÞbÞbÞbÞbÞbÞêzV׳ºžÕõ¬®gu=+þŠ¿â¯ø+þåßCñ«âWůŠ_¿*~UüªøUñ«âWůêñšФ'½®¯øUñ«âWůŠ_¿*þTü©øSñ§âOÅŸŠ?*~Tü¨øQñ£âGÅŠ?*~Tü¨øQ/žëãGÅŠ?*~Tü¨øQñ£âGÅŠ?*~Tü¨øQñ£âGÅŠ?*~Tü¨øQñŸâ?ÅŠÿÿ)þSü¦øMñ›â7ÅoŠß¿)~S|¦øKõ×\ûó“â'ÅOŠŸ?)~Rü¤øIñ“â'ÅOŠŸ?)~Rü¤øIñ“â'ÅOŠŸ?)~Rü¤øIñ“â'ÅOŠŸ?)~R Þ‚·à-x Þ‚ÇoŠß¿)~Sü¦øMñ›Z^<×ËoŠß¿)~Sü¦øMñ›â7ÅoŠß¿)~S/¿YñWüÅßð7ü Ãßð7ü Ãßð7ü Ãßð7ü ÃßñwüÇßñwÜwÇÝqwÜwÇÝqwÜ÷À=pÜ÷À=pÜ÷À=pÜ÷À=pÜÄMÜÄMÜÄK¼ÄK¼ÄK¼ÄIœÄ9qNœçÄ9]ß龞¸'î‰{➸'îé:OüÿÄ/¼Â+¼Â)œÂ)œÂ)œÂ)œúÎ×÷ö©i§]èJ“¾æ|ßwœ.ÿûuÿJ®.¹ºäê§Æ?þýáËo?ÿç¿>ý÷Ï?¿ýãc¼~<ÆÇn©ëËüGzýè<µÓ«|JÔ¨D'­õþëÃ|jPóʼ2ïúPŸºÐ•nt§MjºøqýÈ}ýüë8öéïÿ›ÿþ?–:°ÜëÛ<>®oËøxpÁk~xÝõT?õõ:Vý`Õ.ú0ÿÁ²¯§ú©—{W ®~*rânÁý‚û÷‹×ß¹_p¿à~Á}÷m/÷[(nÇíö¹îòs—r?®\µqéÆûõ“õèWtyô£Ó‰¾î²o‘Èœ"sŠÌ)2§Èœ"sŠÌ)2§Èœ"sŠÌ)2§Èœ"sŠÌ)2§Èœ3þŒ/B§"tŠÐ)B§"tŠÐ)B§"tŠÐ)B§"tŠÐ)B§"tŠÐùÖ«56Ž« +ß;³Þ];žxã<¼‰›½k¯w’xÛópv×Åñìî]º¤y8EÞ&´¡! MSQ5é¦H4Ré"’‚@ŠâPÇy¬c“DBBâGÄ?$¤D%„Ô¡ò˹wÎN!øÃŽÏýνç»÷Ì\ß=ó-JhŽš£„æ(¡9JhŽš£„æ(¡9JhŽš£„æ(¡9JhŽš£„æ(¡9JhŽš£„æ(¡9JhŽš£„æ(¡9JhŽš£„æ(¡9JhŽš£„æ(¡9Jh^Äü(¥9JiŽRš£”æ(¥9JiŽRš£”æ(¥9JiŽRš£”æ(¥y ¿ïÃ(µòXòXòXòÈËc](âxÇ‹Íq¿^”f%²‡¬À³o8ð6Þý¹ÀkÖ8Û +<#ðÌÀ rA#Èa9îß•ä0‚fà r˜A3Èa9Ì ‡ä0ƒf#¨Õ¶ä°‚Và +r4Õ¨ªÒFUi7ÇQUÚ¨*mT•6ÖUëªuÕƺjc]µ±®ÚXWmTY¬—Y¬“~=²KÍúX@,"–‘‡õÔÆz:ì«7@ëiS½áO8^Âs?åx³®âO9Ž?å8þ”ããx¾ý<¦á? Ž ¿ž™%ÿí h šˆb“7Œ¸1‹˜CÌ#‹ˆ%³¸N×Éâ:~ý.d&æóˆ}• œŠ|pzòÍÓ3<4xFà™gžx÷9ð²— ¼ GðMkþÏì¾ |úÊ°„èלrkÈxsÏ9îî}÷Þç²9܃îÿ.,áÞ`ßhö9bqù¹ÙݤЮ.’!0ìm0¬¥q]]˜+—M§˜è­ß`ÎËÀÝ`Ö×çžXsîû="~¤¾J»Ä 5®+3ža9u€ s°Oõu>ö¤}| )ÑKùìÕÖÛ‹°O.Ømeæ¢ê¨= ÑæFôQsQ}Y\d$J.2'½×JÔá9÷>„÷çÒÖ +èŠÌÁ«êK°G§d»TŒ š1Û¹ÇløðN³WàÖmæR±Ä¸Õà´ê[ÍÎtiR’<Ós¼~«SP³£fç,8J¬Æ•éQ+žÞcÆÒºm†ÓýV;ä¯7þîô¥7Yí#ƒ–ùÍô¹ô•ôOÓ¡–ôfˆš9sõȆ‘܈OwÁ‚Ö§séТú’¸H:Jœë`âæÙ—˜²„ YðT¿Ÿcò±Oˆ‹0 ­dÙ©ÈLD™ +Ï„•ÞóÀo=?‰é´g½fo*³KLjW“ÚW“ÚóIíóIíÓIm*©ñ¤VÐèš#•íl׊–޻б­ƒ´^£÷È6¢©³°µ+ SVzºÍêÊ +Oˆz‰VX­„I"J!Ú6 BT ‰qJ Qè_IŠ~ +ðOïguúC¦Åš…•ô,ÑÅ,ú’ iÀï“iÙÿ1$~ñ;^ê)˜öm…Vú-’I %“<çé~Ê3ža…eôäÃO’>I+Ã8†ÓR^â4[¤=$¡ˆ.¹ ?ÏîÁü´ÇþbÕ£Ôcî«+Óû­^§Ðû5ÄÞòØzÎö+ã{×xý\¯+ôû™~ƒÝH×C@¼lHâ;º\ä|ÚØǾ®ŸfoúkŸê“¤W`3§åìex¤£©;ì0,óxê¶Ï_ê±”¼ƒGîÊÞÜÀvK>¬‹…—³­Æçظ>ÍJÆ ¶%µ0¿Äò}wX6%s ¤äôþ<ÜɆÔ4[gL³G²‹ô'$BOeœH-òÅÈÁÈH%âDr‘Í‘M‘ÞHOdE´3‹.¶GÛ¢Ñh8Š*Q]QoÜv2¾+Â1áhCÒ)¢…F¼cUÈCÄ]®V”ÊDÑÍf*õHc—›ËTÜÖ{&g)ýJ•VÜëŸ%•Ï$Ý?M¤ê´mç£nKªHÝÎ +©ì.Æì*_®S²{²NbÆÉn·³4 /1êœ|½[`õäëÕ*YõÜX|¬s˲ü8ÿÍ~l3÷?ñÌ¿~âkݯU&&Ýsk«®)œÆÚjÅÝ:‘Ü;9¯W^(óy嘀êä<WŽ—w‰q:Ϋ Ô1 Á¡>æÓj$!hpºk’¶Ï§1˜ ´´A;K˜¤1zVÐà˜ Þì4+óYÆ$'t˜LKÎtè°ÏIKÎÝâ´ÄÈ]ɹÛ“éº$¥¯(FŸ Ìööa¶¯W†wÞ§üðq?|\†¿p?lùás~ø„3ÿ§ÏÅÿÅ(œ(ÒÊŽÉÙ()VK{}\;¼EžƒeGOt_¡¨¿ K2U·-Ut—¤Šdl,ž‰}Œ>nwÃ0ô{â/v_ ØrAo‡a C› +› +"ÇY„–Âp†â/>ØIÎb(ÃË œã 8—‡Ênÿ~€¯’xù ‡?„#ð9zôè‘#Ϙ OTÜÑNÎêzÙ]½ŸW3åøAþìy~Rqûÿ! &{&vv× @MÅÅÚ`}ÚÚ¥Ðl”@ÄÀJ´‹á⌠s‹A¦h3ƒtÇÿ{›ådÀµîVmc 5mãÿ?07n2)Žd,¹¨hÄŒb°¹ÅÚAD endstream endobj 44 0 obj [/Indexed/DeviceRGB 255 45 0 R] endobj 45 0 obj <>stream +8;X]O>EqN@%''O_@%e@?J;%+8(9e>X=MR6S?i^YgA3=].HDXF.R$lIL@"pJ+EP(%0 +b]6ajmNZn*!='OQZeQ^Y*,=]?C.B+\Ulg9dhD*"iC[;*=3`oP1[!S^)?1)IZ4dup` +E1r!/,*0[*9.aFIR2&b-C#soRZ7Dl%MLY\.?d>Mn +6%Q2oYfNRF$$+ON<+]RUJmC0InDZ4OTs0S!saG>GGKUlQ*Q?45:CI&4J'_2j$XKrcYp0n+Xl_nU*O( +l[$6Nn+Z_Nq0]s7hs]`XX1nZ8&94a\~> endstream endobj 6 0 obj <> endobj 7 0 obj <> endobj 8 0 obj <>stream +%!PS-Adobe-3.0 %%Creator: Adobe Illustrator(R) 13.0 %%AI8_CreatorVersion: 13.0.0 %%For: (pmehta) () %%Title: (RBM.ai) %%CreationDate: 2/16/18 10:55 AM %%BoundingBox: 174 237 697 389 %%HiResBoundingBox: 174.6948 237.6172 696.6953 388.9453 %%DocumentProcessColors: Cyan Magenta Black %AI5_FileFormat 9.0 %AI12_BuildNumber: 406 %AI3_ColorUsage: Color %AI7_ImageSettings: 0 %%CMYKProcessColor: 1 1 1 1 ([Registration]) %AI3_TemplateBox: 396.5 305.5 396.5 305.5 %AI3_TileBox: 108 -50 684 684 %AI3_DocumentPreview: None %AI5_ArtSize: 460 180 %AI5_RulerUnits: 2 %AI9_ColorModel: 2 %AI5_ArtFlags: 0 0 0 1 0 0 1 0 0 %AI5_TargetResolution: 800 %AI5_NumLayers: 1 %AI9_OpenToView: -46 612 1.31 1162 814 18 0 0 50 76 1 0 1 1 1 0 1 %AI5_OpenViewLayers: 7 %%PageOrigin:94 -376 %AI7_GridSettings: 72 8 72 8 1 0 0.8 0.8 0.8 0.9 0.9 0.9 %AI9_Flatten: 1 %AI12_CMSettings: 00.MS %%EndComments endstream endobj 18 0 obj <>stream +H‰ŒWÙ’ªÊýÿ¡ÕÆ TPAT +¡çÇvq[1âÆ}¸ß~ µ•Ý»÷9ûÁ¬Jr\¹2!0PÌ–¹ÂZpt¯q¡£—q¡›ŽÈYbR¥w`È+ïˆfùS$'žÏ›˜*»ÏEÓ^v')ciʆ&Áwl"ò—ˆ Ì£ƒtn3‹\Þ™v±…?’šï½'^_ 45|¨Œ·ÖæŒbÌò¨-ëáC†É[Ê\ +\_Áã@Ó|î›Ù/³êS-O3ñDž튯¿Ã×k™#FL¿ ¼0F+˜Õž ƒ2¡%ðtqP|yË퀢–k®8z*PùG‘”q­Ï@ÓXÃGB£Ÿ©Œ­“݉BæÔuaÒþHú~P±”v~«(Ñ}O»Òþ¿Š´wvÿ¦%ÚVvjUÚ³ŠùÒCÆÑ”öX¶(þ +7÷‚RwÀ½TðTÒ]…Á~Ý¥>¥;Œ¦KU]ÖñÒgW¿ðÉóÚnÊ-ßç+ýasºND¨B‡¸ŸåeúLGBÁ«æV6¸â^ž‡6WžžöG2Ú:~à ªþêè.òÕ\t(—JBTÚ‹û8k´ÂP¡ŒäAÓóUÐn@Žö´(“† #1Ø)!«³=öì#ûzåfÔDDä’êɹ0e±ûô½Ì +ĵއ"iMPÍs蔘"Ûö;\L6Zñin“Î&jã¡@M Ê°;æ7㢎²-b¸L`WÈëÆ£_ì$¤`@=n=¬z¡\¯´ v¢ ‘ï.Wb\Я^L¸%yr{¢³5ßÇS°-o _A‘àVl!_9š"!%þ{yw€Hm%p™rÕ\‡úž,òB n@îÿÑJ«,Å‚†ö£ +Ê_ÔAóØ*s] á’Éÿ¢So°wa|fÀl@“÷òò’¥íZÅäY+?ÌcùQáewu[e8BÕßm -”E¢&·fÄŠ©­j3 aÂP`D!Kcr¾/îÁqá?¤–Ü¢#ž¯=‰ÛÎH¹©O.IŒ_!¸`6S„'Ùú<”€V¼r¿aŒíìËâ¾ì»29é ”òœ³È7¬6†^ûbZó“å=%¡>ö¬¤ƒ'¹ZEÇát’ád±púüêÊìúâ˜íÆ…›jÔê umÇp¥Ó+üÄYV› Ì-¥òpÏÀiã~‡ŠÏaî +j¼ÐD«èeïÏå^a»ÜæÂy’œ](s³þ9Í×)³©ê‘¸zYþÂØËv=[í¡JoX¨Ä*ÉŸ”9úî·HíbLP ¡O((jõ–¥»•n{ö÷ ïzZJôU"ÉbÈbžÜ8öH"({i»­‹²ª7,¹@¡ OT±£©3Äc³¢ÖnH±|ªûËrñêÂäHS‹|u»Ë°9óÕãÇÜw=½º;4B4!ÊIÇÅ«Eªyá[?gwá)Ó®}lï¸ÓÊpøM÷Ke,h¥LeÁR%Ü°GµKô L8hI¢M±ýy ¾”p?¹’èn<ì¨þ³á˜ÏÒ,›ÊË"=úËSO(šó»´<Úô½ ½×AˆþÔ¯.?ŵ¨‡euŸëhV:«uŸNRÔ„éˆÿ¹&Ü°¡ÌIÁ¯,®ž=‘Ê\S©%ï; Êh±4ÊH}T4ác?ê¦y47¶8fÝu8 Œbv3 å¢Û Î=tOÆ)µVh2|¯=G±,¡ K©—Ñ4oCí±”É•½D‘aØ)Ã{GLª3ož®¤¢ìa˜Þ°lôbÄÁy¡RøŽ§0*ÁMp¥‚s¨ëpõv±ç~¦€”µ}§w®­|¼s«îyV íO¥1ÐÖ Ëöñ6I½€JÅÞŠæü­v²6æñò&ϬYð@·,ËQoQõhý ÕÚüwy—J½E«¦Õ\.̳¾ÔïG6:¸‰rÔesq=,–­Ô>¤¨‰´\mŽÚò?Ö×+jŸ;OÞø·(håT5™‘^}ikÅ3‰)‹g;ó(Å+!´xXqû)F >þ¼h<Ÿn©xV³$´ÿçwo3=Cu?ó¼áé&³v7ûœ›èŒQÜÄzÛŒÄËœÏäÝ>úoÜL¡]iš! ÓbÖv$ùÖ˵±t&ˆãŸÀ‡‚¬ˆûµ!(q¿ßié…¶¾ÿ³ÙIì¶=Ï9orôL²¿ýÏÌÎÎÀ»mùµ,œEOUêþžj_¬ù­Ï •çן·­ÙGQžÈчRx×Nd½(¬6¶ÕYÌgÜ•PbT¤h)ùÎgý'w^XTû¹Ø¢í}€ë”?ÁÊ1Ì#|ˆÊ1uuËH`ìQC9÷RœPõY§ › ~sQ/ÁLeîx¢X®P¥<Ͼ:Å$knÊ;³ŠÈ6¸-PLm¸£˜`ñój¶Ð&Õ±E§˜ûº•õ_“ñi£Ëò™7(=ê˜OG;7}üÈ‹ã…gp?D¾iM J…&ϹïA%E$Œß|5V4·)WéÈ*5i©=}“–Þ3É=ò±&rÙ,´A°*èµ÷·—´_óuî-|u8Wlh‡a +|é"¤ìKž_£=kNåŽU´¨…†ËF‹Á~Âöÿ£ø¡ù7YNGò4Ѩ?ßvé ÅÀ„¼I,:}JªœAÃÜÆO‘ OòÌïéã¡5±u÷’ +) +öh›…œVZJ7#_=<”gžÄs½ŠƒÊçEÜ妓ÔæF+(€L’ ÛìÑÄKŒN@¡#4MP%“­ÖÂ>N×ß«-,´hÛID¨³,ñU’{»òä™+C +Ä2÷Qµöƒtèä„ò»»¨Æíi Úp:M1å;‘â‚ÂÓô$"¨BÑ°SyÚ{j ñy¨»zõc¡3o§F„Fü/m'ªÉz¬MRŒ|Œ{Žxh#2»äïôèL&(¢¨gñq +ë"Ê3SO2ñ‰Wj-,C¶Cø\ÅAA¡Ò„gŸ 5l£CÑP†ÎAa×éà¡E{Ê!­ßº7(¬0ºTjF{$—^Ê÷¦BÅɶ`p¯•9:³,r¯ûZšZèÐÁ½Ù›µJµ©z]&¥‘Ï7ÂCù‘¤Ïfä½VK„}´ž£Òð9Š…Ö?Ö%"´â;ìYt gЖg¼ÖJp& + + í<¾?¡íj¸³¸AaŽ°yг*ymÕÜÀB{³Ô‘µÐ=×z¾'i­‚¾7# ¢§ŸÊ9,tÄvf7(Œ‹»±v$„Át¹ðࡵîû6N×XÆe6̉ZwëŒë•­‚…(åñÐB7`œš +3ÙT•äš×¦AŸ‚ŒéÐxbn1¨B—ÖSÑX“âò9˜ +”5A-ôÑZO¹3q…Ø”l‚zâ»!«A·1· J¿Œ­êEÎ ÃÀ£±<´€X¨WÉ1ó­ZóÚ¿‚‰„æN樊âƧB3.‘5@CoΕG«IA{8̬KEKlÁU ö'[@pÌ +$k;PiZËëó.PÛrïD­˜ÀpSqà¿;:‰äi„·rÇådÊ5dåœiWÉè1îø@ùJÙ´j7×>îôA%ùòFgÍyÝ-ýæZu<ÂîXSë:ŸñlK©I°ƒ™·PñIÇJö ú¥›äL€ÎúOÎ\¡“‰ ó£Â°“‰Iü?PÈ ¾ç\NðW<úémA(» ª<…|éÇÒ¶  ƒ¨Î)“ÎøVûüÕè—øSƒ³r6Æ°<¾$¯“ªÍÿžzbಊÇh3Þ0”[ÙÊ9S ¯6%¡IGWpõÐЛ2éìðP~8 C­…•ßÔiX­¹…“ +[[8ç,IЊꘂ-zc:­+‡Ãvƒ¢îÿ + šÜ«tÿ‰ ´øap¯Õ… ×~ŒoÛoZáDð¡¨û'@Ã6¥ûŸ˜ J&kZ Q5@¡ƒ?9"uÿD¨Òý¯ný˜Q+?\“¡ÑZs@†*=… Š¦$ÍÁJO1'A›wQµÅY ~iyǽ§N¯¦7¿Z‘ŸH¿{i Y ¾©Ö N +_¿ánGWKçÂù¬œãœ–ûË–€:8µ¸\<{;ñÙÛ °è3þ3F.ÒúèoÅ­{ý!^x¾”%Z×=4á¸æØ)›SºñvV_˜|9DÓè¡„³ârm‚5@ýºŸ¼2ûÐ+^9£½b·ÿtÐÚc9ÐuÌÊ*ÍÛÁµÐðõq> NE`Ý–Ó"­=ØêNõŽÚµ_ÊñMÌÀœÁ¿C-pË‚ÿm§•LÉ x$n +méú +aK¬ÈªÍŸ/µçoj;Ó;ýü §£‡¦¯átô5( ShÔ‡­âûä*1~–Kðæg÷eóê(…Ó·»ùÏâgœ,T… Ï:‹œ 58PËÏéþS²‡ûïýnùuf—€uøÓÉ!øýz¿\<ÿSšþúä$ÀÍïèìÿ™³Œ¥Ç·»+=Cvo¬2‹ý¿hˆŠŒ¢Î×èJã@YI:•CödÚi7¨¶£‡æ»"åÀúNœôL§R_¸ñ§R]0Ò2ŠD”v=•qi^Û÷ÒÔǺás¤÷LroÚˆZ-¡—‹ÏŸU9TUødŸ‰`}Üöq‚På¿‹)V{ã Äçy±â0õêúƒÊûÿÆ1·ÌÍ·ô¥Wîi°>uýZî\3YP°‹=‰W¾i©ËlXÌxîû»s÷Ä~¡ù¶W4¿`Ï¼í·±ÇoÎÝü섽^•½Ö$!-– ÍMÏB姀*móüŸPŸ±ê–R<Æ +.mSúyˆ;/9ߘ ²Þ¼ua·ì:½bCÂîB²Êßßq¸Ò£tã÷ŪIS·[cÇx‹nUU-ÇðÓ¯ã»*–vg‰x-Y°ý>¾Ûõ“³RO.Ü–nó‹¶)NÅÊøM]wó»-Ù¾¹+oÁûÅåý¹µÛXØ~èùõü¶¥ûEü4-ÈY`½œwÿM2€õfÝ»-…n±¿XÌx¡–‚拽8ƒæ±úº?ùwþ™ÇLíO=Ƴ-]¢¥Oéƒipå„ÁŒX„ÌSÔòMwüZúý(ˆ¿ KðJ–¾­cú»…=•Â`kýG§²äɸÄÊO»Am†bˆ9öZ +Ú鯼Cªÿ±^­[i+Qø ò‚I@¨­X¥ QJ-P¨´ˆGZVAÀ÷?3“ÌLfg&À–‹…{öí»lÖIº3Ê)i®§€káha­1 +¢D0^¤©P‰HXH„ö–8•âš=¡a`O :&º Ëob‰Zž-"Ô.|âWîKé 8o¼ÿÙ'I_ì>­²yºD,0¶ÕO϶iæçâuG•>D R"¶Ãe夀”–Éhp±¯`¬.7!ÎÃ<]ª@t}ˌԇ;¶~ÓeNaµ³šÌËWRµ¥µW}ªoRŸ‡}4ÁGu@’»jïŸ{[8¿3gøÆ-uñÅl–â ÀQ-‘ÍŠq®YÀÝéçÂi°Ž´àÁ~J´m¥Ýr"ô¸sƒøãèÝK¼óV‰ØÙÕeíà+VÓ`vß³äCÔ;šùQq–¹qi¥J¯ÓâJsç´Úqóªai÷ø­ RÕ¤3 éìº7Ð+½éÆqšJ6Ùƒ!C{iƒÐ0¦Æ;{H¸j‹X_pcCQÁ8±tT€{s¿PÙä€ +2°•®”ÛKÎJ—*]}g¥Èj@VzÖd­´°wšl©*]+ºÜ¤‚ ¢WÐÏ›¼Š7" ¥±2¸^*ÝÓtü=çyÖäåo=/Ýau¯äO?^AúA,À¯„–¼Ò»5µÅº[ +#÷²¸;¬Ð›(Q„—5ÑËI×òůê}—zQ€µ¦Kq×ÿ†…ÐÙd#*Û/ +z§¢v$ÔÏ÷jSŽüíÑp0øbr2_¬ØMÅSK§§´Qœ=…l4…8™«ÒÈvd·Ò}0Š¦ÇQ>,_D)†™Ñ+p—N‡ýf÷ùø29£w³hž¶²>2R”‡á`ØÙšÜ༜Ñ\E÷u—£[ù`óJs¡ h£¡Óå0Í1ú® Û8Má$úÝ5U1d+·Æh(”Ïhî-Ð +£õ¦ +Œ&sJ…Éâ¹´죹¨pQtffǤq`‰ŽâÎ%ˆ£BòlXäÔžé[,ÿ8Íœ®¶HìØeXcQ¤Ÿ­¦<®9m!­ºwVPÁ7Ïgèñw*TNxïf{ø\Öo/ç:³~¿²ÍŠ‚$‹Ø2.Í¢`Vó*Íøî‚‚ó`Vªô:-Uƒ"Ü <5?áh†‘Æá¬=Åݱˆ8òsN!Ä0ã]óbã8,•¹(Ú*Ž†$ à ¥p(Ï]ðj8N8^ lj+…›F¨…ÜÅ×ÿ¾±¿'§…‚‹OY Q0¿i¡&WÃá[,/š)Ò—ßVâps¾'Z[Q1g]õaT Å[œl +\ZÅâı85ŒÖBŒJ†Vb_U £µ¨VÃr*RÃr*æi¡¦£Tn—¦?&Ën>ÿ_²œ%2*§„ç(¾Åøf€dê§14šÇrºäzÑ‚)¢;ÛÄÁŽæJèV írê(¢c±ŽK4ÉžŠËtC_Ÿxy¥PJÉ +Ø2””‚Õäd©Xíö9YBßýZ¨Ü• +GÚ J–®Ñô­l×Êú·’ú«ýÇøàòý·,ç9Ém +æìP¨Ãˆ¹Pä‰?¼`mÓ„‚ —8¤/*S$Œ8“ÌÀ›Œfé´_Tƒ¼q¿À'ÎwF_Ø6‚K +HiZ(w#4=^3è¯B¾>®³7ƒÑä|>?»~tþöÆ£ZòáÝ®ï~þ¥éõÎYÛþœ_šçõNåàö£>èÔ~ØŸrè»Bý¼šû÷ü¼šofP°o3/Øþ—MÙƒáY²d¢ß5^rùÅÇ‹dwö÷–V±|ëã«YÈŸößçóO;·¹ÒÕïO~—“¸Ò4ºJöâã×r›^¯Ñ<8¤½Ÿ9½9zÌœü{Ý¢öÅTÓѳ‡{ú§iP?ýèÁ ?š¨?äÁG õNõ+ãaèZ’úõ­1ôè]Ä£Îþ1õèƒa$É£¨cøÙù¢œYøµ–¸ô¯NV:™ö&Rô£ßvƒGçïwVÎà ÷¬ÛàO'/7À£GÉäÛÑûPüèÁàVô(bK¯Á/EQ­^ƒ§ø¨q5y|€G}ÏÏÔºøøåk~Ôþrë€íÝ}-Œ¸G­žýš*‚«÷<ùËÛó²âïN‚ßáWÀ_ ¯”"¤Ûîï<3yuäÿG‘¬0êÝúû_Þ`>§s¤D +ˆþcPã(Êwãw™Wâ†æÕ¶1M·¯)©^ßÖðŽUŸê²³—òufbªoo`Jš+)k¼k^D¤„¼ŽÈœ»*¨a-sö||I ê]£²z¥ÛFLõnz÷„—R +L)p¿+[Ïÿb°£0:|YHûTªô:­ø} +ºänr-$ô«M÷…>Éœ\ð +LÆ6—’‹Êøû4•­&SÚdàLÃõ fÛ©½²Þr†SÍÁ”4=VRÅêãƺx™ÑÚ}—›†¸fÉls¬=ƒö—mÙ?Žììsš†G½a³Œ’¿^Š¨ÕÝdw–‘äÌÀÎTRv¸U0=ܺ÷1î×ðSi¬•B,þ²•.ZIiÃ$)Rd‘ß6ðÇ ˜Ha2u¥U¥ Í­K¼Thæ,øÖ+¨–S»’&íœXr~LQžßm‹¦Ô˜ßåÔð4U8CèMÆWÛíÚÁ²7´¿/N A悺7N\EW­F÷ÅwѼWb½(¥QB~êîá‚+)ÔÎLJä p.¸ RòcòùJ|d%U›>Ò}àçâu/¡W¢š±ÉH:wÔ7™KÀ*=]*·Ÿ°` ‰»!^Ñ=]rˆgýXÄ£>)Ã5è’hÇž.K•^§µqŸêÚ/ì]¹î +Ô– o RÒtµ¤3E*¦än²—ÔVö)t ”(}‰×§bõÇqC5%þJšÑ,qwºÄqn謆ð¦ƒvlܱ—†eÿ\8~¢%±Ð9MääÒn9µË2LøRUtè Þ¡ûò}gèäƒ\¯®¢A4óÒ`ú¥i:*î %¿x ÷â…)LÆ£.X•¦¯êJ¤*ñR¡™³à‹WéžÖñ[)zwÜI;g@n­^¾‰Ú‚&Vc8Ž+©—pùÍ¿ÅÂÝa!tPdÖ,ç=T»¼ÊgÖdÍrø%³ÜäÌ2­ûÌ’G.Cù«.#æ6À*\-¥J÷ƒ)_qÝdƒ6±¾XÍ+)~%¥¥£KÓ6´ˆÞšÀÅÓÇ3Ÿ,ž¿oZÐH²¬»ˆ¤ÙJW?\¸”Š5C*'nLRì7)ô¢Þü>ÏÝÊ79Œ_·>·ÒUDÌ‘&g^ñwü2 ZäRζÿ?R#' Ýbh©ü§ °³î‡—·õ +†ÒtÖvœ¥pf¢` Ú©ïî·µùŽã>¦6eO6}Å8™XÙ¯W'»•ªö¹‹O‡Ò’qbé@ÁrJl c­f:M~Âè^¿ÃŸY^BgËXªV÷u„£‘ª¿È:O _Hµ…©@Ì 8ciþ]í¤‹“‰hÝÃŽoüeYQôÔ0®ìmGpň£“1jÉ/8Ôäã*þŒ£î[µ°iüýCÊh§ ôPKèhz æö‚ï$×ZBË ¶† «ç•” pYŒ`n¼”ùHJÀb„ (ž¥˜‚¾Ø:™6–úæ{Yk^£ÛTvÆÝÊ™ žÖê^Æ8Z‡¬Xî§îJÐÉNÌÇ]É®.—ie…¡…0nPüó¿0L»½†„—b÷UÌ%’Âp´¤€ÉÑü0p €Þ_F){0— +‡p(8þ<{ô·ØŠÏ"ͱh™þzל¾»‹ 7ûÚ·Šëð£'®•þC¦-ä^¬9X*º'|Œ–ÉñpЖ0Æ% ó„€ôð‰ð& ›RJ^ô@’ƒ+Ø‹ÛoM†?[ú·rŸj¤·îJØóÌôªÜ›»“C¨ y˜×h®.®³ûÛ…ñëÌ hX›#©IÆä±½sBýDÐO.N*ÁÉ&«î#":{÷ÅT'&ºäx8ˆŠcU$zyÍ.Ä(ÀÒ­®© ÂØÆò˜¥[c!Ò›ÐÔùjïõf,Ÿß»(7¿`½Û{‘ŒeVa÷%íxSž"gÇDz¬À\ò„eÚƒml~!‰Ôª´ä#¥fr<G(ÞƒßÅY›+æjk-o Ç^òb¯ádLk!& :úN ù¡±C)0ø‹]JÔ—ÇoV1…‡$SDX[ÕüI‹ò¾}K"±L\˜¥þô†µñ +.™H¼øØø¶„’`é–¶Æ +‘RÍOyg4•ªÕ÷ÞýÎLB¿+BÂPÔ¿œ ¬€}vR‡ÅÉÍ>v,+¥—e™eEÑýÙqd¼¯hˆûÚ!×0Û1Š{€0uÌE4°Î:^Þª…Mãï~5ÄZ~³†½9µ†ÖwÎòîÉ5,õoM‰«LÑ…kØÙq )SŠnû±5ì 5ÜSÃzK¬¡—‡,n콆Ȋ»„;_¯–ÞRV,,°³ÛRVàÅÇ]CÒKÿœTY¦H„á³´ý?a üËQ¸wØæ u¾”¶öE#@ãhίþB¶"â.%Ö&Ò2˜ÂîF‚ÅxB‡/s€±¬„fR5ïQ±ßÞ} ÖG%Ž±©ÄVjr+ä?qWÌäx8hKß•NÊGâ¹Ë†ÎeÁ +@aOb"$„!±Ñr[‚ï~ä>K»ïëÍVn·–€—P¸Œ|Os9¢%àñêïhI*F’”©Ô7·/ÀØ˺Xób_øØƲ3®)EÏOku/cæ*~$¨˜Ë›H z9ˆKP°eÊ9€0=ô„¤ç`Hi“·‚¢èOîö^üj|3ýÁX—D´¾4Íô*¨‚cåJUOzªñ9y‚Ï¥ý/^UëÖ|~Ç3õ“ Óºœýi` #Å¢ç­Ü§¹Lé¦5c›ÓMË„Á?V¿â‰åq%HçOÜz&Çßádâ¼»6ͳ@øû8süi°€s%sý᳜B¼q=Kÿ}à8ÍiZ{þï™í4ÛûÃuº ÜLRäýBäš¼ù¸{ä: ¥ªƒÿxNßq§ÙÇ ¢cnÓ÷PHs.µÏ± €ª¼‹à—“iùáô §è”[¢Àçi®ÓäÍIòŽã4§[_éW×)È…Èõ…ëx™å;-o®SE#GŸì\ÏÔ¸Èi=Ìuj.Ûªëò áV½|‰=]ŸlÜ£Ov#®þ¼Éüîuýõ# ç—9íÿ¬WëZâJ|"H Ü$® + +›ˆ‚°°‚Aðþgfr›Lf’þÉ~_;ÝÕÝÕUë©wïø»óuîÔ3™5oUuaYݱðm9¥ÌäHãÐ1í¼•¸ZPˆlªâ›k :(šo³Õà:ÞfÐ@ŸEè¶JxŽãjÎ"nÐãµW„ß 14ÿ hÖ¼[î'Š¤Ü%S‚R £ò~T€Ý·^>з҆ð ‡)]ðNjf÷ê„S‰“ù®“­ó¸Ç‰¹Î8á3fÂ)bÚ+ýž ºRb€8O»FJþ˜y§›Å³‹útÇgªÐ\_ôèú³IÊ{ÐÍ)Åh)AÄö(¼ÁïìK&ÎÄIÍFu(mR=iC4ÈÊpI/Ò¯X &_vöbã ][÷#[{KcgbÒ°Ë9{Vôh+¥…n'µ´\¼V71 Õ³:”=q!ÎÁŒM[§~µo4”Ë­oçªüJUô©|‘àWÌû{HA‘•ùª“ãÑ|P¹æ'æÒÔ¦bÌ0LX”[ É¿âZ$dñˆP`Æö¶Â‹ÇW29[ÓjÞŸa^KúOXÛ‹Ìx¨ß‹ÿ(þÅÚÐIÂë¦ð¡._+ öôÕۀ΂©³pÛgd–^3ò:[Ÿåÿ¦ª´"ÑEÎ#Ž0Ž˜*5'³¸\ÒŒ”Àßvd¦ñðû(’²n»oSrÔ»oíï[™¦ÜìûKGqq¹C?3ƾ´vÒ×Ö„³oȽ6bŽI¥¼6GÍ…7%õqÂJêFo«þA¾·UJ£]€ÖAZÿ¹îSÃEÿŒ}awP֕١àk~Ìɦ¸ž,9ÊÏKF(PË1Á\N¼+Ä⯾@å·3síQˆ™íHÄÎþíò5z^z1¦>ŒXÍì{e3 k±,U…Ô‡l%L—>ÂzR·Ò1Ä"ug›"渱ÖX-æÅ~¯Ø€Ž€Ñjÿ’Kzÿ¶9ž¸•6ˆÂï"ädvˆìÆ‘\´N³AÓ¤“èP¹»´?-è°X&;ùÞˆm•Üá™Ý Þ€»³´ð×MNMAVAI·Y¤–øŠ’T=9di“½*³zÊ E™TOÿ6ï[TÆcÔ¾9,CN8÷“ k 2àŸß92¿‰¸(ÞOqwΤáR Ìïh”¡ß—}Hàäfã ”˜<6¿s(N8©(cªMòq#PÜr‚â+]'Õ_˜’é"£¤Ž'-%ܧ*Ý?'BD0SR9Ù.©é–©•l¬´mv¹×[…-qŸ›‹Wwö–ÒËý®šh~ªÏErÍϼ _ZðáEê°é¸AEŒeªpkXAúª¤0öóêC½©TžBœü]Mø2µ:ÍÝîQZR`–¦¸WôðÃG›‰²8}u« +úQU {nÞ߃ +J­´ÉªO›å1múYr¢ +¾2\Ûöœ + ˜1â”S÷\M9½¦ï,ÀÆØYÛIFÛfUfÁ³ +öUÍ,Ø4ga‘ììj¤`·O/8ýÀ;ÀI:‚j¹eÎOˆ\ç"`¾/ô^òJ/íø®Q«3Kóðö“®—´/Í~>uQCwsÿ 5Í'èùt³x¦„Æiuõ‚8ƒf÷Z\0Éx£øóc€Ñ`ÁÏ$¶/0'öªWAgV8¸Ý^Ð}b•Zµˆ#„þ…ï†8tL=Ö{ÜkïèþŒúæ_¬½ŒHå¡æÿa3éV2”g³ ‚xíðƒxDZDQ ¢˜¾²GœÐ^Ù`Ý'â„¥ªˆ1ÉGʼnºÈÆÃc·äÕ»-³‚ºÆ +¥!¦‰Í`ÐzùzO¤S#7ÇÑ©!Õ§û…§æeÑüη‹ å›ÿ,æª/èggò<c7¯€w饽XF¶À£oüÅŠ±ñÝåWŨ +ÔœCÕn• E%«¶Äl¬`g¿°ûïGï‘2 ×1Ä„®úÓµï¾ÅÃ;Æ鯎® +nåA2ÞšYS˜\.œ±m'¦N‘#£†eDSã‹ £ñ±eøZ;±¸’‡Î‚µ†Ò0"×PìTÖXBý+ØöWG¯¡8 ¬±š[¾† ˜›Íñ¸ZCŒê÷SãïáɱƒÖejzå+ûÄqµ†öîÅ9~ »Ä²¯˜sö’Â^ù +KÚ£`“Û`–PÉH€æ,`…ðGmÓæ€woWÉNJ7–eHÃE‰F°I{sl#žíeŽ‡w!t°`‡ÉO¯Õ´.Ãu"×}j`Žï ¦Š{£zÊ|§hvœx(?/Ç› +èþñü ³yõ*Î^Ì1Ž>êbq"yAì´¤×FÃtu¡/Ȭp:ÀYw\Íi¿ˆG:ñzد¡¶„#-!ñðöWÖT5زyì—Æ3 hÊ´Àv¢† €}j€&Q¤¹/”"‘¦0BŘ}Áþíò5§‘ƒ5XrÉÌ+áËÔêÎ[iR;,Äþü–¯½ìº1©J( +zxH\=ÛK{³—0„)‘s–†Lax–ïõ£„0ï^ÈK¢]1[ihmídcCL8DÜ\ä>¬¥Â¸8ÀÇ€ ¥ªÇ@™ {[tBßë©}É÷Ö>ÄxýÑV2 ÝN$F°-ï|È]i˜~8—~h8]ÍÑ.dOÈ%M6âôuÔqŒq +§1æÉC<ætˆñ‚ç¦ïb"ŒãyŒ}¦[«FïØÇA½ ú€‰-´nˆ³N:â +8´ Á+aUÚä +ÃwšËž7~Ìó´BÁ²©Šo®|¾2nŒµÇû2ÿàô£‚K¬ÏOy1ˆ¨LýÃÑ´ ËÙöçMl{U½_>?ž²Þ"úÉils^ ?µ^eŽ†ý\¤Ç^*Üéõü6œÉ®Sá«L»~ú\,EIšÃ[)Í£i±T{(Š¥åg],?¤¢bý®9ëß³wñþkú!>6)ñéî³+v63±+n¿ÄîËõN|Žvño¤ÿM¼÷âËKo&ŽÚ‰­8Nôâ¸!<Ȳ,¯È›á2#ïâ?]y7º6 ?”í­,ŸmÑÏ·Ê}¥™šÝtÿý}çB~¾×ò]f¾ò|ë±\õ,x¯7k +'_ÿñwÉ”ÿ~2ï¯B :¯$ïÆÜ.²J6DZÆáÏ w[zzâEßô ¼k-©¢ö¥¾9l6\äc^ç¢ù~RŒõÊO®Û>™¯t… ˆá`¥“ŽÝZ_½e¹ ÛTš öÂW‰ÀXÊ×òbiö_U,ß7¿7'/&§‰zŠS¿—]K –ëËæ@Œ•FøÃþƒ(u/Fòæ½z"Æî§b›Ìëã7 šð­Œ&|‡•¯Æ–8[?¼j&¹h,Ðå€1)r¾âeŒ‹onà»*~ð6¹ð§Mø9.z›xã"^… ­ÿÖY™Þ endstream endobj 19 0 obj <>stream +H‰¬W}_ªLþ~TÙeÑòTTÔ4_ÊÒ¬|I+3+µ¾ÿ3»€‚§sNÏ}ßÿðc™åÚ™¹fvfB[Œ¦›T"×ÍIÊox!R€GÏ‹rX™jÖ\N‘ŠQu<.d›ÜVÒ“÷j…´Rðvñ¡ÊOO:¼]mð¢£D0Zö¯N¯?1Ä–x1 ÎÛË"ÔÄ/éù÷~Œn$úAr–!N—qgÙL8ýÌgåaµIc”¬ãâú½¹+ž]_ßáÚô3í¨©ŸŠÓ?À»¢¼äMÃfÞ˜ñƒ€”nú†+¨£ƒ@ gŸŠ® ­î÷¸6á_vøêõeÊçd«®€¿å) n%À ‰än÷ò‚žÊ–šÑç`Ù)ØÈžBŒáŸÄ;|­ÊtSL­Ü.Jà±ò +-ó³TbGš)ulµ6RcŤ l°ì3ؘMêŒ4ÔºXiðï%8?” Q)°(kŒ†Íƒc&î) þä&^FÓ\$óôíWs•Ä‹d\ÍFRK€-Ö«ž¶]‹ð¾‡®å$gŠÝOMƒ¢oæË@IT–(±±‘ê¦SVµ]\–ëó<'ÏŸØ7PÝH¾$RY濪B­^'âë8äÚpšØ»`ê ´Y;N%ƒ{Oß(û³>rR`6Àê¬Yám¤âF’§Ò{¢§ëX›M5öæh«…»’™x™ £=ÑY­Ÿ?fûCéuµký©¦1Nãx±*œ2NiŒýUûM•V§Ñ¿™Fä S¬ºn£åpÞÒ"Ëðœz̵ëGVu^1z~WóøL•Fµêì‚Ej€‡(ºù±ªÊÂÈ8Äê#µ¼Äy£nÇ*¨\üY˜" -­û%9Võsªœy©ÿ6H™ÇX˜®GdÂE!퉨‘qm.ÿ98eȈ¨–ZÏQ‰&ñbÖÿÖ>8å UyÃ@ Ã]´6~fŸýÆnñÞaŽ-^Êb†ˆÂ´ðC‚@S3£Zæ¨}ÆF92ˆÖ}§è9¥•ý(íÝ¥úQâ¢ç«v»‚¢5S_R¾wòÌE’ïUOˆ³oA³¤ù’è}ÒíW°ý1燽ò²r!™a£ÂôRÐ.Y#Bz¨bš} uâµ’+o·ù÷˜70Æ s¬Vù^“èb0è-`)8Õ€&.ýÍ)À¬"Cï•k‘ìH…v©,«Ö½"Cuj!ö ŠÚ•b»Ûi7w»s+éøóP5(bÔ¶/GžºÀäàk»MˆwàŒ4xÌeÚ-ZèƒT4ôù®ÝômµD[°²ýš4Ö$0(#h¯ƒ ªâËémÏKbxCåØ¡¸z–ëx8°p xqŽX8öI(Ü[“ûƒáKr *ï.öÝx¿ÞÒoôΨ>õN·®râýA9æ,ï5*”·L¹oC8êvá:6®¤*”8•ö>y=è i±BßèÍ :4¼öÏ pʯD I„óð9nB«˜ ‘'aîÝþÂëTfø7G:Ð{Ì !iñëÿ€ñòcåºî1ÿtð%ÿ0 e[ü÷ÊCnœöÄs¶’æ û÷©¿× ùò·ˆ‘Éá¨Ý» ufúö¡˜G¥7-ÛqUZ“cöÿRô/‘yä"ÿfÙsû²ž_Žûå~0Ùï$ù—²é‡BÇì#?6:Âöëüü!Í·$~Eˆ|ä1âW…ø"~E>Œ0Ñvlžð™N±µÙ¯ñ‰ór—”†RÅiiŽ¦ w;ݼ·‹{´ï¤Ä… ñë8&¡oídO{F#‚͕òS÷I3y@¤÷¸,U®µâÙ‡œÉ)q¡í{wèÍt6Ù;‹á&àž,±æjã*?xBd •¨f\~"œf -´9¸qhˆ˜¿¢\¥0šn®Øb;y· v‘#ˆél+âÇL/K*\:Á\uÖ(vKc7¡´bj|ƒØ”$²N‰á¹]Ÿ¸üª˜v*¤ÅE¨3{ïd©}UæjhýúèB¶<Ì/½ŒOy…Ú'¢e~vͨ£Ý¡Icc§ø›I¼ø,íÁÎë"­âþM{ +k} ßÂ:mwU9z¦:»_kÐä׺0%)™”_òÐ6‚õö\ ö«‚Û–ÔDŸé²U³-.Ôœ’i}xº¢oGÓ =Åéð–œ<ž9éA9ç$}•åâEí’“Ÿ_.¹xJ³&‘z9õKŒùmZŒiY³VŒ¥GÀîëðc®ûZèfx¹ÝÔcqçíeóÃ×ÊÃj“öüá È}—eß0öáã·}Ò]€ÓmUn´ÃÖJ)-uz>5òpŠ°öO u˜µŒ¢=ÉfÛ´‰u“0MméDÖG`ÙRïêÜr—… \æIÅ/8£tJGtBRŒxtdº¹`ÓAГ tšdÉð?Þ«t-qf _ÁÈ ˜td {BÂ"" +(‹¨ îsÿ§ªY Œ23ßù3NÚîêêz7¼dϹc wä­v=¬´]@Æ"[@ÂSŒªQøÉ—ÄÁ§Ž™Ÿ\‡sÁt@òŸÍÒ8ì¤_ü1Ué,ÕþŒÃ8ûVtÜõäàÐúØr¢ŠÂŠë៥XÕG>ø~q.Ì{“šøX:9…|SNaÈÉ_d0‹Y].I¹N£Ñg ÃKY-KþW1HIå gÇ’’ïÞ«0 ¯Ä͵zåQpP–Ï›ÄC¬ß@‰ý‹°›sƦ?D„-^d›Ó0þÑìspÉ^ gì¸ÆÁÿ6K&^@‰üäío¡Â™×6x쨄ßv òy¥ž´Tz»òe%ÿ¼q&>*žóÿ/¹€ðó=œÀ‡E™à ‘R^|ìD,%Y£m[bS¿¦°<ïý.,zI¦4›Vr¯¬ÛÛ-x +虾°ŽG­æaK tp¡'¥$á¹²l¬Ëqï:ºO²Æ”€àCQo—š • ¨F4Iô¸œdxyQІU:>ûÈ8ºÌ|š$w™·x™6PFÒ’´ÌHEÿX§ªÜ¿”¼)–QÿI!úI$&ù=Þ¼0¿{mà)ÊÓö–[;‰n÷Ú/pP +IÂ)™ é3…uùpäØ:fƒ'¬-ÊžÐì½YTéK=Û¾½fIÎç2°Ô)”ª¨2gÎ…½¤å`‘Ç£—A€x%¸×SŽê½agß ÛùÈ·µ·(r'ºé/ÈlÎçËD?­¾$³ÔÒÌŽ|HgY‹®ÄÇá“N¦ñ€~¬‹Â“扶k’7]«ã,v@÷c~æ·Z$>Ö‡*ðÏì<$æ~á×[À¾£Pæ`·,¨¹$í]žÙ­ŠýIJà˜ã†5ƒaL׎Ø}úmœõÈ• hÎUWÖzã°Gìf‰–üV|xNÅ™“€ö_kRQø•—ü²¬ƒ[ik¬B©0¿ko‡5¡“VÑÒ Q8°ÆðŠÕm‰Ãj}UØ‹3_n¸ 3—˜@ƒSß(Æ­ØX¸hŽã_xéym½$*²vùFÇt8¶:Aøz»´n[0 £J ¶iG±ŠíÈ­þÚ1îê_6#ü†¨Üi‡rP;ö7czúZu­£"G‚µ~ƒVBSG€éöÁ¾{^Ú[›}À%QbRŠL‘ùŽƒã i’_Õãðã†sÄ.‚Íw¸îÊÆ,ò Ä|²AhSY­0cFµ1=„£+ÐGoZükù» ‡¦Ê}ñ ìw5ݹVÖ¬Š§ ýήâ2¼ÚU/†Ù7Чλݖl_÷ +Müýú¨Ñ}NF#sÓI‘inY4 +ÆKÂ)hK`vTøípÉœJU/?žwÁ–ËIÜB®ãÈ¥Ý÷%öú†“7åfCpÊ3$(8`¶Ãt}Í@y#„+«ô,ñ-ÂÙë\0Yà'!<–1Ým(–«·ªuY¼`º€vŸ·Yr$3|ýz;mèŒ(Vl´|4ajåZþ²šÇoÖ4E×Â{GaÐJƒ> ¾åG,äئƸa6V}u(H³g«Ñ¡]ª1âÌDØG{Òà齌˜¥Ü¬Útl^ìÊV'Á Ïî…›¦ÂŽ!æ9˜ƒŽhÓ²gÜÿŽ”M]£óbЛéü—Žê›\7u–ÖSVÁ8É» ÄŽr“##Ðð\£ÁÜR4Ö™´†ƒ}{T%©UW­j8%ž9}±º”Ÿ)Ùlk`èQÍ-n ¦šÆ½y&ÁTÑg°¸"fimc–Ü F[E![îØý€­Ú]>ÌîÛü˜ÝBÜÃw†}OË:üt[þ»oêË®Ýßg–ÿÔÆÂ)hdÿK³ hú?œò›»¸ZròfJ0u‹pø~Þµ©Þ0ƒ&ÎXsFú€'kö9Z¼îg׸s +Jpß)h\ÙÒð–ßædÿÁœ¼Ñ¤¿áäèâ9ù9ñáää© ²ÌÚO-«¨ûàæþ=1ßúbâêœ^¿ £fwb¶­L#æ©E,bÞòÉîÔKøT% ºTmfÊÌ4ø2œ˜î×jVlV–Šþ—,tÌ/'U°/à8µY4¤XSÁ~h§’U¸4¸uf0×·J•j¶ ‘VDv¼Ð ÃvÆd1‹F•-âfö“O„ùÝDCçM`y[‡ÊY‡é¼ü€ÍžFvAÙªZ=Í£ƒ•®.éíŠKbN(2÷Ìlv" SZËN+ÁD€é®Þ•Ãè`+¼y]¥cå)C´á,+>M“£4æyšèy©¶v°6oPb),?ÿÜ • ßuiƒ«S®C+Yñ)e˜ßf ùÂ]<¬ò ·é–üŸ¡+Ûadž×2™¶äÚÖcº«Aµ¤¸vèÛ#î–îNÞÝ¢î‘ïߤ;lyòF­.ÙÎpJŽ°»šª†ºW®8§ï-œßbrÌ™}€j–Š›>¨:¤›–Bîfaðü`!Õâ— nønŽ|_tBaÓ]z*8pI¹ö>h]n‡U!ÖÁ»ä‚±=â~¼Â…Cr•E\𛕴íêñ’@p •E(à¥÷8Z"ü0~³/à†YÀ…û:âþAÀ̓¦Öxq¸6NGàÛš»Ÿï©,ÆÕû QN@I{Íšªô—e£ÇæD V3Qkbz-Úò§ÛV³¿Û1ÃèqY#ü(TUü)“aVÙÅ.à “a¶üûïaöqä;ÜÁoû÷¥Â›™´j5ŠükÎ%p°1üØèËzøú r™Þ‹:}^$mlw.ƒ&*ÛN›s~zñàÍ®+ìKºmø“\†‚g'æ +FJ þé^€žä€¢OKD÷IU•I&[;¦‹­Nc×e]~0¼l°¾oxGARÜ êBž Ý>] ±é˜øÎâÈç0m½$43ÃoØ‹{eýÞn£v™˜ñƒ5údlf¿¹÷Ý'Ï-¶+28PŸW ûJ¼L;P2—gâË";kdšO6q]&ù#/ù»Ó, ½0É.AU¹¨~—1Øl|!̳­: *“Ç €ÜkEÊu®Îàâ„+«ô,±_¥±wèúæ^õ7(qùàÈ0Yè¾hC¹°P`9*ËóD8Fß!,ìlbˆƒa¾"–¨C c¶ §lñ±}Þš‰ÃÒ˜@0ñíÍ WqÇ¢ÎèoˆmàwK È–û2È:H®3È^ÑfGAä„÷š7pã̾±•&]›Vµ{yIª"Ð"bMñ* Õƒ²”k§RfS)0C%onpc(Ýíù’©ñ›v\U®ºJ‰}k 1Qó€µðyß²¨\KÞÈbuV[]Å/ +ƒ›á-úG`OÏ]9™6µ·›M\BtðŽ¬¡FrÊ´mp;àš@YbÁ 0Ò÷¦ä üeߘræÿÍ(üò«Q$CŒî'­’¸Cúc+Ìݼ²õÞ„t Eo͸x¡öÓKˆØã´Ž$é©öL/cF ÿÝ(`‰‡GøÄϾ_}ß÷£ßñýÌÖíûM%((IHn¡Ò_º½#ufRr»}N”ž¦“:ÚVÄr{Êi—Û3ÿãM·gÙËŠ#ªå¾Ûg¶_¹}áê ¾ð¡H5a:2³EøuÑÒ)¨ƒ`Ë7]·ÌpŸÖŒîkýEƒíÝžXŸˆº¸å•Â¿$C¿VX¢YðÈãˇÂ‘5hYGB ̽Ҝíå©ôø Ôjö­Ó[hjÕÛê“€¢rdz¢SaÎïÀ«Ïù}X!s%.ö‡ëß[},0‹dU.;e·ÀÃl9ÐXé4×fJ.;™Éóç¹Bó‘µè sÛ œé²›q»-Ãf€ï–‡kà‹r_„ÿ²À’_–3ÿ²À—_mG•lT?†øDÐléÆD‚ì„Ù÷>,ïšR¼W•êÏ‹ÌoÁ‚ìØWÀø.,jª¯¡*‰ÄÉ_ƒÙ‰;#r +1a;À +È•¶&7 î^.a9fÊó$RËÉÍiÿLö»†Àå>ºÐaí­b´OÁD¼HúÅL,˜k¸æŒO`°o8‚ƒÜ·\À•Qã`2¹`Ký9NTw*yî0é¶0߇XYK4" þ ÃІ®uvêóÛlŠa_y|ªöËã“órå¦éoðéD8¥I»n·Ô8ƒòóiõÙÇ=³ j¨¿´ÓhòÓ×TEg®¦Â„l7Œo>ì7}m¯$jªÒ"øU"å™î8DfvUÙ¹´ò‰ÈŠÊ"Üå½ @,ÒÜúžhðn¨›îÊ·N?4hWõNz‚kùl²éÆè£D*q€ùž¦ê|]S‘žòSÎ÷¡ºk…:õÏéh*Ÿ»?O {óv×TCŸN–e¿Hú,›ìᓨ+ U¿.¦!² E›HuXîc8¢ú/cn˜BÉüv[8ƒ¬^*øc¥l/‰Y˜Ç’Mÿö.ÎwFHÚªÃ1)i,z›ê8wÞÅ ΕçIú§Ž,?´ß4ìQÆz±ÌšÕ0ý§`¢Û6Ðë(}WqÇ.vgCk ÿ—r˜ ö07XÒˆ‹Xå6RQo1\Þ£E Åys%Gìlù_±$ÊLÎ戙.~›%öc²\L£}Äg¾‘“ä]uª*PïCÎ2P¯Õ±7TÄcD%ž ÖôØÌè +“{}_¤H@wÌÕ¤«0h†gyh&¾@2‰éœ‹½¡hž´ÏeÌ 4¨»é¬ô¢¹Ã4¬wy0%^–­Qúï0‘ tÀ >t×ÔÉ$bŒË†¾Å’q.]•›õZm%GHæßgIâw§ žvÓßa‰K“ɳV±›3+ÇùNKT–ÞVí-zËëBà£n†29zûIì Kª)®@aÊÃ;ä8Ô[Aøúç%*€ˆyë.l€`ŽÓ±,’Ÿ@QB+"B¥!ÚŸ_œM€àø^®Ä,:Ðp +ŒVäppá?K¶»vã«ÿ<Éd:ñ#O'H½×ù¦³yZ<-ž¬G*Š2XÎVµÍ|~>ÿg§¯î?ÞæËÝÌ©Ô¯F*¡ÏïW³9üOgœdB¯ AŠÝ˜ÝеïTÛTËúCºþÒð÷òýAÈç{Ç^.ÖÈ{£R¨ç xùÂKΘ+-¯ðøÜÇÇšk§á&ÿ…jì¨û9öïŒøõù¦üQåOÚ—zËðö÷BQâuZ]ëWµDD©‘Èô"ÿxM +ŠdK+ +ðx%và6Ù†/ÉýŠ“®´½ÝlÔ]ÐËåN㸌 œµk9ÒS’Wà4 ß6½|×wêåïÅSl(z£¥ØÌ+œ ÞÈÚ¿±Ã•j¤)ž’Ä‘§ÚÝ¢³ÉRä\ É‹Eä *nHüÚ&ŒD1…éÝóœøG(´Æ\Õàéc&Œ›Ñˆ’G¥‡vœvñ0Èì,öŒY°â]HrŒ%1+(äy¦arTÉ[ •¨Ë’ÊWƒ¢ÕdHŒ%e%«¡£8öCµ¥ž0–ÄŒ\¯¬%±c× ö[ˆ€*üä “À@ž? ¡§GÈK^x<Óœ¾ï¨ÞœÕˆ‹D@oæe‚6R Âö¦@$;p´g‰ °>‚Çbú’é~g×1é¤ÿCI†í÷çüÄ—¨òŽÚQ{б5 +Aüì˜.MrÁÌbÔrÕø³M¾•?‡]è,`/´Rìw€.+N‡ú3ï·{´Ê<Ò#×LuÓ©zM ÍÏuc–÷ZÖ5TÌZÖ¬i󈸩ÜëŸÂk4‰3FÄÀkÚ1‡fƒé@¢°Ÿe¨ î®j%Ó[µ6T8«é$†îélQM0åËuŠ¥½ÍÊߤ=r;gêX/E“ûÂú½s˜ö¶®PÚKA\ZØrÄ“%]Fg1Ñv[ä>$LÀêcá%“.bÅËê„Ø=’]¤Iã +$QÁD%¨V¯yRsX§ŒNÐ9+"vgˆ£¨[o©ÞlëÐT“heã‚Ñ \^HdW4iq/¶’•’Üè 8™•¼¥'$U6§n§/åWß²¦ïÙrÉ=Ñ7¤ÀòþSž"™9x«.8gŽÖ4ZD8'Ü~¦~b˜U·“áóýê #\`©o6ùû kVwóÊf“k]’ÿO{õw¥DÑOï0¥eÑ0I&¿Ô¶J«buE»Zµ%‹® + +¸Öö³ï{“dfȵ=x®Ðا.Å5]Nbí»Ù$vU¨`ÿ\aÂ:²î¸ÆøêGÕ÷?,~c9.f—ÿšmñW©Ã€9Ïo>>P.}<¾–µÃ¼Ä“¯,­›´úþð”–¿£{(ð˜eñúà &ZÀ¡£±Âl—Ò¾xoßù{å/ ¯°6,Ô»tÌq˜ý¤²ìè[ìL›Ö·–vyáä—Ò9‰ eGT\zÓ—yÃ:ßËÉûd<¹9±h.fÎEL/.–šÞÂ0MS$éU–oö°WopàÞÆX¥û­Í¡Ìš»‹4oœ6m7®ù@ÃGŸnD¦ Fq\ë 7ð …ÝýSPœªç;Û“:9i~˜Íä€+ù©,²íÂzå×s€Š +µµ¿%ƒÿ®É+%¯Ø‹d€Å¸ªæPœfÖ¾ZòñŠd‘9µÍc”êi½æí¤¸£ã Î(>•ãKD–ä’£ŸW›gbªžÄņÿrI’Ú¯‰–Šîãjrƒ¼?b¿dÉÁÃ$×èkZòô´|]KŽSþÊ–\ñü¯hÉ"7å¯lɱ/,ô-¹z¼ž%‡YД¿²%Ç5¦\XòšlY¼S¼y“¼¾6ÞíÂ&ßÌ•Û Ö‹‰6/ò“!ŽV¹Ùˆñ@À‡Æ»V|ØÁÇû[xô¹ Åó¾É’Ý´PºKwÓò<®œ2îµo`Œ¿bUÑ®Ÿœ°õŽyØ}ž +~ÄCß—O„wN>pÀš›¹¨BŽ[eYck.áfߪ¨±FwüÖ|ì·æ¾û‘QLO›- ð¿~þôãj¥ƒ'èulŒçßm\¤/e¾TâÊ®7 +fb÷×—+h¶S<üUÚ°¬ÖZ¯³ý­öÏïo¢ø¿"DšÑèþn«u/{öc40,ÿQøÃO?$–ÛuáÁÅhãÌ(q,±Ê¤Ñ3Š­êÊ`´zy>ºì÷ÚƒG²€¡ÃíÆÁÆ*Y 1¶ØER‚lh Ð0T&pN´ ÖAÉ +ü>÷0ýÊ-|ß1(‰ÿ®àÛ^òd‘-ƒ&¹þ6áË?Ä!„‘mr|JI©ö ÇñL/d˜yh†¾ç‘5æš”º!qÛô<Ðf»4F9¶Ÿb’ȹáضéÛÌRPVe’‘P¼'b3Cæ;D2yAv¾4Jd.˜4}çƙᖮZÆÒ\}?«Ú·uÕžŸUí9Y&‘ªELä*˜„1Ÿ¢Zd%P"sÁ¤éCÕÌÆZ¸ŽªZ‰%¹2«êI9ÌòSˆÍ,Þ Eƒ,‘Œˆ$d,MU2¥rä|R´HJ€DÞ‚HSÇ5û–®YÆÒT=?£Ù³uÍ®ŸÕì:Y"‘šEL¤*˜„1Ÿ¢9MJ€DÞ‚HS‡š]Kï³KRu­LŸ]ª÷Ù¥ZŸC­ÏaNŸÃlŸ%S*GÎ'5‹¤ÈÊöYW‡š)?M›RjÙdݰɃ!ÏŒÐMëÐ4꘶ {yÏcrÉ#iü¾Ãp&ߊ±¢UÏcåþ}.YŒ°2W‹êçccR;S&!õ8©-ßIH½é©ey†Yí±Š>‹Ô¢…–RÊÆy&e^HÏdu'ræ+bŽGŸíQ.2·GyÈ|E¹ÈÜå!'( +èKå!óå '(ÊCæ+ÊAæ+’ÑÀ“{εõŠ4sæz+ÀUý VÖåØœ]2 {ï7¹ I©Lÿ2ªhzÑðÂ%Õú Šz>NþÔ¾‰6zè'<[hŽúƒÇ4°h€}à~ó°cx¿L˜Ã÷Û\á³\h”S³åôiËpð¢`žv8NhZÌ•ËLŸ)"Î- ‰7=?=zðøPp"¦Ði³6à"Ëm[iÞ6iÈœÐ-£ØÁe¯KJµÚÊ9üPÙëÚˆU~(ähNÀ4=iHÉS¼©è Ãd©¸4¦Ði³NKsôþ0›6XÉ@Æd¦ò]©ˆ1¦)’1…O›wjŠ‚œ+Çgº"S2 ô±@ï‘Œ)|Ú¼ÓRôòýLù,q¦x–°)r¹Säò§ÈåM‘Ëž"—õ,W£fW6ÜÖZ¯Óh?Fƒùy£XÜmw£ýAûò:Ýaû߈´{=\ŸÑ-Œî kD†ýŒÀ+)¼X\Û©ÿô¹’. endstream endobj 9 0 obj <>stream +%%BoundingBox: 174 237 697 389 %%HiResBoundingBox: 174.6948 237.6172 696.6953 388.9453 %AI7_Thumbnail: 128 40 8 %%BeginData: 5498 Hex Bytes %0000330000660000990000CC0033000033330033660033990033CC0033FF %0066000066330066660066990066CC0066FF009900009933009966009999 %0099CC0099FF00CC0000CC3300CC6600CC9900CCCC00CCFF00FF3300FF66 %00FF9900FFCC3300003300333300663300993300CC3300FF333300333333 %3333663333993333CC3333FF3366003366333366663366993366CC3366FF %3399003399333399663399993399CC3399FF33CC0033CC3333CC6633CC99 %33CCCC33CCFF33FF0033FF3333FF6633FF9933FFCC33FFFF660000660033 %6600666600996600CC6600FF6633006633336633666633996633CC6633FF %6666006666336666666666996666CC6666FF669900669933669966669999 %6699CC6699FF66CC0066CC3366CC6666CC9966CCCC66CCFF66FF0066FF33 %66FF6666FF9966FFCC66FFFF9900009900339900669900999900CC9900FF %9933009933339933669933999933CC9933FF996600996633996666996699 %9966CC9966FF9999009999339999669999999999CC9999FF99CC0099CC33 %99CC6699CC9999CCCC99CCFF99FF0099FF3399FF6699FF9999FFCC99FFFF %CC0000CC0033CC0066CC0099CC00CCCC00FFCC3300CC3333CC3366CC3399 %CC33CCCC33FFCC6600CC6633CC6666CC6699CC66CCCC66FFCC9900CC9933 %CC9966CC9999CC99CCCC99FFCCCC00CCCC33CCCC66CCCC99CCCCCCCCCCFF %CCFF00CCFF33CCFF66CCFF99CCFFCCCCFFFFFF0033FF0066FF0099FF00CC %FF3300FF3333FF3366FF3399FF33CCFF33FFFF6600FF6633FF6666FF6699 %FF66CCFF66FFFF9900FF9933FF9966FF9999FF99CCFF99FFFFCC00FFCC33 %FFCC66FFCC99FFCCCCFFCCFFFFFF33FFFF66FFFF99FFFFCC110000001100 %000011111111220000002200000022222222440000004400000044444444 %550000005500000055555555770000007700000077777777880000008800 %000088888888AA000000AA000000AAAAAAAABB000000BB000000BBBBBBBB %DD000000DD000000DDDDDDDDEE000000EE000000EEEEEEEE0000000000FF %00FF0000FFFFFF0000FF00FFFFFF00FFFFFF %524C45FDFCFFFDFCFFFD0BFFA8FF7DFFFFA8FD06FFA8FD70FF527D527D7D %527D7D52847D7E52A8A87DFF7D7DA8FD057DFD0FFFA853297EFD08FF5329 %7EFD07FFA854297EFD08FF53537EFD0EFF7DA8FD05FF527DFD05FF7DA8FD %15FF7D7D52527D7D527D7D527D527D7DA87DFF52525253527D52FD10FF53 %2929297EFD06FFFD04297EFD06FF532929297EFD06FFFD042978FD0DFF52 %A8FD04FF7DA852A8FD04FFA87DFD15FF7DFF53FD057D52FD047DA8A85252 %7D7D7D527D7D7DA8FD0FFF2829282953FD05FFA82906292878FD06FF2829 %062953FD05FFA82928290654FD0CFFA85252A8FFFFA87DFF7D52A8FD04FF %52FD28FF7DFD14FFFD04297EFD06FFFD042953FD06FFFD042978FD06FFFD %042953FD0CFFA87DA87DFFFFFF7DFF7DA87DFD04FF7DA8FD26FFA8A8FD14 %FF53292953A8FD06FF7E282929A9FD06FF7E292853A8FD06FF7E292929A9 %FD0CFFA87D7DA87D7DA87D7D52527D7DA87DFF7DFD3EFF7D527DFD08FF7D %28A8FD08FF5253FD08FF7D5253FD0FFFA8FF847D52A8A87DA8FFA8FD047D %FD3FFF7D525252A8A8FD04FFA8527D527DFD05FFA85252527DFD05FFA852 %52527DFD11FFA852A87DFFA8FFFFFF52A87DFD40FF7DA87D7D527DA8FFFF %A97DA87DA8527DA8FFFFA853A8A8A87DA8FFFFA87E527D7DA87DFD11FFA8 %FD08FFA8FD41FFFD04A87D7E7D7D7DA87DFF7EA8A8FF537D7D7E7D7DFD04 %A87D7E7D7D52A87DA87DA97DFD5BFFA8A8FF7DFF7DA8A884527E7DA8FF7E %FFA87D7D7DA8FFA9A8FF7DA8527D7DFF7DFFA8A8FFA8FD5BFF7DAFFF7DA8 %FF7DFF7D7DA87D53A87DFF7D52A87D52FF7EA87D597EA852FFA8A8FFA8A8 %FF7DA8FD5AFFA8FFFFA8A8FFFF52A8FFA87DFFA87D52A8FF7D7DFF7D7D7D %FFA87DA8FF7D7DFFFF7DFFFFA884FD33FFFD05A8FD04FFA8FFFFA87DA8A8 %FFA8FFA8FFA8FD11FF7DFFFFFF7DFFFF7D7DFFFF7D7DFF7D527DFF7D7DA8 %7D527DA87D7DFFFF7D7DFFAF84FFFFA87DFD33FF7D52527D52A8527D527E %527DA8537D527D84537D7D7DFD10FFA8A8FFFFFFA8A884A8A87DFFA8A852 %A884A8527D7D7D7EA87D527DA8FF7EA8A87DFF7DFFFFFF7DFD0AFFA87D7E %7DA87DFD09FF7DA8FD18FF7D27A8527D5252527D7D2752FF527E27A8537D %527D52FD10FFA8A8FFFFFFA87D7DFFFFA8527D7DFFFD087D59A8A8A8527D %7DFFFFA852A8FFFFFFA87DFD09FFFD047DA87DA8FFFFFFA8FFA8FFFFA852 %FD18FFFD05A8FFA8FFA8A87EA8A8FFA8A87DFFFD04A8FD10FF7DFD04FFA8 %52FFFFFFA8527DFFFFFF525284FFFFA8277DFFFFFFA852FD04FF7D7DFFFF %FFA8A8FD0AFF7DA859FFA8FD05FF52A87DFF7D7D52FD3CFFAFA8FD04FF7D %7DFFFFA87D537EA8A8597D7D7DA8FFFD047DA8FF7D7D7DA8FFFF7D7DA8FF %FFFF7DFD0AFF7D53537DFF7DFD04FF7E7DA8FD047DA8FD3CFF7DFFFFFF7D %FFA8A8A87EA8A87E527D7DA8A8A87D52A8A8FF7D7D527E84A87EA8A8A8FF %7DA8FFFFA8A8FD09FFFD047DFF597DFD04A852FF7D527D527DA8A8A8FD38 %FFA8A8FFFF7D84FFFF52A8FFA852A8A87D7DFFA87D7DA8527DFFFF527DA8 %7D53FFA87D7DFFA87DA8FFA87DFD0AFFA8FFA8FF52FD047DFFA8FF52A8A8 %A87D7D52A8FD38FFA8A8FFA87DFFFF7D7DA8A852A8FFFF52A8FFA87DFFFF %A87DA8FFA87DFFFF7D7DFFA87DA8FFA8A8FFFF7DFD0EFFFD047DA8FFFFA8 %7DFFFFFFA852A87EFD38FF7DFFA87DFFA87DA87D7DA8A8A8FF7DFF7D52A8 %A8A8FF7DFF7D52A8FF7DFF7DA8A87D7DA87DFF7DA8FFA87EFD0EFFA97DFD %09FFA8A8FD3AFFA8FFA8A8A87DA87DA8A8FF7DFF7DA8A8A87DA9A8A8FFA8 %FFA87D7EA8A87DFF7DFF7DA87DA87DFF7DA8AFA8FD54FFA87DA87DA8527D %7DFFFF84A8A8A87D7D7DFFFF7DA87DFF7DA87DFFA87D7DA87D7EA8A8FFA8 %7D7D52A87DA87DFD54FFA8A87D7D52A8FD04FFA8A87D7D7DA8FD04FF7DA8 %A8A87DFD04FFA87D7D7DA8A8FFFFFFA8A8527D7DA8A8FD53FFA852527DA9 %FD06FF525252FD07FF527D7DFD07FF7D5252FD06FFAF7D5252A8FD52FFA8 %5252A8FD07FF7D4C52FD08FF7DF87DFD08FF52277DFD07FFA85252A8FD50 %FF7D277D527DFD06FF7D527D537DFD06FF7D5284527DFD06FF7D527D5252 %FD06FFA8527D527DFD0CFF7DFD04FFA87DFD1BFFA8FFFFA87EA8FF7DFFFF %FFA8FD15FF52FFFFFF52FD06FF52A8FFFF52A8FD05FF52FFFFFF52FD06FF %52A8FFFF52A8FD05FF52FFFFFF52A8FD06FFA8A8A8FF7DA87EFF7EFFFF7D %A8FD18FF52A82E7D7D7D5252FD047DA87DFFA8527EFD057DFD0BFFA852FF %FFFF52A8FD05FF52FFFFFF52A8FD04FFA852FFFFFF7DA8FD05FF52FFFFFF %52A8FD04FFA852FFFFFF7D7DFD05FFA87D52A8A87DFF7D7D7DA8A87E7DFD %18FF7D527D7D7D527D7D7D52527DA852FF7D525252277D52A8FD0CFF52A8 %FFA852FD06FF527DFF7D52FD06FF5284FFA828FD06FF597DFFA852A8FD05 %FF527DFFA852FD06FFA8FD057DFFA87DA884A8A87DFD18FF7D847DA8FD07 %7DA8A8FD057D52A87DA8A8FD0CFFA8525252A8FD06FFA8525252A8FD06FF %A9525252A8FD06FFA852525284FD07FF525252A8FD06FF7D7D59A87DA87D %FF52FF7DA87DA8FD29FF7DFD13FFA8FD0AFFA9FD0AFFA8FD15FFA8FD0BFF %7D7DFF7EFFFFFF7D7DA8FD2AFFA8FD4BFFA8A8FD05FF84AFFDFCFFFD1DFF %FF %%EndData endstream endobj 10 0 obj <>stream +H‰Ìkoã¸ñè?¨Ø H(R"EEêµM{y v®¸. +ƒ±hG==¼zìnî×wHÚ²o“ÛÛûpÀz`Kœ7g†Ãñ›¿Ü-ÎeÙ=êóð"ð½7oÒ^«±ë/}‹õ¯êzÆÞ ÞÞŸù¸€I^ñÕžñGÝU×^Z’%Fúí®ÑO£:óßžjYµä}r}¡ª³ƒËÔ‚0C˜û8¸¤Ô—×ÀtS[Ví6é>ƒî8òIûLÄ~ÈÿQÝëáKž &"n/Ž p3ÀÐDø…ˆhrY·žÝŽw}·ÖÃvu×—~ú¬ZÿZm¢ü¤VëŸOx‹®çú¹¯T ‚ç÷z;Õª–¿ú‹ç汫íë²jôp£?Ýwjï×Ë×°çW£ª«µ% Üh]êòÏ2#¯èª¨j yiÔè “$y…É*™ªº¼™šG ‹fÐáʆäa€X@XÌ»AÇ««0 =ŽoðÒ$:½þé_§q„ìáí{ð½²u9þï™Ó¼ÔÍ®†|Ût…ê‡5¿Ç÷='¸ë’pÿœ>ã‘ù:ê1‡úc¥?]ú7]«Ý>e?.ª_Àóˆ>æÃÞOµîÚÊ„—”p»¼îJ]ïQV¶¨•Ýœ|üu KÕoõu×ÕÓhkž{Dñõ¬M)agàv§Ûe÷£õïÃÄÇ!D3âs{Öì.fÖŒ‹]à4P«ÁÈÇò;HÂm_m«öRDþy3—w}U“µÏÝõþ‚Ÿ|ÅáëÜ„£n÷nCQ¤×'I.®`5oË´kLÈs2!µ-ä½î¶Ž6¿[ +ˆO;ï½ +ôaêF=€®Zû‚¡m¯>jŽdYé(’=QºÖeU× +åj=ÝŒP™ÝØ<ô0K(Ç¢¬6¤ÖUÕ°©õg¤ŽÕ³håÔWŽ§:á©fžÖª÷Pçx;ÇÛðv3oç\™ëäX§#«‡¦™·TÛ­îQ jÖo4Œº¯Í.½65‡§ºÖ#Ú©ÞD`÷„@N|ùXC”z{ê@[‰ÖÝÉöiDp +Kݨþgä|˜Í]´ÝXê ’9ºj5ë5Ó—hïßáÙõåFƒªªÕæ½QÃzªÍâÀ ¿·þaÒƒÙdÙ}j‘þ¼®Uc_¡ºªµªA`–ÚÀñ­Ú—nl¡§Õºéà¾ÙŒÇ•‹„®ëjgv>ìÔZ#é’!÷Ŷä¨35Ò–àÒ}ØB†[¥‡…ÓiWG¼C–ÕÇÊÈ´9æ?Ío›^¹ŒæSßYOíI™ý¶+«ÎC› +6¼/°Œv`§+MØ\ÏÙ£ôì ]ëøÔM”ˆ‡äI‰æ'ïÒUF>;—»Ð\9üÕi¹^ÍLWŽéÖ1Ýžè;ìûÖq<8Ž‡S5ŽTšL C…N¤]"µîM½C“·ljm„;ÒîD{èij·ªŸšZM#œC¸~FkrÞ2·]”þsµà¶8^áʶ©¼]wfv¸ôW/nÜWîà÷è%z)ènŒåœ!è3nœZ>ïô7»å®üÃ͸7´Ç»Kê5SË~úc–^L¯Ì`ÿ½üó¼š'šßu^ú8Ó¾tõ¨í<~Þwe3CÕ­?Lª†CU»î6>ŸôLSÑ–®<sḘ…‡96­Z÷X©º¬61h쌄v}WNkhìhÍê¹@·Þ*ßÂ"è㦠úXÄHí@àóÞ ÁQ¦k˜\ Žö½zæ/ºÝjŸDa®¡í hõZQþ¤ãŽ˜¼›g'ý»zÒ»¾›vWí¦óÞº¿wj|‚Y:åc¾Ã¹¥ï$ûCõQ;Ìý»³¯ê[BC4a¿}üÜu ¼GßS5êßW´X›™¦÷“~žüe×Õ³¿&ÍnîÑkø¿wV ½m]¤_ZÚ3|i ÙïÎ +pÿ¶ ~ÏÚSsƒ¶ªõk^¡Ï–탼×Gy»<<¿¡,íÉ®†æX'˜;ÓyÖµ^<ÃØü¾¶Loà?ãIØ,6o?êºÛ89c`Èðÿ­úÝ×T›4AÇ,á„Øó| c×ìº 4,žÔN[wǧÂr.¾¡ŒkÕÂlñ³JÓ~ ~¡Ÿýº%9ܬ”š¶ÚúÎÏ¿Ò©Ÿ´'äw¦-ÃdriT·Úúà0>>ó^ÁAÏ%~RÂ_¤¿yEQäEV¤ERÈB¼ˆ VÐ"*‚¸ò"Ïó,Oó$—¹Èyç,§y”‡9ÉqdE–gY–fI&3‘q/‹3–Ñ,ÊÂŒd8 Ò"ÍÓ,MÓ$•©Hy§,¥i”†)Iq$E’'Y’&I"‘ð$NXB“( ’à$ðd!s™ÉT&RJ!¹Œ%“TF2”DbˆBä"©H„Bp &¨ˆD(ˆÀ"àÏyÆSžpÉç9ã”G<ä„cÄEœÇYœÆI,có8ŽYLã(cã8`ËYÆR–0Éã,fŒQ±†YàÑ‚æ4£)M¨¤‚rSF)hH Å4ˆŠ(²(’HF"âQ±ˆFQF$ÂQafa&¡ EȽ0YHÃ( Câ0 ÉIFR’Iá$&ŒP‘‚I€ œã §8Á ¸åc˜(Žpˆ Æ8ðHd¹’Àž€cèpx÷ÿÍ*1Ó3W2…!2°˜…÷fu‚I»ÕL×@K²¹šõöm…J^)Tb +5:*¸Ë,Ä@X @j!È-AagØ} H oˆ Å̃1T á2¥Á3U¡4õ 5GŒ.÷Ü@ + Pä„Af p P/PÊ=™…ÿó^n+’G~‚|‡¾ñ¥š<À7]]UÆ–…À F,–,dI#!Ö¿½ÿø#ë4ÝÓ3»‹\ÁÎvdefUeF|ñçDC†KÓ—ãŒA®(fb™Vh•ÖhÚ@»ÒFÚD›Å¢íÆï6ú¶1tÓ+uË´Ò­vkÝ.݆ÕÂFþÐFDÐM›Þ‰!û³B¡@Y$J)‹L1«RY´Ê¢V½²(–E³,ª¥7®éUi¥[î–h Zv“òcMê™Æ¯3È®‘¦6Ð.4}Š^…–iÌ…ãZ8”|’‘F’’À˜im¤ñ³‘¨“õ„mLÚÊÄ-=y%}Sfj¥ª¢O)4Ó&é}eŠLó S½1Ý+S^’—aæ'-KZ3q­DU¼JaÂH.\Ɇ|¸M9aômuÖ¤—Ö)‡'<ĈK@dî ”Œ½Òiñ€é-¼j~1³ý<Øk—}ÉÌRX^±é¡›™½ó¬<Ý·á¡]ÌZÍîY{hõž™;ÅðÖòCK›)øTYÎ;Ø-¸ÛoØo¼ô´„±w ¾[ôÅ.Ó +eÙ…lDlÎòšaëÆAä1AH6ŒÈŽ_å;¹”]©Öz“qDrÎX‡ä ÈæT²A’W$ý¸ +“`‹êÀž$$kq4€H#05cO0€µ„”.Sƒh!ÛÕ€¾S›±©Å3 ƒr¿€` ºcrGè1á K²,œk`ìÚŽ îŒˆrXÀpß–Aª +T_í+à=]gì”ÕR67Øl`ÿ€0¢̈g‡úÖ P* kCöÇ)èNöl½ü)8‡¸çg¢—îãÔÛ}ÅŸây3†³ õTÜ9ÅÛ$¯uäl¾-j~yT°gÁþ6Ó£Nœ%Ë}y’Mzߟ%bN»Oz¹Ë‡œí¾Ä¯w?mç8‡³Ým¦÷ËÙnžç µú:óen(z5#¾ŠÞNó4Mãt]˽{-õZèµÌk‘¯†5^+<ë»V÷^Úµ°kY×¢®%] º–ógÙ|šF¾Èæ™B¾¯oÕñsi¼ +cÃ:¸×Å›*Þ4qWÄo•æ ª2/ºò-²ÒÜוr¹l­$à’~1‡^csõ%”õ÷>·–O T'Pý!P–ª·Ø³êg^)Žl¯Mø¸Rß1óáCØ úauÈxxĸ­µ¯1â‹U¶—XsSc_«°wêë¾¼­¯]]¯ÏK«yP[™áèØßsŒäÈ¢œT”Sùë£qš §’É®œP¦›ebÃør¬Ýžõ;¨R¦Ϭä…Õ\êycM¿ z]YÛGÖwTx±«5,ôŽÅ^ʽ˜pUø™•¿°úKýoÔ¢*Ñ#õØh …£8ð<ú‰DˆzlàÉD?•‚¡Q4\ôµ8—^“H±É* G5!z"PSDžFòs5aC£VpMŠ,â3-EzWuR]‘D)HW±Ï6—xÔ%ú|ÿöôôîçï¾=}¿°fàñ^ë’û G…ß?}Âÿ+q_·g5âÓ‹ŒN#YŽ$„z¹+^>J¾˜E¿¼*_Þ(`Ì^ÃÜ‘0/‰˜û2BÆ篺²¾^ΰ¿N?þÞÏþg_»¬Šš<7»s)2ƒ;ç,mMZRuîÄë ÍéÇœÅÙ·Ó?tD¤ ˜ Ò'ƒdìc-Ç`.>%„´ I@“4yyø±ú(¾…°ä+>‚Ô;]uLñÒætHÃ-¸x#§S” ¾¼DŠHdiJ%xv‰¹Òw%«¯OÁ ëÇÄ೴ו}@zñ‘ü¸ä,­[×°È´Ü°ì!ë¢Õ ~F™¿¡@ˆŸbéÏÁvm ­a¬?Pú”+¾8eñS +…¾+ËØ£"+áµ ÿ7D”¬GÀ åBÚ[G,onÒæ¬gT$Ç>» +?c=ä¶[Ö,œ¡é3Ûš‹Ò¥´Äo‹Xoñ[Êœ"¡é˜x–Âï ò +ñŒSŠgŸ"{`ྠaƒñlªXn¸Å:úµ¨‹¢ÈU–ë#J +ÜM—g-‘›é]©tå[Ň(X† ð¸ÁA2DüZ=7<É:ÀÏ…>¢nQÀ8Bôœvs TÒµ=ìJˆë›ádÄD„ªÐ>‘}Z+ú‹¯ätvS­e`ßõÕœ&ZDñs (˜šÙ®+bÂ<ˆ•(ÏÁvâ{ÄOØ4 ‰‚B.>÷ªM*–}¢cÔ@Ì4Náe‹ð¿EÂÁ‡¬ð[hV>¤ËclÉL£Ö,ï[Wó’üÄּ׈LèŠØW?Ñ/mËÝ1|"ûT] UöÔ¢K€é QÆ&uÙúŒ~°~ÊU†è•13þ¥ˆHGÐI›+éÄu~#| y ´y}ctB2䲑N|óžthª9n¤ߥtâ‡t$]’´Ì+餋OéèÛé8kÙH_±ÝIG?I'm\»N:ñ›ßH'sBH'íF:.šÛHßwò­¤C »‘>d#]’¤ÌÒ¥ Ä…tð|±éàGߎ¤C[¶m#|èåt bȺéÐ ]VÒÁÇþo¤“ûµI‡¶@ÂwÒÁO%o¤ƒSÉ‘th«-­¤K+®¤·¸#餭ætô¢ êà;ùØ=ê¤Mpµ ¾VÔ‰oíu2¦Ù u»9uâ†tD´¹°¡Ž}ʆ:úùˆ:¶… uâ'¿¡N>)x@,d©긟nCDH GÔ!š +êàÊqEãÎQ'±Y܆:øÐêàsìuã2~AÙR6ÔѯGÔ±­m¨cÞ„ ud‘= NXÓÚ†:úuCøµÝ¢²?†v 4¥œVÔ‰*ð?æ˦玣ˆÂå]:RúûV¶²A,@D¶–c p"ËYøßSçôLMU·Q¶(‰âSwî\OOÏ3çQÔ!‹':ÔÉ,·PuÈ! Eä´ú÷M:™ÄÙ‡’Y¹éHWÚ 9|ZgRÈ1¯vCŽ£XrÈqŠCër 9£BǬvGÆ!öÕ•q¸²ž§2Ž«U†2Nr‹q8ÆÉl„5#ã$OyQ)ã +žFÙÚ†q2ÊýmANb[`%ã$ö¶*”2Nf³ä¢Œ+C44eœäœKu“Y½z!'™ksCNrK©9Èɬ_ì$ä$Ë»½+ä +þŽc8ÈÉ,¥ñPN²|µß”“(¯á(‡YZô!åãÚ¤2.×R3©IJ9ɵ®¦EÊ!çÜå0K«7’ræ b»6ÞM9þôr R¹/˜’rë;ÝQŽ³U=H9þÌzÈH9¬ÈŒÍQ+e?Ü”ÃÅRÜ”“;^ÖÇ +9ì¤Üó ¹¶b/Èqã¡6Èqs–®Cέ(ä°¡kër/¶Úr|ÂTÈ1Çé ÇY. +¹…™¡#†r³#¬y9ærÈiqÒA.S\1{Ä•GqEÞÅUfN\‘­¸JÞÅ#+®’¸âó]\óp⊯Xqå)¼¸rdÄÙŠ+ó&®˜YqE¶âŠ¿Ù.®¸`+®\4#®’qÍÓ‹«d'®yâ*##®’œ¸–pŠ«Ìœ¸Jvâ*y×¼¸Jvâ*ù×½¸JvâZâ)®2³â*ÑŠ+â.®˜YqE6âZâ)®2sâŠlÅyW̬¸šssñ×½¸ò#®Ì›¸rfÄ•?cÄçÜÅëhÅ·ÓŠ+6È.®ØGF\óôâŠm·‹+·¦Wd+®Øλ¸æáÅ•€WæM\93âŠlÅ•Ù‹+FV\™¸"M\s<Å3#®ˆV\%⊙×¼¸"oâ*#'®ÈV\‘wqÍÁ‰+±âÊìÅ•g5⚃WæM\1³âŠlÅçÜÅhÅ•‹fÄ5ÇS\sôâ*Ù‰kŽ‡¸ædÅU’Wɇ¸Ể«d'®9â*#'®’¸âó]\eæÄU²Wɇ¸ÊÌŠk.N\wqÅÌŠ+³×\NqÅÌŠ«d'®È»¸â;V\Í9€:Ä]\sñâÊcŒ¸2oâÊ™Wd+®¸¼]\±V\y?¸b‡ìâ*ÉŠkŽ^\¹ï6qÅÞ´â*Ù‰«äC\±Ç­¸’-F\™7qå̈+Ÿ#®d‘W°ÆŠ+³W䯉kʇ¸bdÅÙŠ+ò.®2sâŠlÅUò&®2qâŠlÅy×”oÈáS+®Ì^\92âŠlÅyW™9qÅ1F\wqÅ•Yqåjq•|ˆ«Ìœ¸JvâšÊ!®2²â*ÑŠ«ÄC\eæÄ5U/®’q•™WÉN\%â*3'®’¸¦vŠ«Ìœ¸J¶â*ñW̬¸"[qMíW̬¸Jv⊼‹+fV\Í9@9Ä]\ùÓF\‘­¸®ïxqå̈+ƈ+VdW¬¤WÜQ+®rÇwqÅN2⚊Wn¼M\¹9¸"[qņÞÅ5e/®|Œ¸2oâÊ™×…™G\‰!/®„•Wf#®È_×XOqÅìWaÄyW™9qE¶â*yWŒ¬¸JvâŠÏwqÕ‰+¾bÅ•§ðâÊ‘Wd+®Ì›¸bfÅÙŠ+þf»¸â‚­¸rÑŒ¸J>Ä56/®’¸Ævˆ«ŒŒ¸Jrâû)®2sâ*Ù‰«ä]\eäÄU²Wɇ¸ÆáÅU²Wɇ¸ÊÌŠ«D+®ˆ»¸bfÅÙˆk§¸Ể+²Wä]\1³âjÎÎ!îâ‡WcÄ•yWÎŒ¸ògŒ¸âœ»¸b­¸âvZqÅÙÅûȈkl^\±ívqåÖ4âŠlÅÛy×X½¸ò0âʼ‰+gF\‘­¸2{qÅÈŠ+³W䯉뀷–î@7à­á©sÁ…> n@\ët¨×Õ¹£;Ä5ME]‡¸V‡ºqMMQ×¹²¢®S\«C]wâÚá­ [$]§·FK:œ¤¹I×AíKdA:äëÙQÒu.]VÒáAOë"é:¼u‘PI×é­YI7è­é½59Ò xëêF$Ý€·†®¤ôVWêO²IÛºIqŠº q]XSÔMˆëhŠºIq­êÂa®2ræ*Ù™+>ßÍ5o®1xs•|˜«Ì¬¹ÆèÌq7W̬¹2sñ4W̬¹Jv押›+¾cÍÕœƒ¬‹§¹ÆèÍ•ÇseÞÌ•3c®ÈÖ\c<Í iÍ÷—~³nžâ:è­ñFÝ€·.:uãôÖ>§’nP[³’nð›Å‘ﶴ*I7¨­II7NmÅèvN€OMNO£´ÖdA7h­O£´Ö¬ ´Ör€Nnv³ùJè­‹9D]@ê½+ëÌ5 ßëÔu<î ®\ž‹vîVoÜ=µ‡wúºHàúëF¼pÃ;^¶gTÚ +l˜w˜E<~7ïÂZѨÀ ¸ cQ‰Ö">ÈÃ_çÚ½d^€ÇöêM6PdÁH½@“ME±¨²±9î¸,Îuƒ—KâjÇ£ÍF×ñ`³EÉÍFÈl¾âAfóýD™-ÍBfK÷6 ™]l\6Ë%ŠÍBfëf³YÓñÞ÷í±ÙL™|™2{•<ù;fÙG×*|2»^RJ¾L™J¾ ™­UÉG7ÊÍ‘/CD×~"ù2dvÑ“äËÙ••|™2›•|™þ‘nòeÊlqäÃOË[TÉ—ox\äãwzväË^f3]ö‘Ù|ºl¦Ê>Ü™jêOÇ“¹ÊØÓñœË¶!wå¶] s8òas–‹ŽD_ Ì¦¦ì ´Ù<üð¸I)J?< +±„¦øã¯; @ ó(C hD:í2›R»!hµ«n‘‚ZûPðÝEÁo_úü݇wŸ?üüñí§//¿—Ñ«×ß¼|ûýçO>þóåÕ÷?½ýåý›¿ÿøã_Þ~þéß}óò;ò§ÿyÈß¾üòþ:ˆ`½hŽ}óæõ»w¿þç¯?~‹Ÿ\þAþ“ÿN¸Éò¿¾0Ê~”?´—Wß¼üðùÓ¯ÿùúÿüK¿îïáñÌÓ‘£ÚžË'xm8r9æè°µ0ëÈ—‰ê’ÓºU7Óɉµ³‰tdÙ¿Jtä6¦úzÞæE™ºHL 3¯qóœ£Xç„Õâ4iN¸ __8¾T/–ã˜ÕVIrÄÞ<ÈqeBrå8W« Å8ÙµøªDZ!>—ÿ Å›2 ˆ[†“CŠ‹á¸Nq,C[רÇJ•\â@ˆP\!Î{Qªƒ8nq]Í“ç-ŽS!Î)yˆãi“V¬ÇSæq<êBqqPC(®V¤ÇÞt²…8fiÕJBœ`*M!Nvåâ ŽÙˆO}ïhlÄ‘sî☥e5„8Ï‘´¾"¶™ÄùÓý©¯È}µdB|}§;ˆsf(Οiű";Æ ÍØã8ßµÄ8.k}¬ÇjŒcñr(Šqn¼Ź9KWˆ# Å•á¸gµu‡pÞëÚ•àÜ/kÿà+OÇoÎòƒoƒÒ›ÊÍ›°ÊOeNQќ֫ð7ÑýæÿÝ¿‰m¹Ó³r/¸J^…Vó!w¸Õãiä²wך›FΗ²•¼¶Rª©ä²g✾’§šå¥l*y +5ôa*¹ô=V[Éå3Þ:mæ rXL3O—¤ØfžB_ß¹‹y›×³~óVCö,'䩬¶˜õ¦×b.ÏÀ²Ó˳¼¸ú0½¼†€:oz¹ì×6÷^^ûj»$º¬A&ŸV.€î¾•³˜VÞÂê±W/q.|Û^>嶛^žäikù ×kê©å-®÷ñÕÊk¡ÛV.×÷V>îRÞêbó]ÊG™m/壬â~•ò’JVž£±†YöRq±¥œ|}:9/Õwòš{3•¼EªäSÉ—¾úJ>B¦’KK¨ý©äU¶ñÜ*9¶R4•\ªòÚlW%—UòšÒzð®N^ËÕw®N~Õ|_ʹŸžRžb,Ù–òžcÙK¹ì‰fZùh5)αëâ™/—$IRˆÞ¨ „Äg=gšû·»"¨YY/ÚÚ\•Id€ôpûmÊ{‰{ìxò%æö丫檿<9CaÐ÷xr4²­¯'_qO<žÜ>v<á³Ý8h@—óºqþšùuãsʼHG«ËßÌô÷¿í¥»þ”R+ÝÛÏ’yî¸F=ÇuÒ”l›ë¤•µ˜ËÆþ¥ +Sɶ‚‚[Ê«>:jRjúhh=0è¡›À‘o )]Fgl/tYm¦‘î{NÈ×I³&qͺ“†öö;Nšº¬üYàxe­¯pªtÒ¦á–ÝICC曇“f÷V¹Nš2rG8iv Ù5+QûNºÆD¤“ÆätMcV5‘‘^š×_Ì}xiÞ~#mTxiB»ÈõÒ5–J/Í{õuÒøßirœ4§¥¯ë¤1_õø»ã¤Qó;ñ8i~†×ïqÒÐàéë¤ñ·VÂ7vÞV¬·k¥»Ûçk‹i¥yEVZiênšVšgoð±Ò~8>ÍÇJû™×™Vš=Ò4¿ãVš5¿ÒÏè5Ó륡1mõã¥}B¤íßB/ÍöÕüÎðö¶ùXiN/ªc¥9U­¯´Ò>—#¿ãV5¦–V—¾+ é§õµÒ¬õh÷Ò\bÄT¸—†æ=ôxiÖŠ¤•†T‰as+ ÝzOûíVšµ¶fLAqª8Ó•†®z†7”ÑÒJS¯È™n¥‰¡Ñ^/Í­àñÒþ™ˆfÑ{ÿüÿØÝ4å„À€<åÐغ…žÖ£&úbóµ¢Ob¢º–š¥?úr”ƒ:êå(¯îÙ.Gñ÷úr”c¿ô´ҵ\–R’$_–²VÆe)‰` –Bë”—¥¨µ˜ñ`)YbíÂ?­öú´†;K˜rËT.LÁ‹ú…),‚ÖzaŠÁ©¶¦|­è¦äAë3aÊ÷Ú¤tšr¶iU¿4åvͽ]>°K‰RîÿØ›×ÏÇý'”R—Ú¥ÞÁÖD)›§ï×$J}zV’ÔnÊCRÎ(ŒK’”s^.H)=–~@Êš­™ uÀsRÏØ”)jH!-AJ½49J9ÚƒQ–z¥c4WØ¥.³?õ'÷žåg8£þØ˃Q/uIŒúcl$F¹d«E¹VkR”«-íIQ¾£õö`”{3‚yŽÑ~€~0ʦãUúÅ(S“¢PÃæ¥hÍ}¹­/¢ìþ8Ó€(um/Dë>¬QêX6 Êvk/C‰*±ËPêj—¡u¿ãËPÃSkìð!K-6Û G-f‡pÆß í!kmõM8ÃÓ&RȇoÆ•Fâ’>ûàZC'ÞŒ£÷³ãŸióòÍu{ðæ¥ÒoÐXÁoÔsÖo¬9ð7Þ¸F1I¼Q[èÆ´V“n¾]špƒ”U¶¡ä¯·ÑÙÍf¢Í8…K´™1 D´Ì0†Ž6hUÑm¨[çhƒž»nÖá;ÚøÐ •˜ÐM7è„0gºm‡›tCíÜIN7hŒoУ¼>%\;#éfã§h_‰7ê©ÝXZ«'Ý\G4pºAW‰«+éÆZ¯5é ±oÔUÖÃ7ÖJ¦Ïq®±ùF­S¾ù³#XùîQ[¤&ç[|g}ù楞xó§ì #Þøv¾i¾aÓБ+ùæ§lr¾±CV—‡oè$˜MK¾™F§¾yçi{øæÍÉÑ›ÿUë 8#Ål=€c‹›®$œO@©I8×æ$œ×¤'á\÷œ3HæC8GU€Ç 纶$uÙüNÑ 6ìK8–ÜoÂQ÷Ȳ$e«ý!jmÄ!bd©j˜Œƒ”7\2ŽµÎß!]{x +Ÿê:!§œÈ¸=rü̘’slIÊùºk%å¨[I9júPŽµ蔣Vµ¤õ +S—˜ãûlz:æ|ÏÚåtoat¨M¾Û!ôZV“tʉŒ™¤C|@y æ ƒÆm9СvL—ƒN'¶iI‚ºYûp:’sÐ:‹%ç ûæ^rµ¡Ö“sÐèIÎéÄ"$èP“m^tÐmvIÐéÉ5Ò±¶mA§'mÐQû»~@‡š•b :êÑ$AGm"èXÓ8N]®±AÇçÌ1Ðù³m&èü3‘kt®wLÙ ó’i’Σ+IÇ¿s‡¿¤ãNnÓë¤óãl3IÇiféØIXrÒéaî&w^é¼;›ÒQƪ:öó°ö€5[* :G˲ÏDÑt^«+Aç:ÖuÒ9‹DÒùºu$é\—‘¤£Öº~“®¡E×Ð/éPšUoZ¥.:éøm“t¬Õ97é º…ï"é(KÓ*jfÛ¬“tÔeÜ´J-ãM«#9oZågjX>'ky㪯 ÇIáI:êy%IǚÓŽÚ3Ø&×láï’t|Ç&7®úžµW¡ñ›Þ¸Ú(5'4®É]ãDÎ7­¢6,­“‡}ÓªN^IWÒ€é +¶i]мexÿ3ætÒ~IW.‘tøûüº +÷e7®*óù A‡‡ÍñæUÔlö›W¡‡ôzAÇð[xAÇ€2o`u-7°BG +ý’5-7°*S«—t<©/éP›r#k®qH-ã¬þìˆ@A:~&òM®ÞÄrHÇ’ÝÈêÑYùzör9ÊM¬~˜Á(çd˜>œc#-»‰µ9rgrÎû®½‰Õ{³ÕÃ9Ê¥7²B—jode‹7­ :'˺™ÕG¢¼™Õku&è\›$èE/ç|Ùz3«ër3+õ¬ÿɬ­Ââôñp%w²‡s•Ž&1UË|1‡6·8BÎ+æ`MKÌAjÜts%Æ=1­}\ÌA÷ÐsȸEsøŒ ½˜£®ýÅ×]õb:RLP²ôþR®D‡$å8åm\Ênƒ½”Ã+nzåjeRÝéYáK9pe-»”«è:msÇ×Î5,·L.å$ 攃VRäK¹&i¿œrÐÓ/‹¹†Ýk"Ì¡"ÖVbºÚ‰¹¶ÏäË9ÔTt&ç {ÙŽœƒÃúùÆöê-9×ÐdÕá\;äÃ9Öê”ä\;1fsŽZ〓s¬5’s K49Gݪ>œcMâ@s¹Ææu´gbÎ}ƒ«"2ŽS.¾a_Êyé“\©§Ö¤7d.}8Ç,á’œs~œíbN°:^ÌaÏ[0)0wx{0ÇÆ“õb®r51ÇV¯˜Có˜Ú‹94åhz1G°Ìq1G½ú‹¹ºë`®îìt0GU}9Çuʺœƒ–µ.çjœÉ/Î ö}ôþåKKor¥žmÐQn£˜ C Ss’+U ãEÐAÊ´7¹Ên‡:h„Ò›\ù÷Òßä*Éq“+¿³Âó9è\×7¹úº7¸RZY :jíö€Ž5'ÿuo=AÇŸVÃÞ%èøƵÞàê[&7¸Bw]op¶sÔt‚¨* :á<Ž—t¨i‹p뤃î;É:é qg·t¨mûå Ã.Í\¡Ûú*Æý< ƒN"sx ûö¡ :ÔF×›\¡—Ž– “ˆõ7¹¢öÝä +­ÅZ‚ŽZct¬Ùõs.å&Wêùb%Ø×\©—´Äõ¨íÁk½ÞàšklÌQ—þWt¤ ïlÿLDëñW/é ®ÔÒnpå’²ÞàÊÔuƒ«gPÊIÇ1íéØICopgp‚ÎïëeÓ«ÙMá¿r—3šøûV‰²A,@ +ˆlGɈŒ@C4ó頻¶ÛÇÁ +%Qnõkûœcw?®NwßÊä,)oÐ%+ýÍ9¦ó,wßÊXÌÙ9'°ŒÓ·J¡s*“ƒ9ÉRsÑ+;§eÂi[©Û}+òé[ù{wßÊLãô­<κ“ݤC&ÕrúVòµÇî câ©çy€ɉželÐ Þîéô­Èç®nì:Ün3ÙŽdtÔ£Í t¬“`<è„šz@G…ÕYnÐ!Væi[)ÇÌι¡Û¾¼æòsÖh_½AP©Íãè?¶Øóˆº@«˜òÍ:#|MY°C1T®7¯Ðxc{AÇ* +¥dÄï•gº‡’pŽ‹x,M´É™‡Q±fóœ‚×Q‚>©Ç`¯É©‡ú¥û.Ž=jÈóâ^àMÀCØàC~íaò°nN3^ä# ”œ}ÜÃ0—bHДŠOøX‡>ãñy@@êñø<”hËýöy(œÜN/ÙÌèáŒtO£'Wöhf±mœ^öOå ?|{Lù´²„t8ìÃÏ@a¼[YX²œN'‹¼d5oôeRmÞèÃqšíYèË*«ÓÉZOrw²ìcZmŽ¾Ý×,ò±õÑ—>È—ñ#—ÇŽK@YäË*ør‘±uE*±|E¾ÌžÄzNGb!?\Çt+ ±OsŒ…›} %+=ít=û0²ô^/öá Z¸<ærpæò°#9–ÛåJyµ·ry¢p:.S½à‡äl‹»ÌÈ™jN? 8¦1ã…?¤y6š/þ‘8ÝÈ,þQPúÀ`§–œ€BÐÌN@êÙÛíõ0‹¯4‚ Œ4ŽÛ te¦×üæÛÏ_¾ÿøÓ—ÿüôþó×—ß"ôæÛ·/ßüðåóÇO{yóÃ/ïýðÝ?>|úùOï¿üòûïß¾ü†Cþð_‡üùë¯lÐïðþý ßï„Ó?~•Ä?ãöòæíËÅ_ÿRøëGþówŸö¼6É™— +#Ž\²Bº3–:/ 肦Ö5 C@ëbb¹gˆ§cd©gÑf%w~M‰ØÇ— +óÞÈ‹©[†`ÒÍ4à¦Ým¦Tµø•º”°tÖj9âZÙO¨ÈUÄRëz©ˆ›[:F{B°'ÄÚü­ûÈXÈ„Æ jA‚[u§Ù\SU“±:ú´ý*“‰z¸Ã%•£kŠàP‹]1ÐJÀŠJmH¥Î« +eAW6ÿŒ§ˆñ2äí8oH¤)$h®)ìXSt'1ÆbírnzŒX&DKŽÐ–5•kÓ:l8¦´¨¯uK¶#²õkNçÍÂ/ÎQëNT”Æ4¬ÛyM Û‘Ð×sˆaT‘v©sHn]Sj­¡q¬¶ÑíL©‹ƒ´§ŽE¯2ÁIi}.ôÙ猨3 ]‰Xž9&j¶“̉}NMYs@\ÍÙkÈöP·•{mqSÏîÊ-¢–SºNŸ¢ÇˆO{ +ŸEÊ´ÇX rÜ’‰[†µYKÝsVíï.(\#eâýØiœô«Îjå@b2ÆÂŽ™›ØÙA6K½ªjà==¹I =4/¤êî.)]›~ï4/kNíU±&Î*" ®¥UQZ^rͱÃö mú[nD?o»&^|2]•t2xçLÞÙ€Vûÿ„òwÿW(Ó¾À¤Ü(¥¥©ó zÔvPɺr£6g?(…FÚ”Fb¦Ü,¥•êñ°4ò²Ê‡¥ø=˜v–Fë0¨‘×äá)eN7O ýð”†(Ø<….#ÝãÅ7ü­[xÃÿ[­Ãñ¦GÙ:Þø:ÊÖ…7•å2ˆÄ›v!• oÜ)™»…7Öû¬M8žGÌý"O™ é&Oy¡X•£ [þwŽ‰ZvÂíÜÞˆc9„Û/²¨‚DNeÛ¦#Žz”‹p ÍÙœpÒ¶y"Ëx9Y'c-F'qÍ + +qÔ¬Ó'ã 9:ãkˆqBÈ(ãôìV7ã(k(Î8›2/Æ)ÖqzŠ™ǯëÖq9ãˆæfœN“|ZŒÓG¶t1N›Óª3®Š¶Í§Ì+ùbœ’ÓšHAŽgÍår:esA9eC™N9U@¸2g/Ê)–Ú¦œdsȉBi\”­Œ<¢œtÌN9ê°(ø¤\Ášµ×‹rŒ-£,ÊQ7ëiE9êlvÝ)Wø¦c:娣‘XœƒÞùîœc,¶ì ƒŽ­Wõ0½AWX–v‹ +tÒ͈ tÒ©\¤Ó²føD:ꇓîßd×ꨂ àWÑÒ¹Xÿ•o² à'#6È MÞ>çÕdSZ]3Ós§ºêkÖuõ${b«t¬{.ëcÃSÇŸ£‚^êtÙš[‡r6ËtŽz›wðÑõ9C\»ÎC¹wÒ½.Ó±Cé;Ôxh®„z6z]ìúÆ*êØ¡nü1/ëÐËžOjê¾m‡ë B=¯}nz«›ÿjêSm|Ó{ߧmgǽڗ<Ø¡l{VÇ®?!ç¥{6¾)výÉI;ÖwÎtìÐÅ&+ÅŽõ2,Ö£Ö„{݈ŠÝ늿g¯°Óï¶'±Ó·œåØimÏOÇN{£»vú5Æ…jÇ×qr“v\ÈvÐNogÛ®wH#iÇ4ÆqíúãîÕNw^[I;Ý­ºv¬ÍaÅŽz–°Coœ^;åå ÇNÏDé ;íÉy°ÓÒ.«Ú©Fµ&íô²²\;­ËríX÷{É·v ;ôX¢tíÐÛÒ#¹²æÍy´ãgî¼èÚ±'{»v¨ç8Õµc]fN®è1«kǺ¬H®¬ëJɵñ\îH®|‹ØL¦Úi]stÕËžíÚ¡ÆùZ®ëi_ãÚ±7VDWÖ»w׎×löqíø[èªËÖ<º¢<«äèÚ ¶žj‡OKÇ®ñPîœ\Ñ[ÃÒ­j‡ÏñH®½ðá9³våŽav«t»‚ÃxÇh×®ÀŽ>C»â.›vx}ƒ` ]Q×yß£Ø $[9»¢7öôìŠrÕ)“ˆ-a`Ç´²#¼j]#¼¢Þwæ íÐë%Âkg̲±Í´Ãë§JÖ½]#¾¾®aÚ¡®+ÇWýî±\;¾åÌÐN<·„v숯ú5=â+ÞÈÖaWñôª7³-·ŽD“ØË:n¤3"½6ew»uºïZN¯º7›¸u¬OøŠºÈÈñ•{¼YVPìT—ùUDÉùU{²ì´´õPì£l^U"¿j]"¿²Þò}~m‚1g®lzw¦5ë„cMP‡Rl²ê°ÏÇ– GáìÔ¡îcgêp|ÎAê>WP‡zZíÔñTÚÓÔ¨Ã[† eFk™™:^öHP‡ÚÒˆIWì€'éŠÇ=“Ž½­õ²9/¤Ã/TAé¸h–¦T:ñ‰;¤“ýœÒ •½u<‘y¬ko5¤«¡¥C'çIÒ¡gCØ•õÖçÅ¥®5üÊZuèU~þ¡µŒ³œ:ÔíÒçÖ¡×kßnêYlSëP²™¬CïÔ|±ûVb—j]{òÈË:öÄæEµ®=™æZÇúÎœn{ËF,µ®ñ߶ Ö±nÒ“uìU{$ªu¯k¨u¬mw:uúÕbõ ZU:ûÀHÒiï•bYoÃB¥ã‚ìÓ“u\ÇrÆcÞÎÔU,G_™:l¤6$¨{Ì}¨ãÆ«'S‡§P©=¨CÝlf3êÄþ£Džn«õ Ž¸ìÔ±>3S'÷v]êäF¨‡:Z$=[Ç˔֡®ç„u¬ï%ßÖUlÐeéÒ­cïôH±¬w;Žë;3:vèµ³"Ųn»;v¨ë9Ų×FwìP —èÁŽ¯—™Rlå¹\‘bù‘c3™b§µä«—Ër”ãرîö-Ž{mFˆe=ÛtìøŸ‰yŽ±H„X]µê!åì'‡Xô°®Ý±«ø+½:v•GreíÐëÍ‚®j‡zÞT«Ú¡Æ£s'íj·!ìbWñKÏŽ‹º­* ;ôÿ¥;ÔËz„*ïÉRìÐ[zp/v¨që›cWqïçÌ)=¼æ)e/£9v¬»-¢cÇÞˆ¹NË)–õÎÔ¡…96B,ëc#›RÇzIKÔ±7%BìëJë2sˆÕ¯îçÁNß²·c§õÊ!V{=B,ëÚ"Äò’õäËuìÇC¬ÞÎz\;îÁÛðÒŽ;iõ±UvìtãÕœa¹9{mŽk €j÷óé9ò'­¹uŠËŽ «µeZ·N‰S§•-‡R§}3ÖéUJDXÖóD„Õº|aË.ã$êØ»­RǺ¶êÔ¡6Ô:vxžèPϵ–CǺŽäZÈ–;ǺŽãαîV?ÎÉÐãð0Çw4È”9­ËNÌéUwsç„'¶TwŽõ;9ÇÞšâÎñÅ2’:ÇzØçÎñªŸ×9]2ËQ\”g[ìrçë·örç„›ù•_…çqÕäœpŸõîΡ~ #sÂÍ]$1‡ž `×9ÔC×9ÔK¤'çÐ;­‰;‡z·}Ü9¾~ÝsçdãE~EY·ÍpÊj¶sèÍ2ÆÃJüGÓ™“'‰¼˜cïب¨ÎÉ“f®s¨÷7:ö†=ë:aβÙ_¡c-e&èØ+öÐé[,¯*tö‘• Ó^ÛëiV(tüy@#A‡^Û{=ÐéýäP~¡ã9í$èЛқC'¹W:ÝzU’tÜž[¦K‡ºJ êP—ÖV¢ŽÛ|ÔéÔ).F…R§õÞ‰:=)–jiY³Ü:µ¨Ìd^¦ˆ[Ǻqë´¾—|[‡:ÆÊcz«ÙÏRë6gšcÝæĸs†ÝËZ2\»ÍU¼™•ÚátÌ~Ǽ‡;ô0ñG†Å‰š¼ÞÃÝÒ“ÖßÜñ`Έ°x‡ —;ÖŘrîxÕm[[¹ã¹-F¢r‡Zì›»¥€êÜ->áVdXÔÈ9Ãòü—U;®šÌúp‡Z·Í/ÎÝƤ°GdXœ}3;Êev9w‡ ¥NçîpD†=|pž<ÕN`wò#w«tÅ4î +î™´Ì]Á½¯‘aQ?2w¦Ý)Õ¹+¸ -2,êYçï +¾oä ‹ÞéÃ3¬"[ᦑ™½c‚Y‘aµ–È°¬ïÄÞ 'ÙȰ°eC›y‡z”‘½C¯—È°¯k˜wÂ’¹ãWwqîøŽëì9Ãj¯E†å—––¯—3,×±nÏ°¼<–vØ!@(k‡4zdX»d9vÜx}^Øas"ºlÇnó _#ÃnnëÑvxÀjŸQì¨ XpìXïyv<'ÅP%vŠÍìˆQé’°C¯Ÿˆ°,yµÇº­OüþuØžgˆ¬„]Á¬óÄdW8Ùô›j¹®…ScmÙ;4ÓMwðpx#È¢ÿÛþM'§ +§Ò¬jÏíè¡xö[=žODŒêîáM2Z«/sÎ7ò±¹ìS*Ž0'ñîô±1J;ɾÂ'Bµô¡ø¡¹í5ìậIú‘ƒ¢¿ãòÇe,Ç̤…¿†ú¼,˜ –]Éæ=|c]ónÉä“ï=ï!´´¹V^ºéÀ»õ>¼>Í^Á¿îȵyO2€ßX[ÄZR]Â?¼ %ÇZ,G«‘jq‡©ÀÃ_#m'ó×pdu2þH°Ñ"ÕZ@É©–¡fš¢z¹'ä\ý˜ƒô§¾ôÃJµ]bÚãj**W?¼¾Õ®{÷1©ú½®¡‡«1šØÖqþÐ+-¦=¾eÙvSÿô#æ¡ûÇ^µÓ§þ¡îWõ?'u$ÿ°sÅ´Ç›y'9›ö° Çÿ=í¦v£®N{*qiO,9€ØœóÒ«¢>u¬í·mÈr±ÏÛ#ºHun +UYëÏ#Xì®ÕGAEÈHSYŸ5óÌWøùjC–BÈÆ®;¦¾Âi Ÿú…?ýüåë/üóë~þôåþŠÖ?ÿøá§_¿~ùãó¿?üðëïŸþûÛßþóÛçýãÓ×ßÿþËþòÖË^g—œ·rBÈܶÛ?"ƒ€Äæ@„HöžiYA„vôU¿cÏØî~ÜÅG~ÿ_ùãÏÿü[<ôüÃÿâþ=ô履%ñÇ_ñÇøñ«_ÿøéÏøë_ +ÿáù‘ÿý=Ç•wÜ@SƒLë¼¥Ð*üã 96—1T)eGC¢»4zifÕ_bH@e§ëê£>)GÍÄëK•øŒ°=#„F•O𢡶RШˆ¡Í\::Al5e 9ÎDŒƒO†)„Leˆx§ÆÝMí¾ãeiFaúÒc i©gÐBŠ”°GÔ˜i½CP‡˜†Éí»Å¥"myÁAƒ®ÒäÀ;f•ˆÚå´ +fòžB“ÕioâËj¥Ê.YHm7àB§Ê+µ×ÏÝGŒulðTç/A?B âúÝûÓšöfhžJ“݇iq¤µj¼Þß1“× +côF9I“³â†^Zì@“c€6YÚ$.tScôi•YûÌ£{†ôZtœÈMÛ™<ß1FËÅѵwÓ[¦#V¡C»Լϩi)™"`Í;ÆkÓ˜±ézÎÉN9"ñàbz¾TTnV¬œÈRê¶mϘ©ÔÆÜ9†- +ÓàŽ×“s<ÄU<9Ǽx¸—œcjvÉ9fw}Ú6‹*‚¿êèXY}Z‚NÅ;ê ºþ›é H týõFè¨GøÖb^Š'è¨gtúýiFÜk½@ÇX»P ûÌ1£”ûšë^í+A§gÂÕtÒEyA§˜÷$^Ów’Ž¿£j/Ò #¥%étžm%é´L÷‹tÚžÀI×_æ>¤SÞµy‘N¹Ùj’Ž:f†ˆÇìíS{÷š Z¶'èTþb¶tÒÞ’tBQ­_ÒiZ›I:é2“tQMû¤k˜r‡LÒ!¶¬§[¥,}&é8äi“tŒÙZIºÆÏØ5IG]ÆåVrúu’ŽºÌãV©ë¼ÝjcJ¬t«|Ä¢é¤ëeW5kX‘Å5“tÔ# K’Ž1Ñÿ!õê=IÇ9[ôwI:.°ÕcWµiíØUè=ËmWÒfDL¤ƒnÍt%¹.·ŠÐTÊé q}·Ú /Íq“®dû¤+Ø£}@Wpt³÷qHW‘E:ü¾þt†öË]…®ãyF 3lå¼ý*b¾Æñ«Ð³FvëËè>‚Htt(ëVéz +ô*qÒ!ÖË1¬Ö*è¤Ãï»ÚM:ÄV=–õ3‡HYçmYõêÀHŽÏ‰‚tvÜJ’Ž1?–U¯éDzry~sû8Ëq¬:Í€”8Ç™Þ/Î1Ôçš»’sʺv;Vef³äõîDzBóÛ²2Ã[·ȲgUE”Û³*×­@'"ЉDç4«Ï*]Žg¥^öKÏÚ µ:æÍ9‹œ9g¬ƒ9H‹†ú`iîqˆ9TÂ^~0Ý}]˜CñìèÓsÐ=>$0=Ḇ(ã&æðˆG7˜£¶qaŽ³n;˜ƒ ”ƒ,cÜ”+ع±åXåmÊAÏèïå°¾RPŽ[­VPéY¢<”3¤âöC9ÃIE3˜³(÷å*ØÄ,x(W“`¢4nÌ}Q±·ùå —GW.̵†¤¬õÂbÕÛNÌA›ï™˜ƒnö’sˆáœWrz”§#ç Á°qq±]£YçZÇ:‚•ú´×ƒ|8ǘE /ε×Ç<œ£îq¾É9ÆæžÉ¹F‡m¿8Gݬ_œc¬>-39÷™ƒœ£óÂœÞ|Œ«ž‡#ÊÅ¿(§ØǹR¯nI9îÇÚýâ÷±lOÎé<ÛÁ\Å~ôycΘ“–˜{yûbΞFè‹9Ü@¥öƒ9è½Z`éìÝoÌáf›­Ì,Ok/ÌQ÷æì9®s–®#0GY¿8ÇiÊ>œƒ®{ÎQÇ”ç*t†¥LÎ1¶{:WÊÕv‚Žúitˆµ=s¥n«'è ëò˹2Ôž¶) Szœ+/ãv®•e9Ó¹rÈŽfL “¶Ë¹jÖc\)=ú{Žº¿@ÇXǸR6tü0‹ö.AÇõšãª=«Ç¸B¾o㊶µ'èjçBk‚®² çE:„ºr6H9ºç ++s]¤Cìi¾”BÕ±EëWè6«] CÌw[ :è$2«z<}h‚±9úq®Ð»Ï– «üÄq;WÄÀ¾ã\¡{ñö‚Ž²tŒùéç$ëq®ÔëÆBè_q¥ÞFaŽzZ»0ÇØ°c\?ss”eÜÆUoîûßt×ÍŠ7àW¹Ë¶JÿJVÞ„,pB¼5¶‰MÂÄ “ż}ΩR—$†`LiºûömI_×qèô 8 +Öí ®:–WpeÓ +®¼fgpåsÌcWOcJ¥ã +)œ‡M:.$mCMº¨;tºìâ™[¹4sLë–Ý9®æ‘ÏÜÊ1lwNaé+·jm9ÖÓ]²˜ÓÒž‡2§íœ^%¬ØʺŽ[µ/c«NÕ8˜ã˜6²Æ˘¢3‡Ú@sä8’l9ÔµµæȱŽe7#hÄšÇ:Úm¨q¬³Õnœ=}Ç#’ubJœÖ¡ïÄéE{rã„Û5D7Žu·MãÆq¬Uqãx0ƒc]¬¹sãøõŠõQjœ>0ë³Ô8ÔCƒÂfœ€ÍwÓ8Ô©l™U¸[ÜÃPk\fÊé—'¸¡ä cWã¥Æ¡.ú–˜Æ¡n"ù0c#%qãP÷ԇǿOóÜ8ᬭ̊2öÙ»‘8á²(ý cØ Å‰CïV/âäJqÖ½«qr%˜iê-[:r+öšSä„›Î:~EŽµ„z DZ`¯BEn»±`™k<ÓÎÑ‘Ócl+rvN;Ó±Ô9Ö5%GŽ_¯öz ‡±Ô{sätBctä¸DFr2ga"'—·S9]xQå¸8»TWu”´˜CRjs\ä%VgNa1'”9­Í}gN7Š%YuΤiîœBêîœ^&ˆ;Ç: qç´¶KÎñÁ—v¶sk\âÓ¹În&¬v®³QìgnÅ|¶&Å¥#OÍrªJ‡ù¯ÙÚ»‹: A¸•[±lôÎ/êP[(ݨ㾬[q„µ“:ÖÁŒº¨ãEûL²Ü/Ü´Á4TêPÇROêtýŽèÔ5¾ÚÚÊ­¨•³:nþТSÇg&5:uøŽ±[çâÔáñI/+·ê̤êÔq´ ;u\g™‹É¨dlåÖÁæ8»¹ÁÎË:>¥nà)–uáU(’Nê^ÿqåVÔn²R‰õ¡‹º€/‘VnE]c-˺$Ê™[16rY¹U9-·NH=­cji+·j-+·²®ö*[Ö ;Ø•[…Ë|4ëP—PNë0–ÃÊ­Û5Ô:aô8©ã'gYÔñÕÎÙgnÕ±´r+?4¤•[ù÷ÐÏÜÊçûÊ­œO.°KºÁ5ÙNé°JöÜJ^›4‡ŽËNÏn‰¥;t/÷¸r+w±F± :¼ÜF´s:~Q˜àбîuÐq›èêœÐ)5eAG‰Â –: å±b+Ë>’;×õmŸ_8‡å9ŠH;  رïè{šlŸ§Ô6‹1ÖaPÐ×dÇÛ¡è‡]Úawmwyq‡=UÒ0´Õ; ¼qºƒ‡¬ÃÑñ¸=U®i’’Rtôx™1Nõ8Öì=û—íwvö8 Þíî¾ +bKЯmM®›âC>ZôkLúøÃhÍ탵:üz‡6dõyX4±Éêó°G+_y[Ÿ‡å›ªgYY²Y£‡½¨“°7zÚ“maO¯ö•e_qFNü"YL+Êé°ì‹Ê›œQ3‘âJ²xו¾DÕÆI_â^«ÍéC +Ñ'<é³Pr&Y™ZªÓw›)£~ÕM¾GzX]—‚2åK:ùcó ©òm× |‰yÄŽÓ—¸†¶.Ç4ÛjŸžcº}‹¶óÔ>ÔYõ˜öáëäÖÊa¤>`²˜ÍÙÁY—‡'’$Ÿ]TJ3Þ²ËS…ãêòp‡Ë–fÕ§4ñC=biÝõÁ‚Ç%XåÉ4ŸþQœ™<Õ?Ö=X¶sƒÍZt• óLd=Z=z=œÄ;êŽ zì«Û ìòˆ/|ýæñéí×O_ÿyøðø|ûCwoîo¯ß==~}øóv÷îˇoŸüûóç_?<}ùéíýí;òóÿòÛó·ÏvÐøÁÿßyƒ7ܸþ¼Ö¿|Â/õvw{ÿ~ûW‡™ä¿¿ü¼pW1gÿ 9©È endstream endobj 11 0 obj <>stream +H‰t—ͪ­· †¯ ÷°&…ʉõçŸvÔ$J'…t^Ihm™äîûJú,/:Øl^-[þ,K-Y/¥MùõŸßýþŸ}}PëöÒöaõ5^®µKh²$ärõ×·9EzS· #"­Ís…fMÉR3h°õN1„t.תCB÷ÔÒI^_ÅœMâ›È»&ëÐîѨ%Œô% û cH$të¹DÃÐÓZ}V[øb؆`®áîÒflFŸ.u6Û3xHxáÉœÑÒÓúØôü°¤×œÙÒ6æŒÍ¯Ñ»ké*¯Ñ>07Ý,w›¬anãŽ@{Ä!•0rt‰)X÷”¥’ca !¦-–QÁv E4\Z£œBÌz¸1|¢Qô”±\wæp±ŒxÏ<"fM}ÌjÈÓæ0ì3#Òž)Œ â Ü4pºÐÒLj ví®µecùgŽr‹ã4•p«¤q6Ø”†ŽÝBOî5gRΘÆ>ÆLc©d"-]jÏZëÖãgmƒ÷6™me´À +Oyª‚BÓ<ˆóö©gfŒŽuÔ$#àýÕ3G$>·5Éu„SÓJ8±HF-fü_üù—O_ÿøí§ÿûÓÇ_~}ý¦Ï¾üüõÅ7Ÿ~ùñ§½>ûæ‡?ÿå¿¿ÿ黿üôÃ_¿þüõò·ÿ;ä¿þü}úþü»Î1åÂ(LËÖÁ(yöÂ(¤±Þ%Ï…Q…ÔF£89ôæ(lž¥ÅQÊœr8Šß[êÃQ¯û¥¦YY,u)|³Ômm–B[‚1YJ‰‹¥”˜)–zþ˜LñeÔé†)vü :aê1S>0¥ÄÎSÆÍLQ8¸ +¦Ýk~õ ¦°¡€fÁºHé4í‚«í¢)lø¦Q4…¶B)D2ßQ +Ó\f…R× '°Q +½´J»bŸ] ¥Ð¬«H +‰¯æw’„Z\ERèÑH]¢ð.ºÍÖ,†Î,@ê:6úRØ&“H]/-ŽºraÔM]¤0º=lŒºn³_•{/ŒújV9‹Þ1¦Î…ÑXÆFaÔ]>¼Øõ(QQ¡K{Qԓú\õ$ɼÀhß@ßõ¤ÃUzaÔS‹¢PÃæ¡(VÃùߥ@CAÔ³¿É¨k’¢”‡Uun¢N!¹ê¬b; uË&C¡ûãò0Ô¯'äÞáÜ$ë œk6Û„ƒl‹å"œÛdõM8C œg¢+|ñ 6œcáÍ%O*¼¹ÖÔ…7óZÌ :ðæcd¾…– oaj½ð VxsílyÇ›Ûùoî£o¼¹´~ÑÍ÷gBE·ˆ—Ü yÑÅ6˜|wmÝlÚÌ«pé…63€ Bt$ÏF´*ë…6Øö«.Ð=eÍ¢›u„JÆ;Ý`b1-ºAo +GÑ@ËóÄ-ºÁ¦ÍM7h›óà z´ë nQt³äë«ðæzêE77­Õ‹n¡eÝ ‰óæ*º¹Í³kÓ ˆUxsM¼.¾¹­e|Û>6ß\ëÔ‹o±vvVÁ7×–mSð-ç¬w¾…©Þb•,±À›ïÎcöÎ7­¯|{ßâ4“MÁ7ÏÕùâ2 M+¾Y¶ß"óT.¾ErÊÚ€ƒ4²^€ƒ±uÎSÜtá¢áB'š‹paã^„ Ý p!žá‚U,E¸Ð$E8×í!à!œ"=ñˆ}'œ›âuüÎuÏN6@ )Ô/ÂÁ&#Ï0fBÒ’b4÷¼âŠqn£|úä ©+ȹž© rê%™×g@ÎnjɹÐù¼,Ê…ßµŠr®…fQÎ5½(ç6Ê + ʹVµM9—+ßt…9ßÎÏÀ\Mç »dëZ ƒm¢œŠtÐ Ü(Ò©Wäœé`Sê:È‚˜ƒ·å¸@Û~sèt"J‹ tÐbòÎ9Xlè(ÎAÇF7ç ûýâlC“ýÁ9èÅ“‹s +»fœ7è`btŒ:h™ túô5ï¤s[B*@§Ok´A§ÏÐwÐÁf­YÎõ.й6æ tnS¶Ýö±AçëÌ1/ÐÅÚ6 t1&›]è§KYg7Ó´HËè*Òùïq¬o¤ó@J>‚tqœ2‹tž!bv‘Î3ÉKA:ÝÌ}H™'ã"]d§ð&Ëôš0@>ã}6[ʺ`˲]ÔDÓ ta£U  ~“p#æ‹tá—F‘.tE:×úø<¤ó*[CßIçùLzºU×HÒM:ŸoÄ7Ò¹æܤƒì–Ï® ëÈÃ7ÒyÙóZïY5øÚÓ­ºæqw«^ò6O·êc(Ÿ|AºÐ|·«á7[œª÷A£HçºgÃR¤s›ã“ÎõTݤs—’Ï»"oQø´«49íªühw»Šß¨§m?)qMè|5<à/ÐùYv´AºØŸnÕ#ÁIÂCºVï¯$]C”ÖW¯‹t~ÒÚéZ!9I‡ßço@‡üñš-Ð!»û3¦·çšW¿ê%4ûéW½ì¸Ó7Â:oPæiXCóiX£”ó;¤ƒMÛiX ”tLÒá÷Åt“¶É§eÝ>ŠtÐÀÙM:_;{ $É'IG§aYµ´¨§eeô´¬¾=»9‡8Žv:Ö8ÌdTpÎ792þÅ9Öþè–Iw³8y'wǹ)´9çréiYýÌÈî–5R[©@hY§g’hwÏ6šºÐyFº`Ñ͹pK§g ÝNÏÕD¿íYÅöqq¦`øæù{¦0EmÞ˜ÃdË ÌaäÊ +JÌAk^us˜¾ò•ž˜ƒÖ>æ {êƒ9φ¼Fsþ CæžOº0ç~ÌAg“”ƒl½ß”Ãg®¬¿¤œWy¦GPÎs0Ÿw‡rØáϤœ‡,;¨¤ŽÞ{…‹räén‡rȲ¡r0çåx?焽ÊøPŽÁœrÐÚ3Êr°í×WPzZfT`N•œ`Þ˜ƒ‰-3(0Mö\x^;Ðò`¯8›òSÎ9èÞ¾Ç'‰C¬¿s¦ÅO¯ëœE°•Á9yZwι&çäic6ç\kžoqÎmcâœx‡ÅZœsçüÆ9·1âÜö±9ç:³³0KŸÆ5Fd“”ËöN¹0½u®®§RQÎ2—^œó8¶|%çâ8å`Ž7æð›$“s›·sžx¼nÌoµ0‡’oµÄòÙÔnÌgCô`ÎÉ2ÇÁœëÕoÌQWaŽ²u*Ì9ŠHoÎùoys&çþÇzôúuQü«¼e"!j{ì±MW-Ý  D·QѨª(,òí9gÆ_? V¨ôÎü¯}¯í™Ÿç@—9ç¨×œ‡sùÙUŸœchÖã\©‡Ì :Jë CLfçJ)Þwè Ëh·seLVÓNÐAÔçÊß“Þε°&ûq®3½ç3Йηsµyq¥liè¨Ñ*_ cŒìß £VÑ :~Yöî.@Çç|Œ«íY9ÆZ±©èöÖ]©<Ì +¬Ç~“±*Þ1é u9Y#4îÌq‘±Õ}èJÃ&c\¡¥—üBmÊÐAo$[õ@ëjDtˆu£Óô\}¹®(vB/çŠÐwœ+tMMtÔÕ71@ÇX;ýœÉrœ+õ¸1‡P³ü_˜£žŽFõaö9Æ4ãºçؘ£Fa^œ³W» 2ÐÙ3îq t¦ûe\-Tq¥.rŒ+§,ó6®ÜÇ:qµãtJé˜!Í»á 3©×c\‹18@g‰WnßÊä¬E6è(Ýõç˜Î³Þ¾•±,œ3²Œã[M;¡ƒsV&s&k ̉^µs6M:¶•š”Ú˜3^ÛÖÜÐÞ´ùÄCÖÅ.ÌQ)s´€#âHÈAjï= G]ÚÅ8„Їõ`ui3G]]ã2˱âøˆ¨âL{›ˆ³i‡ã2Ë5•`õð¢ Æ1Ö½öŒqœ#¹1"ã(Ûnýã¸À…Ncœm˜{'cô¤Mx2.+úƒÑƒqÐÒž5³{¹‡Xïµã 7¿ˆ¸ÜAì”/Ä!¶;/KèÆ‹b3™^ŸŒChŠä`ô1ƒqü}1/—:¯ãY!ËP ÄAƒ_ã‰8„4µˆƒÆÒ4——ùx"Ž±Ùk0./³=ŠûÓ€cͯ9ƒ\æd ÈQç¤äKiäörÔUË9{w-9{ÆŽAÎÇô'ä,$# G­"9.O‡^CL†÷H9;ÏRrL‘)ó‚bšI¹¼y»(g©WòE9¦çȺ)Y²ÌA'‘~aŽYÞŠæŒ,}æLqaÎ*ŬqÎQÓƒs†¢¤çlž”ƒsÔiæàœé5çá´µ~µsu‘œlfR´sƒ]â¸}ë@²v¿þH:ÈÜ—O%éPØz¡1tùÇ·¢š´õ¨ƒÞž4PÇ¢Ôc[ñH_m¿¡Ž:i¾PÇi‡çµ¡ŽEkýÎBtiz£®úK ®³$zøVHØ€Û·²öyàuÜ3Ô@ é‰ÊÏêÀ‘<Úñ­¨|Fu|´;·u“®¤h ncÇ·NÞ˜óîæ&[¯Õñu“UToÍØ­–åB‹µß +½¡ì¨Ã¬Fô Žç&Ç·BkÑvX‡³¯íò­ÍÚŽoÍÈÁQÛa½YÇ +èÇ·šÎÇ·R«_e‡u™ ìñ­™Ëù謃n©Ý¬C¬¦ã[÷Á:V«Þ¨ã«Ý9êøÈiç|Äå[ó*ù ^šäøVþžÆí[¹eßÊã,ÞÞé!"ý&©ßJ¾öÜtLxäÙq²îIn'K£NPCßò5›|´>¶Òù]ÐH§Ë£ã"P6ùðû˜®ƒ|Œ¥eH¾=Ç&ŸÐ’¸ç ô!–äÑåñ™î%aì³1ÎÂÍ>†Š×ž±º=û„UÑÛÅ>aB=º<æêà¼ËÃŽˆ—þéò@%YöÖº<£p9]V™K»à‡OÐÅ]Âr–ÖGÐuËðÞ*ð‡uˆÓ|ñk]ÆÓøG=Rí“ŸZ ƒgF@êÙõîõ0ŠŸ4‚ Œ2N·—Ø ÔY^að«o>}þîß?üëã»O_^~‡Ð›oÞ¾|õýçO>þýåÍ÷?¿ûõý·ÿxÿñ§?¿ûüó¾{ûò>òÇÿúÈ_¾üúÞúÿðÿ_ùy/à›ýûá‹IüñþЗ7o_~øþú·…ÿ´~ä¿Ä¸´Ç)’Ÿy.ØO”Ï?W¨±«@,ƒ-Ô•õ‹‚ÒÒMgš4Áu=ý`¥ CúLÜÙÔÖâQÃÄãÏ,Å5$Ï䣮tK¨tìªi]'ý{B¥-ÛGyÿXp*ei¶F MCQï7ðYÆҴʸ¸©­º¨“¿}G,¨ÏŒ%²“Ï°DöKñ):Ù†Lg*¬!(Äɘ–9}¿ª=SqÁAâ.2 Å[À=‹e^ÅJ6д&/˜Î‹Š7‘·¿¨sÄbXŸ±ç ‰m„ÌmHŸmî!~#¡ºp¼/Ý7 ;eh˜›†q5æ¨Ç¦6ÏdCÞ³Ó‘ŸÊNºs•¾#fQÖ˜nKfËÂG&êÉd´{Y†­VÈh‚3`‹e›ÄOÃi·1v5@×ä›8ÒS —ŒØÌ6oeµòpØ{QÛrñ>\ä1fd?c–:tkàƒ9梦›dŽàâÝcZ#£Ú˜=‡ÍO­ž{ÖLì÷°Ã`~ü²ä•b¯±;iQõÀþšUgðK¶%\&ëp®+[ÉD*ï;;ÐÆî¾âëÜÎÆÉ~µ³Z9Px(Œ±:{n"fh^“-Þ‚jl'=¹I‚Nö XjÕ=Èq¦tSû½·:÷˜Ö›Çx°VÉÈbÌ7M79ðö{ 4¶!ÖúˆÄ7@‹,€WikŒ׊[‚ºˆ8ÐOÿ Êßþ_¡¬(‚>äFiöb ”fözP +ÝJ½QšyY÷ƒRèšòA) +tôz³1ë#6K3ï*9,ÅïiéÍÒ¼üåj¶¶!xJ)åæ)c©ž²Y$ +6O³÷„O³7ÁSâ„»Š/Ëšo ’"¨Ü5s5{Wx¯¬9 â37~UYöS/ *k“鶀 +½iiDUl8zÍ‹¨Ê¦·‰ +ݧÚe\8U–PkSê”%p +=^§ŠÒT­§¥Î )$¾º\4E õ8ƒ¦ÐÖ¶.˜RÒr=aÊX›#`jÚo1ƒ)µ­ôSÄàU$`Jí)d,¥tJR¿r ¥¬ò©¹¨'Jíͪ¥|$§(µ!~CJ-ÆKk¡Ô^ã02”rÊMRnccÆ8I•­=ï’ER&GS¹PÊ$êÚ¥º¡¾Qʤ«íF)³’Bö6Ié4ß$E,·Rf’Rê,7Hó:¬Òì  %†äæ(iUÚá(µ¿Õ9 +­>ãÅцµçQ.Ê1&¾ÛF9êÒZP:Í"å“©A¹†=;í"Uéåbbü8Jc#Žº.½×X~OâøˆŒÃ8Ór!ÎBIqИ¡⨇¿%ÇXw¸â8Gj%GÝô"××$álÇêT™ùâB¶¸…7HmmÞþCv½%º ƒ@Ý’@èµÿudnÿ:4×Ndéâ±ÂÛø:s%oÈ>"}¼!›©ÞP‹æîò†¼ûÙ)ܘh?ú*¡¦}X +‡œóä ÷Ûë¦p¨Y›ý“ë#yµÚ/¢„WÏJáÆŠ“þǼ­ÇÒ93…óÜw +‡,¯¯Žµ)’Â!£›8I³è)Ʊֺ¤q?×pã˜m[1Îï=Çgãh–ÆÅŸœbœ×fçw¹‡ŒÄñ×­˜¸Ò8,Ú<ìÁÂ8šôéÇr¨æqØIX«‘Æ ‹ýŒóg½ç›3†HGyHt9ŽrïÑ%rÜãü÷§œŸ€èr\9Ï"E9¯éü”ó89WHwQε +y\9ÏÒS9ævüUΰ?ÇE9Ön£ìÊ1Ϙi]9æíz*‡Z_û¤rÌ»sÈ:ç,α&18tÈ2×Hè˜÷Í:㱌·¨CǬhÄ:ÏjE:¿l4|.sç1½Ò1ë²"k²VJÇl6R:æÍ]RÇŸã‚^ê|ÙzZ‡8{Ìt‰j[[OíÏáÊ_팇rï¢jÜvˆdŽ2^š«`‡Z´^;ÛX¥£‰rçù±¥±l¥uȾ÷>ëçµ/­CmYøOëÞöÖÙÁy^V°CMmIb‡Ü÷ÔÄξ!çG;Ö¢}sì웓.vÌ3fØĵѢ³ò“üº&vÌCµ`ǚŠѱû¹†cÇûìµ v~ï±?ìü#g%vžãý™ØymXjç· .\;þ?NnÑŽ+ÉŸvµóÇIî¯vÜ!}Œ¢wÒ'µ³ÏÝ«ï¼¾Šv¾;»¦vÌá°cÇ ½F/Ø¡†c§‰órFbçg¢YÁÎkr>ì<Æe}ǹFªE;¿¬¬ÔÎs[©³Å%‹v;ôÄD™Ú¡¶ÅÞäÊÜl¥vü›Û/¦v¬ÉÞ©òGS;æ6ëäŠÚ·s§vÌm½É•YW™\;Ïå~“+?"Ñ“¹vžµŽ®~Ù³S;d°•Ú1ϸMjÇÚXoteÞf©¯Ù£ÏKíø»¾ÑÕ—­çèŠxV«£k?€8jÑE¾-»ÎC¹ëäŠÚ1ݺvÈgê›\­ñå9«ví¶aŸv «tv¸ÔÑSµƒrÞr|Úµç²k‡úþƒàì]uÆG;áΫ³+jcÏ7»"/ò°ã$Kø°ã´²ßðêYßðjül¼Ìžv¨Y{ëqÌêò´ãµTªv¨m}ãëÏ5B;>úUÇW¿÷X©?ræÓNrnyÚ±6Þøê·±7¾òçj×·åô곯´Ždñ1üXÇtÆ›^»³»Ó:ßw½N¯¾7»¤uÌÇÞøŠÜdÔñ•{¼›$v®Ëyó«‰VçW¯Éþ°óë¯WbT­ó«Ê›_=·7¿2où~õS6WµµÛÓ†u¶æQ‡(ÑY?ê°ÏÇ–GŽÂÙãQ‡lcWêp|N¼)‚º; ©Cž7ÔñTÆÛ4¨ÃGF4eA³ÌJ/{äQ‡ÓHH‡Øæ¬Òщyžt<è}=éWôyO:üBô“Ž‹¦3¥“{÷_éð_çŒ'þtYÔñDÖ¶Žßè }Òi*æÒq!f;E:.–7aW:ž4_\ê¸þ]µPÇg}WÇ©ãsg%uÜ—¾´Ž»SmÖq·ÛÈÑ:ßÿsëx¶4æ^·Žgo—n]ÿæ‘ëX“hݺþÍ4×:f‹œÖ±¶¢ÅŠ™ˆÓ–ZZÇÜÅŠu¬i¼ݺŸk¸u̱;“:¿u±þXo—.þ`é¼ö3Å:P…Kçž+Öq!ÛŸuþ8û£?IlUê¸o‡<ê>s?ê¸ñôTêðjj:ä=[P‡Ç8lTê°V·G7ö^:æ3+ur×¥NîÕß$á_ù×:^¦g²¹Ï:æ¸d±NqÍÓeZÇÚ±7Å2ï~;æÛ3&vʇ±ÞËÜ·%vÈøõuŠe­ßžØ!cB}S,ÿ¿Í2Å*ÏåzS,ÿäDOæØy–:ÅúeßË8ÚIì˜-î’ØéGôÅŽyö™Øñ›‰YÁŽ¿Xä ±¾jšC,â´S‡XÔ°®–Ø)öž˜&vÊ#¹ªv¨YA×µCžwªuíñêÜE;Ô¼ »Øéˆ_úa‡Ü—JÁµqúNìÓe!äy;Òĵå7°C<¶zb§Í:Å¢ÿÞ‹lmôÄŽÙb;ÖÆëë<ê›b™w¥%ô±oˆe>Ú“:æußu¬MyCìÏ5œ:æ6ëë·¶óaçÙ;±ó¼êë5{C,³ö7Äò’zêË…´“C¬?N=©wÈ°Y´ãNZö†Xu‡;ßxZgXnNÓžØ1ÇèÖq?«3,kÒ{Zç¸ì7ÃzŽ™6­óc’ÔyŠåpꜢ?m_¥½–yž7Âznÿ°2ðÍÇ)Ô±vZ§Ž™;à£9PKèX¹¯r‡y®Û fÅ9”ЭtŽYÇIç˜íæëœðH®dŽŸèѹ)žÛ.ÌùUwOç„'¶i:ǼÇ.ᶦ¤s¼F»3cÑä¥sü#^\îœ/™ìÏ9ijcìJçd¢MØ+Cîãg~žÇ¥Å9ÔÖ2Kç¯aÎœ,¨Ý¤0‡Z4`×9äá/‹ë2ngÅ9ÔNï’Î!GCxãÿ_÷Ò9ÙèÀr~EÒ}{82‡ ÂvaµÙÆHæñÛf2'ß$òÃk'zEwN¾iæ:‡¼5æÌ„Žµï:‡N8gÉL蘥Ík-Þ‡ÝÏ5:f›Z ó{›~ÐùGÂ\‡.þdè¼ÖwBÇ<à +‡Ž?oFKœÐ¡Ö÷^tþ<Ù”_è¸EN?:Ô¦XOèä3÷Jç[O¥HÇí¹e¦tÈ*ýQ‡Üz_…:nó¡3©s\‚ +§ÎóÞ…:?)1ÕÒº°f¥unQ›Å:¿L“´Ž¹Ië<Ç%‹uØ¡c¬ÚÖ¡=÷µn³§i¯­Ûìwa7\X2R;dY1³ºv8Ó¢ÍKîPCÇÿfXœ¨9–&wÈ߀úqǃ9ß‹OD#q¹cnÁTrÇ«î;ÖòÈðܶ ѹCƶªÜ-hr·ø†[o†EÆLPgXžÿ ´sÇU“©wØ¡º£Iî6:…=Þ ‹ÃcfrÇC¹Â®äîpBÑ™ÜRöfØÃç©]ÝaŸ“u°JWÌà®aËK¯Ü5ôúfXä”Ù¹khÛo—šÜ5B3,âÔ9žw¸:âvl¼V„x<ï|™Õ;®Öz3¬gy3,óŒ÷ÙóNøpß +¶d<ïGÕ;Ô¬½öçá—bVîxk“䎟xm]üAa½Öß Ë›¶þfXþÛu†åBú„Úñqj[©vªÚíXš;»d%vÜx>úü`‡Í‰Ñe'v›ox}3ìæCŽ¡,±ÃîhüCê;æ=OÁŽç¤ªÄα;bÔîŒùa‡š7Â2îÓÓºío|ûÏ:lÏ3DVÁ®á ê<¯³kìl,îèÜ5vÚ«w_v4KðpNïdQÀ÷Ž¯™ä57'ìvóPhxïìD¯ñ¹s?ý¨Çó‰CÓ=|H‚&|¼Ì9äcqÅ_¹|8ÂìÄ-éca´~Š}oéÃñC}ÛO³‡ëv=Rô#ÍÇåËØN˜ùõ²ÙÙ¬6‚ð­|K¢Á¿m;YØDY$‰Â +(É& î>UÕ>í× dÁh¦ú=ö9¶»W“€eë €Ó‡Ïä~[F>~Uj¼ù^ýš–j§¯Í‡nnøž:‡WÃ'oöÒØbÿlž¾öÏä`qE[KT§Ã?ü æ»­ÅlµDW‹¯Ïø«,Öuã¯r+lþóH´ÓÕzƒrwµljÌ1*ü=MΦû -õ…~¬¶™ŽÛc÷%¨lúá÷¹\ýÛפè÷2‡èWÙšxêþKõ¸=>2<ÝÜOpˆó0øÇXñêÿ Û&ˆø‡å´1úÅ?DIÝ…ŠÔ3y—L~jå¡  äH©×°Ûóa¿hPE>$Ltm•_¡ð³Ï?|üòÇo?þø¯÷ß|øùí·}òù§oŸ}õñÃïÿööÉW?|óÓ÷_üãû÷ßýé›?üþËOß~ÃGþð_ùóÏ?}ïýðÿ_øo ‡þ|ý³$þñþaoŸ|úöõ_ñ¯+üÇý#ÿû{ŒKÏ8ƒ©A¦•Ž4B•þs‡:Ò1ã…KÝð5 †¦€º· Ī!ÛkôÔÌpêY5†™G™ç3"#‚Ûë|‚7 uæ)Q›k4m¾»ÈáÒÞÑ‚ñWšHÈI\Jã–I}ÞÀ|D 5­oÓªtJþ†45eo-ÖÖgÆOŸÏ€Ê]Ž$$j\g±T"¦az›ùvq­ÈQÃ… –J+G÷˜™<¦+Zc¡ÑµÂó¦‚Í*ÛÌãX(Ôîᣠµ×¸ ü$@^¤é^÷ÝFŒ±N‡|›^ JAƒ]zõý– bš§ÐÂB7ËZ˜ÌÉàíS4‡µ]ûˆ ^*\,Ž“Ï,“ž¥y e™¾øeû=€ +‹W A£ 4†w +dKÍ÷¸Ÿ!­$¥N›ÎEi668©Çˆ!3ëtQ§ô2ð´+¾˜šý$³ÁÖzÆôR5F”{™cø‡Ùιn-ÞÒ†rJ ¢¸”µ8á=f(«yq>cÌ<óW5‹×:4-®òòaM¶}cp鮊¡\“ö9s¡ Ïhþ³Žj§@Ñ]i„"ŸÈ•[™¼R™Zd´lšFC©«iŒª©ÜM¿gÚ‚=¦Ã0&C¤ôOªÛÅËE:K’‘þ~\X!] F;2™M"PõØ6MœrèÝ”t‘†ÿ[Ò¥V’÷Áø‹ÿ+Œ %or#4»? +„B#gB¡{i7BQ^ËSÅ + «}Šòœ£] ÍžËÁÐÌKª†â÷äú04{sù€4óv<¥¬åæ(ciŽÒ%:£Ð ØÍQÄ*küá(YÒë)¾,[¾AJcÆ#|@ÊMk倔.,—¤…+ͤ¨ÜW€ÔXõËÝÍA +”$I—~‘Ôè#Ë’B÷À(„:/Œ-MïQê”k`zm¬F VɬF¡K[AQH|u¹(ŠJqE'¥bŸõ +QÆúšQi§ J­•ˆÓ"×€(µçŽòŽÒáeÈü’B_fú¬I¿þŠP½Ö,ÊgrêPYùB¨bV¡zKPÓÛ/‚r{ÎAP£™ç² ÊÔàN½"”Ù3œwD¨=0Êdã%úŠP&d;…È´ (»%Ë7AËý”iŸê(u®7@ó>ª y7[@ÉŸzñ“”*ýð“:÷ÃÏì3ý’Ÿ ”g¹èÆXõÝݨKïA7è´J½èÆX]tëÈÓc©Ê(¯lCˆ¦úA%Ý÷ƒ6êæ:ÐÖY~Am|¤ÎÃ6éz¡M¡d6hLÐmÔsæ mŒ‰÷mœ#õh£îv‘Ëë5Ù´a-ÀÉï{åBZÝÆZç÷õXë¬C¢WvÇ&(OkenCH¬u/°†Øcæ„5èY× ²udc®ã"[gR÷dƒ~à«’®ÛÕÙú®•‡lÐhSÚ Gºý!B¸nF­o_7Ú(g»ÈÆÐZd“ön@dƒÎÅo¬‡l YÎA6h؇h£Îe]lc,Õl{™cè»ÐÀ´‹mz±õ`u÷6Iló1ëb›bhÓK¼¸„6®mxglÃŽÙ¢çÚlÓa:˜Ä6æÇ"._؆‚ÉìÛº(kÁ6e\«Û””uÜ {îpƒ®µ¯ nÌíÞVÐM©ï®Ft“Îù¢›bÅ‚nÒp~Ê|¥›(åÄݤ™¢urú]tkXmý¢c2Æ›nÔ潫èF]ݞݫÃÏQt£Î`ñ º˜Ù+ßÊîú8èlþ)õt€k,J¿> 8>2Üx pÒ¥]„Ó´nðD8êšgŽºŒvŽ±¹Œjî§E….¶Ÿä26î1,ñô¨Ð£Xȱçpȱ/™§I•.§I…žÉﯠB-&µ±Ÿr0:åðû*ù¦b³œ6õeŽ¡C Þmª^ÜÇ¡Ÿñ†Æ)—Oƒ”c¬Ÿ6Uoi§MåâúÍ8ìâH§KÕY: Ä8¦Çèíb3hõèR«p;ƒqÊ·zw©ÊÉšƒqÔ«6:å~·©ÌíÚr@NTY§OU-¤»OU,Ï€œ´oˆ ' +]ŒÓ¬ùô©Òéô©Ô3ÿºO­Ùnãfbò°ã2}ÌAdv'}—Y•ù .‘Aý ºõy!µ³Ü ;â ›‰#Ú\ı&ýâðHw戣Îv#ŽÓ®|í½‡2™Ý„KØ;[‡p¬ò:á ‡ûºC8,p¸rÂeM ª>á•p‰·ú!\F²¶zÇ‚ W šæÀ&\9ô"á qW®‹pˆ=¦K„ƒžº!6â*^ÄjxEb¥×ˆƒFj@tÝÈ Æ!ÖJ›Á8hKÛ¸‘qÐà—]ŒCl•Ýà’qµ½Kæœ$ãêÓy¼0Ž±ì¾]Œ«O÷²GÝülÆ14ÖÆU¶UnõÅ8jrì•qŒ¿Ÿ—9†˜ qÿa½|zþ«‰(þV~KHŒtú¿ºÝš¨Ñ-"DECpÁ»÷œ3½Ó[ˆ® ž3ßÛÞÞvæÓ9zï1«zÂMç#ÚE8Å^n•zV Âq7æª㸋iµ`œN³ÄeìF7âAÅy$Ä=¬}Ç„ËëFîž”ëAtñ͇DnµÝˆÃ6J=ˆ#Tæ8ˆ£væÄÙ>¬q¶ÍÒƒ8BÈêÅ8N“Öat^ë0ŽzOùfëk¸‹ Æ1¶êq«ªÁ²rÔ»G È!VÖ8n•ºÌcÉÌv¹U†Êî× 9Ub?n•¿§~»U–áVUœÞƒ rÒv»UM{Ì*eK+ '&ôvAN èǬR÷Òr\™y[㛳ú€ëœ˜°n³Š¿±¯õ<‹Ì¶šr|_A9.Gë”ãì}å¸ ,õ7å¸Oó@Ne6YåÖläxª«Ì€Ol“Xµ£¬Øýè9%j=n•É]G È1ù{¿Ý*‹¨÷ãVUxž9k»«ê{c¬>N2·J=/ıÞm³J½œŠB5ÊãBcÝŽY}Í1´¬–úmVõÞºrzÆ= '=n³ªX=f•:—cV9g^·Yå.Öu̪æsD‰rüÊVûE9¥l ³šÅ߀œ.ß^•IY}‚õ¨Á8a½½*cVJ0NP™Ç«J»w Æ)çâ$}?„81ènã4K:V•º¯cU¥ÓO­ªaJkëBcj`7â¨sÉ8h‡YŽ‘â‡(ÀA÷1FŽ:·7ߌGé¹øF}â›íÙß|³¦ÜÝxãÅ0áM:Í ošu–à›±\S¾QO/šàc£[ðs$7Dâuó¦.øÆïkÞC‰oÚ1ï±Ä7è%‹ðâ›!GÆÁ7èÒ^>ÕXŽòwÏãCÄq¾A>ì"Þl`Û’]xCìi¸Ä7è¦ëaó z˜Õ‹oˆ­R,ø=Ë\Á7þ¾y|³É5Þ óÜ=ñ tÍ oˆõä·šð·õoöxŽÞ[Þ²‹oöø–Í7è™ÝP>€c¨ùí¦ +3*ïò8jKýcÉo@î5ÇÐÂ`5ò8½¸æœžñâà|̸§X™8ê^JŽ×g¿‡X™sàtœ9à˜ «¬ pˆuspö°vN)—í"ÓrZÂAg+qЩ”q!ŽéÝrÄ *Ž!NÚ™ˆS‰¸{ã2#'¥þfœ¦IŒ£NË‚qÒ{Ê7ãŸ@ÈÝÆ!6J©Á¸É&&6n²Cœ·Wøºa-(mc{SRµÑënë6æ²úòª(§ÞFÌA»}aŽeÙêâ ï6樓ã)0ÇY§'¶0Ç¢MŽBa:·~cnˆû907x±ãU¡án¯ÊâOB’cŽ›f=æ¡(}»07Ñ Ìv¼**„é9Öä¹ôø¢a*9æv¼êb!¯»‹[ì¸v§Ç-æX=^Õ^dåÆ\ÂGæãU¡;æôÝæN½¯ +Ýso‡s µÚn¯ŠØÚ ¿sÎÀ„Ú‚sòýæ½Ê8^UÚŽW¥î~…çŒëñªFWåhtÎA·ÔnÎ!VÓñª¯9†/Ì[ìƒ9¾·ÚÁ9mœ¸½ªbåxU¾3•ãUù{š·Wå.æy¼*O3{Û(Ê!?@Ÿ›rH¡Vë­ÃF@Ž 'ó‚’eä&¯ô|¼*yÈ}½ ‡kme#È‘*cX@ŽzöuAŽ¢ÜÜeÚ)”ÜL>C¨®cU)ç*Á¸©{¾þ„qHÎÕÌƹ„ÚË}N.±•©Û½s‰]b.7ç4´35@‡jh}•cXÀ²}™êPR­,'¶X‡@Âu3v ×¼hÇê„£ÈïðµRrÓ¬õ#â18|ˆ‡úeã]y ´TÖżċ ÐC}Ú«¹Ã¼…/}S,HúŽ=îbZc÷@‚>V_øjg,;ýîÍݼ{‡7v^x»îaOJÿj/ª©Á®uïO½ØËÀbëú<þõç<Ž|ý@.ǾÎép?ƒvÛWôb%÷ŠK®¹Áö +‰¶nìavoy6ö°îI{nEn÷Jû"£»±÷Ø™M=}ê¡—<Óéîh³D“M=ü>—ë cûbõ^s ®«1…/ì!–Ê«»ã3Ã+AÜÓç`p±ìU'îAW‘csSåü^ÜÃ6öñêîx–»sóîÉ_݈TÜÒª»óéîp–Û>$e×&mðA¯ÜÆ ò¿–ç² }Èïâ$ßì#m¶Ûû¨gr;ðK~f9è'ú8ËD?jÔ ÄÍ 3ÏÓå%w]ù'üäÓï¾ÿõ7_|ÿÍ?¿ýü»>ü¡>ýøÃ'øþ»o¾ýë‡þðõçÿú곿õí—¿ÿüû¯óë?üŒüö¿>òÇþõ•?ôKü‡ÿÄ~ÀÂõß_~Ä_âþá£?üåÏøëß +ÿnÿÈþãÒ3®£•)2<ŽìHër)‰ §®¤ô$—FxQªE`ÿgFÍ´ã¯,­=Eì1^ÅÔØpj]3Ô}k´ùÚÛ.Yª/ÂMì& Šë¢éŠ.^ÑL‹J05îl9V€P:Mùœt, Œ7yVŸÁO´/ÙßʷѾàyF 5K¡ûÒnU=cd t7ÜL´&cÆ”ìÜÖB¯ôƒVÁ ^PI¾¢ÔMÄÊr SvZ;Ž!tN€¸†Ô•ò3DWbºÈ†:6ÍD ç¡)[uú!ÔZ×4=sZ”{7w4hx©×kmb 6d”:ÑQùÌJìK¡QfFþyÊS;’˜j¬°»â.ñL K¯Õ©kªÛ(ž15ó“›MóV«~8t;Ôú\œ]®=ÆL÷¤ªtèÖªÞ3'槦‰dŽàhŸ1-åZâ½s 6ÄCº{îeöLÏ{êPP¾ðKé›cŒÞÓ^Cz÷ +¨xT¯ÙeÆŽ†[²À=–áJó£’Dlà2Ô‰ªåGÚâjeTÿY§µ“@€e¬zÞXá.bŠæU´2êN'ùY5vš€¹Ò’ÊX?Wòoi£y wŽ—AWÄ|i“FåD±ù¡—Do±A3¥ÙóDÅw !)~õàªh¹‰m¢tvîM:—â<ËÿÉŸý_‘ÜÍä R„ä®BOš¯ RÈ–ë RÔ×ò<H!k²R”çõ&©msõÔxQ•CRüž¶’šÛÊÀ©ñŠ<4¥,ù¦)ciš²Ot4:M¡ùÿ‹¦ˆ¯r§)iÒJà+³n7NñÅÒŽSîYͧ€ ¯ò §™m½œ¢r¬­Àigѯ~á1¸“8…~X)žv¶»¹]<íl'±î‡§Ð-` +ÑG™o˜"4WkSêd%` +½6\¦½"{ ˜B纂¥Xu¾XŠªqK;íÉA)%Ö¥Œµ5¥Ò~‰ ¥ÔúÒJ;m•@)õªARJ`€”¡^J€4fØ ¥N³_ Õ›{òÃö> Õ˜eo*Ôs€T¯i#@ÚÕc_å66³àhg[Ï«ds”ÙACð)³h8õÒþ ý)³®¶¤ÌÌ…mŽÒ7u»9Š˜µƒQ¦*£ÔVnŒÚ>¬£Ô>­cÔÄ’‹¢„Un‡¢ÔþZ§¨y£ü#Š6ø$cöÆ1TœûbuníadZ¹\Œc¬¬þ0®¡ +N«H•G¾‡Øÿ!»Þ²Ü†a€nI/ê±ÿe*œ~õ€M”±-]“+€cdþÇÇjX7x$÷NÖ¡6ê ê?ÆD2^˜+Q‡Ú×w‰ºÁ]tZP‡Ü­ÿJ‡Š­±B:d}ò“y^ùB:ÔÖpý%òi·u£t'qø}êPkcÕ ¹ïÙ‚ºñM7?Ö±æL‰ºñ H—:f]ìu¨Y)Ô1¯Þ‚:f<ÈDk£YPk\êø;{íD~ÛvP§Ïøt#ê”}4±Ÿ¯la~fœ°Žÿc›¬ãìÞ$É:=Oj­ãéfÉ:n%s˜dÝøÔ½Öiëõ•¬Óöìí³ŽÑWuÜÐËz¢5;£uÂåXP§CQF¢Nµz‚:e_WÖI£Ö’uZ·®°N¹¬°Žyøš¿Öu\ìYã×:”vofe.c}Öñj¬c­îýY‡8Í[/YÇ\fžYQ3»-;­c.ëͬÌm噵óTî7³ò3ÕÛ>Y§ÜòЪu½»”uÈ8^+¬cž«'ëXÓ àZǼÇø¬ã’Ý[¼°Ž—ØÛZuÓúZ‘Ï*yhí {MÖ!ãMÔuÉgVÔ–ù\+ëñ +3ë(|qÎl]‰Ì­+¸KçQWÐ&´“¬+8ic>ëÊCYÖáÿ÷ê*ö½¡¹Mo•º +ÆVžZQ³=ßÔŠ¼Ú¬:Ž ~uSö[•Û[‘wñWÙ³µQÞØ:8`¹n]壯Ù:Ôv{ƒk¬ñYW¹V\õÛ>uºuüŒO9n]©Å~¾BÂ:þÌxƒ+/ϲt¸‘«¼¹UO“J]é¸C–$wÒ±7·v¡»C:m¼žçVmÎ^?éÏxƒkçÏY\¹Çû¨Al9orÕ™(yrU­î NÙZP'Œ²tZ¶¾ÉU¹¼É•y×ÿ&×^ñäæJÒ¡t»Y—®²§ è*ÿ¤¡Ã67„‚ál{Ð!Û=èpxŽ·êò˜ëA‡F¹sØž~«ÀèØs®b›þ ãyÌ-]oØñÖžs- “sÈxkžäj_&ç·ÞºÞÑ#´ö Jàƒ¹ÚYr¿ð…t¨6vH‡<Ëmâ(2›I:ÔpÏ{H×¾ëZJºþ "?Ò±V½Y”týf®tÌÃpHÇÚ:+¤ëœ³Úé˜{I:ÖZ]!]¬q¥cöíÐé§ßøªOø¤#çüöëœJ?ó+ó5œã Ùg$éx#‹wJ’Nϳ?èîÇX:ì¤î*9tŸ¸tÕKÐá-TÚè»÷k6´ ËÐáåÆóБ–½tÌgfèê}\tõÎOt´¨Ž,×)çI‡ÜÎyÒ1ûš¿Ò5ìÐ5ç¯t,ñæWæÝÏG£šÅêPëgÅüÊؽ÷uÈm[ž_Yë·s'uÈMßüÊÿ/3ϯ§r½ù•ß9Þ÷‰:åšçW­ûÆWF+'¨cÓu¬IÿKóìó£ŽYõ/¨ã×úÆWݳöÆWä9N_QÃmA]þ®£ureëPÝG\Y‡<ï<+ëñÚÜÉ:Ôn&êšá&í7¾"÷Õê/u(Ùé;¨köPæñAž· êP[:µ—:ä3¼Wuõ™çWÔ ß›_‘G±Ô1¿‹Aköz:ÅöæWæ¡CɈÀóqóª=AÇÚ¬o|5.tÌ<ñ¿Òé§}öuúŒ:¢Ny¥ñU¥ñÆWæÖßøÊ%ÛÉã+oä8o|Õó¤S×:n3YÇ­´Æ_›ê´óZž^¹;GëuŒ>ú麹ŸÏÈÓ+k|ŠŸt¢e¿éUÙétNtŠct¢èOK§eÊ^™çyërùoxåÞ­v~¡ÓÁóVVÐ1·Þ>è´`Ž•îOÌñx¬µ‚9KÊñP-ïÑ¥s³ÊéÐÝü)Ç³í« +9Û99eo59-»{(÷ÿO9u?5¡kËŸ”ãåNGÍÖ¼¿ åxO)§æ””£š~”ãÿ­½B9~×~&W®VKÊéO#”ãe¸`~Öq'F© 9Þ­Û}I92½*®r|µŽ_åø˜{¯¡œã>¡ÿÿªÊqc–7¹r+ïÛ¿ïØNÈñP³@Ž‡n8”B®~#Èr¬ï¥\ýƘ«Ïqó)5˜cÍö æ*'¬:ƒ9æZfbŽµRN0k\æäÅl‰9ýöhÁœ>ãcŽ˜óï¬_æTê;˜cž½s¼¼¹gbŽjlï“Äœ(;òË/ýô“˜ãí©®’˜«Ÿ¸×9í½V“s|,»ÎÏ9ÄVûƒŽO²÷• ã†°6:ѲN@§¼w‚NGÅçYIçÖ¬ÎÇLÒiRC:æâOIÒ)ûš¿ÒaE³•Z:þH÷«’t› M‰–n³UÜyzÅ'×ò7 ­ã—O«²_ôøƒŸæø™^ñÑi«vÈßdØñXÎ7¼â#ëöþÂŽ¹Ìš°ã²û´<0ü‹Ôó\ìq×3vKø·Ànñå¶bzÕu[ž^yúËjïY-°ãÃÞ³&ì°iê¶7½bŸ˜ØñH.—+°;MÚ ì!{ÓëáKóäŽî°ýò®OØÜ¥ë¥cWp·jOØtíM¯ÈÁ²°+hÙ/á]ÁžïozEžÍ{e×®à8[ž^Q;ÃÞôZ+7¡=í4†Ì¬G—õ¦Wåú¦Wæéo³§]eû¦×Ê1Ë…tíÍÝÚ¡6Ê›^cO»Ê$cÇŸöÙÓ±ãG^KçßHÓ«JýM¯üQ½ù?ëðÿeçé•7²í7½òy6þ¡×:l”­ÃV²ñ¦Wþ³ê +ê¸ó4÷üP‡Ý‰ÁeÔm¾àÛ›^±¡—&²ê ØiþQG[–¿\Dóž'QǃR\TQ'lìQÇÊ/?êPç ¯ŒÛ{\I·õÂ¥Ãþü>—ßÍÁ>ÆÒÕ5ðˆb‹}ƾÄ;Ï€Ÿñ-Nc†×„è§9Nö§-è]Å‹~ø|;ègÜà‡ÓãiÊÅÝN;b^üÛéKv5¹rzâpÙN™—K;ð‡ììy‰¿Äicÿá\¦û« òÜœçÉœ«ý©gªã@`òS+Á@AÈ&R¯ÑO¿—ÈŒÒf`Yæv|‰f€¯ö¿þæã§ïÞÿíÓû}xûñóËozõÍë—¯¿ÿôñý‡¿¿¼úþ§·¿¼ûöï>üø§·Ÿ~úýw¯_¾â?üÏ!þüË;ô[ü‡ÿÂ÷{áôߟ%ñÇø£¿¼zýòÃ_ñ׿þãõ#ÿù9æ¥{^‡¥1¹‡jØÈ^¡Öe\¬áˆ¨++˜Ø0¨³Éµà÷“AÌz’+ɆB¤V†ÓÙ ÍQæA²ô¯)y ÅNCxÏ𾵶ôXp„)ú;MÀß~[ð7:È… ·êcQBôgé®eÅ©xx¥1äÞD§F´~±›Ïˆõôy¹¤ìcð®÷aØÊ¡Ë×,–¾¦ uÇ´|³t¯b?èž³îs&è5c¦ìWoÚ+eª´1-ï©Dëì7Šü/ö~aãs ·›÷㜚–ËmÞ잢étíì'†Œ[—#§9ƒFÒÔ€¦>'Æô¢†b©=óãJ!¡lnüé¬÷œAðá‹á 4f¡œ8¦òÖåM*S; ‹@s€•"׃ÊBº@[œÓu‡@פ]œ:»kN-r0¸ºÉU‡£ê¦æ÷âG.{n,žç û¡[«Cgžùí…w–l~¯wwR˜TœSE¤½®sÝ=÷ä.î瀋ÌOj¼swÐíø>gÎëªU¨{úã€ü1^e¥wß’Å«—•9Ú=G.J¾]§;![o;Bã¤_qFýN4¶æîº(o²%·Î)]¹×ä¾…÷ÈO2 dYcUªíÙÌSºuý>V‹bk£yŒ~OeÄ•>òUYMh@Æ/lãÂŒ6V×hæp›5ÑÊÑ£š•öªÁÔÅ´‰µàûHþöWErÏ~_ ;ARèÙúifÅפ¨°µÆ)tMyƒ:½ýÛ$EÌFÞ$ͼªl“¿'雤ÙûËÀiæ¹iJ©NöASÆÒØ4¥UÄ M¡i{š"f³mš’&È×À)Þ+÷|âßKJß8å–Õ²q +¸°ØŸ8-¸7rÞ8EáàrœvÖ<óÀ)bxƒ8…¾Xé<í†ò(íài§¡„E¿y +ݦ}˜"6WkSê”-` +½®¦½Â[ô0….uK!ñÚå`)b¨Æ,…¦%ºQJI=QÊX[3P*­;ÌQJÍO}¢1t*(¥V +9I)À R†º.\é^ÁAJf?@ª'÷ 嘜Z€TstAHÂGÜ ÕcE)—tdG¹hUoŽvúzú׋£ÌŽÖí)³hô íѤÌ:ìÃRffÝ…mnŽ²sêùä(b¹mŒ2ý“mŒRg;1šý°nŒRúªŽQBÈNŠ’U¥mŠRûS¢ÐýZñIÑVÁœYÆ1f¾Ûbui-V±ƒqŒ‘m7ã +a[Eª2ÊA8ÄxÜ7à(éÄoÀQWé på¨;ÚÇ67á¤íœB©à ±B ÀQO=cŽ±!°:à¸Fj%GÝúÁ7~]³|óMÛUoeå'Ýá—Ýpƒì­Í€[c®zÀ­5”m7è2enе–zÀ 1»nÐP +¾µŽ´¶qð ±b­ß /;ß Í}îæb•àºù¦uz¤Ó+"„‹gßÚ@ô€£žõàCzóMÚfð :]^›oŒÁ©ß á$VŽšcŸ„c ­mn¯á„£Æ_áôìÞ‚pÔüÿM8Ÿ³ž„S¨àô/1Ž_7¼× +ÂaÓú¢ÿrÂé8N"Sd‘™Â!•à7[® µ=§Ô«vNÙéí£ݲ;!ÚÌP Ž9Þê +Æ©ÜáˆqÒ9ŒS¬ô›q’='•y0N¬rî¨\¥³㨓3ð`\í|B;ǘLòÅ8êîݬGmnÕƒqˆÙ˜+Gâ´ïI9Ʋ·Âtî£樧ô…¹ÊªÔ ê˜ã€!易.õàœ•ÙsÎQªôæuõàcxëàu­-8G½dì6èø1äç:í™mÒAwS7¨Ch¢œ‚uÐ,×`]eEÎy°±š{ ò˜£÷å8PW‡Û®u•Ï[%Pmø–'éj£Ž tèƒtÐÝÉ·I‡Ø¨‚¿“z·n"]|ê¨ê+èuÐ6{ ÔÕ«½y²Ž1·nB]½:¤uÔ]ÝëFb-ÉT9ꨇ•@u+å@cUס£n¯á¨ãsæ˜êôì6u³F NÚû”u +µ¬Óc*þŽº=XÇ´d7ëtžx£›uLkí`S©µ¬«toÖ)õl¬SzZ ÖQ;……:&4Lô:ÄP·%P'¸¬¨SQ¤z N±¼nÔIú²bXTÊÁ:-›G°N:`u½–|²Î¢Ë{É`b3×ݳR#OƒuœsyÅ`cyÎ`1_W ÖQ§~ö¬ˆµv¹v²Ž:ݳR—ñèYmú ݬã€,?欓.gÓªE× ÖA£¼F°Žºë!›uŒµ±›VêYk°Žkš<Þf?ÐJ4­Ú3ÛM«qn:šVÃò]!g4.Ê@±"çÙ³·B}­³WøîYkâ½ÙOÖ%·`Á:d‘£ÏQ—¸™ëdÒ›n#X—n*_¬Ãïó Ô¡JX´:TZ÷1Ž:Tâg׊X›}w­Ð£ô¼QÇD{ø@û”¹ÛVé²ÛVè™t•=X‡ +e³Ž –åÍ:ü¾J>Y‡J}³.Ö¸X—¹gãªg·±YÇ1«oÖåhY‚u µÝ¸ê1u7®ü¼v’9Rô­:M§”HÇ „çƒt̤Õvßjbî Ò)ñìì[•œ–ƒtÔ«îÆ:åv6®Ìq«9P'¶¬Ý¹ª&ÒÙ¹*–ç:Iß¡N(:I§Uóî\¥Óî\©gþïÎÕðµÖÇI:ÄägoÒešš :Èì®zƒ›ÔfޠíÙ6è k›'è°iËï 4 ³AÝ¥oб(u“^ ã®Ë] £Îý]yƒZ}ÈÅ9ÈÔûÉ9^_›s¬s›sÐCïÁ9|ùysŽ;Vúæò0©íÚœËo*Šss.cjµ :–ãi鬠7ies®Ü sÎAãÒ\ç“»9=y¦7èÌ`J9@‡Xi¶tй­ ƒ6ß&bµÔ¤ƒîÉMœH=XyOÒ!†=· Õ7ÿ¡»|zþº‰(üUÞe"¡·ÿ7¬J»A,@*ˆn#ÑUùöœ3cÏ\§Bm¤÷ÌÏö½×žy<'u¥¥‘®l'ò$c¢Í¢‘®l7sHG]õ€ƒtŒ í®Œt…>+W'5þ¸HÇXÖ ÑHké¨57túè°¯:BíªqÎf´'ç4ôð¯ÔÓP¡œã†ÌU/Òq#Ój‡tzž%@—±uÜ C&•&º \3/¯t¸„R®:èbýšÏkµÝ Ã3G©:¢eŽ.Љ×˜}rБDRoÒq™´‚t܈µ‚tÔ{É'é22t˜¯tÒ1¶jøWêY–£Žz÷‹Ž:ÄÊá_©Ë¬Ž:è<Ûí_+»y'ê áNÿò÷Ôþ5³,GøWNXÚêTËí_uÑ°¯”èñuÔUŸ¨c¬ô°¯Ô½tGßK´Å Ôñ{Eܾê–å°¯Ð{úDBØÔê¨ËHkàÑQ—Yãfbµ¨Å5ÖAwó³Æ:hÜšóbblÀêrÃÍ°¯Ðed¹P‡X[e:ê 7• uÐݺÑ@bƒU{P½ê(ŽºÜ±~ûWÄPŽá_¡kjÅQG]uuŒµèéTæð¯ÔóB|@G½rqÐQ)èëö5Ö0ÐQ§~ÛW}t]Ž:3§£Nõ¸ì«†jØWê\¾rɼnûʬËí«ž§qJYÇi¤çƒuL¥Qþf…°£N3/ßî•ÙYsqÔQ›õSÒ1ŸW½Ý+cRŠ“NÑ2ýª67ë¤Ó:qЩ²íPÐ)ˆ¾jét•æ•ýƒNuúµy•†§­ tŒi3»AGKvÐAÒsŒ”QsÐx;˜£Îí¢BhdžSŽ:·å”£®ª7å„9rü½h3fSæ9]s§œ°`SvÊQ£Á¹(ÇØèâ”ãÉÜ‘RŽºiƒ”ãç‘ž›rº_2rÐkªárÊIÇÆÌᔃ.íá\…Å8òE9ÄX1N9h#˜ANä‚bÚ|ÊA7Þ‡rÐ-ÀE9ÄV)â”.8—SŽ¿õ‚r‚ïHá\!ó´þM! €Í rˆaošC×r²=ÈrŒ-ír²}Ì¡ôÌê0sŒ5½é s2­ræ¨%õ sŒ%½ s±†aŽºö|aNŸ]³cNÇ(r s6g<1§¡2sÔÝH¡˜ãçõÙ/Ì!Væsz èÈæ˜#«¬ sˆu©Å1'¸‡sš{Y.Î1?§tçt– ƒyÆ:¦yËÝA§h1P(èTÏyNKÅü,Ig¤N:%Qêét™$N:ê´ÄI§z/ù$R´µq·t,¿bŸ¥¤›ìhR´t“Ýâ¼ÝëDriílÖAËØn•¬cÝÔÝâØ!†\ ÷Š’ê؇ëÏÌé†벇yeý²‹8°£N +©€לšØ;–mR ìXó­ß° +ÿì°¼ÝF¸Wh¸Û½²üÓÈvÜ2éÙaGÈMm^vMÂlá^1„é;VäPrì­Iî»E…{]¼3×Ýѱò’v};lîæå†6¿I¹a‡ƒ¥qØ¥Ãå »DÊÕvH –­ÓùÁA;ä?öâ¦ꨶp¯¬µY[ÐN}H¿iGï2½ª–p¯Ô]o³íXã3Ü+90¤í ñ€›vÂä÷klÚ ýÇ ;>ºJÀŽC¢¥³—{ÕP ÷ʇ¦îUa4o÷ÊÌÓÝ+Ï3§á¬Ã7A7ë°7­†{%`‡ G3¶ç‰ºIÔÔ騛¼ßs¸WãØvÌQ‡ôÀòÓQG¶ +Ž:jTÌ…:J2¤uŠš¨cq¤*ê«+Ì+å\ÅI7õ¯¿"$DÆ…:~°"™p:ð3krûYÚ™®5øm{sØGÄo}²¯ š):=ú."å°¿Ï¥:Øǘ]’ƾXÃØWhJ4u~ˆ¥òèô8fh¾ýtŽÝ?‡~ e«=¥tÝüPúásê°ëÇé‡äÅiî.Î:=lH‘zwzÈøb&×:=ÅpŽN™§]ãÈξÁ«øƒ^v lþq’§5X@äy9ÿáïŸ?üçã»O_^~‹Ð›oß¾|óÃçO>þãåÍ?¿ûåýïÿõþãO~÷ùç?|ÿöå7òÇÿ;ä/_~yoƒ~‡øÿ¯|Áy±?~Q‰?~ÂýåÍÛ—ÿ¶G-WÿÕAÚCùß?}•tVéÈ@äªMËíß;Ô:3½XC];ë¼£­ªÑ °Šç˜vNÉž8¦NÀŽ’½qÞi Í4Ì»0÷ ¹D€‚[uWmGõΡ’«þܘµ€$Þx§ 'š©ù#š(Ô<“ŽúžWÔ%š:ñ–~å%*g. ¦0H2ôc…Í?0Ëú§„W"Kɼ=eÉU’vŸº]•·ÊÄm'Â@a ³ÍÈgÖLl¼õåô;Kïó2xs%f^{6m07Q¸‰?÷Ä«JªÚ€Á)v¿0ç÷h%=W<‡¡­œ>Ķ ø’¦—Ð({ßl­s)*á;¸Zl0YtžµD”ç{ÒàmÃ×acŽŽcÖ 9Î.O~2°ÉºÒià {/îÖT‹‚„è´W¯)ñÚ†®‰c³˜ë™T3{?å¡c¤ê)©­ ®z›UtaÍçLMkà[#7^õÒ¤T£ +ÉòwåȓCÝ—PÜ¡TÝÔ8x%ïãÕê`ªã²·7Ö˜Õ™=¥M¤l!ôΔ®õUáš=¦³2ºY–ˆð +³I?s;/^Wº·Â>ƒwÑ‚²ì§ø3ÌS‡g¬w+h°ÑÒk¥Ö]È,^+EÉêYZ¹NKË,攬Ogf·ÎŸ…¯~¦´Ñ4¦i å œÃ« lÍB^žÕ+Ü<­¯q¬%R"Ýmå¾;àR + ô¶.jTÇ°|¨sá.þo“qRÑ endstream endobj 12 0 obj <>stream +H‰t—˪.·…Ÿ ïðO6„mÕM—däc™„8ss81±!¾`N~û¬Uê–Z'öfïµZRëRõµªµjíõáw¿ÿöÏøýâË_?~ýÇ?üüÓû_{ýÖgï>}ñÍÇ_øé_¯Ï¾ùþý/ß½û÷w?ýóoï?~ÿ—¯?ýMþú›üã·_¾»•·2D¥=Ú¾{÷å‡ÿùñï?|ÏWΆÂ/~ª¼µÖí¥ùG_?^ÞˆA¯Ô/꺆`Ô¡½õ‘Ë‚WÇhð¼7c“êEØDÃ5Ú²‹ÇõîÖØÆFóÊ6Þs*Ò-fŸ’ºŒVëë«ÙÇlpœâ¥xö=^òÖ›”15–zÇÔ¯÷ø(íÅv}B+µkôÔÞÙ'z²úXŽëµµ\²×0hSÑ97©­RûÚ16)ÒãÚ5×WyëË#„FmÚCg/¬Öñ°`£Ü³ÃvV_Þ¤LˆF¶ ë2ÆêÖ /.|<Üœí½zdžž4ºÇ :xÚµÁSmyö–-æ.AÕf­Ù½möÖG°• β²U/bœjs•dTáTkoW'+µr6­JcSÏ&£)óæ&à…ÚåîµÜ*öc«–'VJ=%6Zò¯>-ð 'ЋÎ.\NȵMÍ%ãàë¸ûtQ¨xŽÛL çNÕø‡5ÖÄzÍ3mÍÜ_{äJ±Ô¥3”úè¾'Ö+³¥ÈœWl)ÎYµL …V Ûó*U™ ¾&˜LU\sÓš± ÒPêÝÅqco>÷ÿ $‘aöÖâE5hÖ +šhµÑ›Éƒ÷áX8†gb0ô< ­òìV|:,mÃØBR²ˆdZ ›9 áÌï<ÂAdÓ p"EFØÔ¬;šež•b0£Ù¦rÁLÈÌX¾Kb8|usK¹­Í‹Îa›¤ŽÌèšC¶DïWzÃ1¯®ïèYn÷äµF,ÞA—¡vðŽžzó. +}ÑŽJ›´ƒ×[[°£døß°£öÔvÁ´º`Ç6Ö}Á.µ¬K«ÔÅ:hŒ‹uÔ½ËÁ:züܬã%t±Ž:êu\_˜,Ôå†ù&´9Aë[œƒ®8©Í¹`2?9ñV²å…ÿ”¶8Ã=ߺ9¯ªÖÍ9ÝFߨ‹Š9[;QÉèu0n*OÔÁ0¬õ‰:Xhju0¢÷ë`´ÒOÔÁ÷¨mÔE{˘¿YGg¼QG‹at£.µõ…:hÑü mÔÑ«" uÐj‰Çd%Ïö ;z%OuÂn1aGí3üìòÕ5nØQFñ»Ùe°K¯.Öå[f²%ë¸8îÚvØ´:ð7ìòTTvŒ’~>a‡pxÁ.’»uÁ.ãÏí€]ƨ›v!Qí ¹ú¤C=|,Üe*Y¸K-rà.=­ w©ë¢]âHû»ÄV"hâ.µØÂu™8½€vQ>Ÿí ¼ëvQÏ;¶iè¦ äß =xÈ–¶¡#/èÁ¨“‚ z°šGÝЃ14ïsô|`!ÍOêÁTo²©ÃzÕE=h×ü–mìÑKbMê¥L°Ç­´¼:Mìå±Z_Øc”COì1œ"5±çoìeüY;°—Ajzc2GÔc\#ªêÁ‹áº¨—˜±¨—É‘ßM½ôd,ê¥[ØK,©ØËq¥-ì¥.maÚsÌ{†0YnnìÁë⻪¥F¤.ì±Ï¼AnìÑ“ÞoìAÖÈ ÙÄu©gU /"ïò{Ô¥íª–ZÛYÕӳ殺m¤Õ…½ÔzVµ9nÖ?{ÐÈ°¶°G]³œÙأǯÁ=jÆô=Žiyó[ØãMwU›»fªÆh哪Ö˜œæ…=øtnê³RÔÂlŒÝÂÇøž?ŠZ/üŒÖ{従ÝØÃëƒ7õ%Ÿ`¯ &½>°W¢/ì¡E?©‡Õ÷xTµ0´Î6õ°…½}RÖúÜûõp€ZeSEJâéA=‚¨ïº6µîºº—ü¼=°ÏË®k%X¢rbìR9±¯ë®l÷ö𶳲ÍWg4±Ç&Y]Ø“»¤y`^ìÊ6_㻲åêâ„ÚÊ.lóPXz ’~@Ñ4b¶–î z}v¶¢&7ô(‡ïʺHœ•-CÝ\õ’2c—¶™å,mÓ“¾¨—:7dR/©tB/‡•]Ú¦.»´¥îò¿¥­!Œ°¾zðxË]Ð^t6ó %/Ûæ!:#q2Ù0zlæA{~ûÌC„¼Ã_̃öœÉÅ<èšúÁ¼Ì¾™WXeŽÍ/õ,p•éÙvË>£öE½Ôr¸9î®o)±m‹zÔ¸HÔ£gu×·ÔÕꢧ&yñ[ÔãŠEv}››¦úFŶžÔƒ‰õM=Et#Ì7õ”iÙ>ÁLgð>®.>¢ýÀ¼y#»¨§‘ìØÔƒaMå¤ÌÖ7õ`܈žÔƒQç5uQV«þ(pa o¶©§<«Ÿ¸0ñìQàÂð¶¨GíI§M=z±oz)u¸Ôýd,Üpw}K=’“ÉÀ• \00 J@DBYŽ!™™ü‰Œ3QÞ~֪ꣷ¢¤úß»wwuÕ×kɱ +²{¬’H9±`å”Ãô·Nˆ<¨'åçv{Ë DŠRaz,ëv{Ë1ëý€ž@¦L{+qô¤_&ó$”|(ó„I„žLc¦»eŒ"Ì“Ø|ìnm„ì‰ucÇ(q;ó;ïó Ý&ñ8âå…xxŒ]Ü€‡!DÀcìd +<ÆAâ <ËÎ̃w|Ä‹RÞIlÊÆ;™¶ø<˶5nq‘¾™ÀãX–yxœÃ¨kà1Ž"ûð¸Aå¨O2&¬•.bžM`NÉx&À³ìI±‚ ð0˜3ÊÁðgeYãÍø—±ï0ÖÔX"¯< `‰a«÷vÅ—:Ç'”€x¶`ድE슈ºÆ; €deçQeqòØ_¼³Í¤¬¼ã +gÏ6£Ó‡¸8ñ “x‹rû)ñ,=˜Ø!CkÒF<Ž9X%ÞœB‰Ç8$·O>\'ž<"ݬÄÓWòF<óeqò~»K%mÄØ/¢ž”xr®Î â±Lª¯ñ0† ý žmôíÈ“tvC«´ØÔ‘‡ÐY?™‡ØxÚ¿…y¬öHyÚ˜'Éu0Ob¹&ó¤eÄð*ô:y@O dÒ=™ÇØ=ƦÚ=‰eÎz(Óó.ô0–½z…*ÇL¡W¨!ËnoË¥ÉÙÆŽ=„6‹Uì¡AT†,ÜÃ<À´·èª³ÜC\p^÷ØŸiº[<’c:±^MîqÚ"¥­Ücóa£r±‹iç^–›À îe^uyÚ[Ä•'¼p0Ù î1i6¹É=Ô(`w{ 6ir™d ÷* ªÊ•H[ìmåZwW)ÇD 6îU¤I°9¸‡Ùq´¸‡ˆVgrÏ H7îá'•©“{ýã{‹äR\À‡ŸB|do1XC\ì­E± >1*iÍMžöVb;í-ã$7Ü>™`Ú[K#&°Tð!Œ&îàÃX0ÓÞÎ)ø, ÊÎ=~Y¬’rOL¡§/ìöVÆü´·ü¨ñÓÞòwSv{ËLº2í-•ówì¡J€£{(§¦½%l³Íƒz,?º¢•z¥¥ªQ¯ðÊwÓÞ¢®3ýÚJ=\vÕÉ+J=RxÔc\(çê±aX¥zB8©G*µŸƒz uº[†¥ú½" |=Ôh¢Rꡪ­Kuj=#ZÇìÊHçwîaЦ* +\À‡Žˆ)„8ȇiÜ }è¬èQ”ƒ}0¸~Ê€{U¶¹Ò] +Óáÿ ýfÇ©ä<øÈN@æèÑÆÔæa Ñøº1ƒQäy‡ ‘¢ª¯Q´plšIAÁpƒL"6'‚”kª;ñ³4à€¸KÑ™‹Dù%9†U"©¾†{ië + ¯>ì +PtÚjyuéÓñâ’EV(åÁâÄ Çœô bœ"H:É©œã†Ad’éŸ~gªÊ®©?Œ{võ@ù¦ú"»©þP}ÖŃ8”Ä,5"¬.æ2@ˆ“´®Hþ& qÞ^ÀÞQHö€NSrÄË QZ„“C +BÆ‚„•„˜…Ë)ƒ…ºê04TAH¥4¼m4üäêþá‹W·¯Þœnî?ŸbèüêâøäÙÃý«ÓoÇù³—7oï®ÿ{wzñÝÍÃ˯¾¸8þÎG¾þÓG¾ÿðö®=„f­¸·óòìõõÕíí»×Oß<Üð“úàgøƒÿàNigüyþABüãþ‘Žó‹ãùòÔ—g»úÊÿò¥|ïáîþ¤ñõÝo¯NmäÓãüÇ›ÿÝý~´øâìQ,Šß¡þ¨¦ ½TQã8ßäxTz7–cærþþƒïpüóøégs¼ñçOÕ!ls¾^ÆÚ¼Ç7ÜÌþ­m¬¿ûÍÌDZ“|î_rè8É(DôJJQ~†(ÿÏ_´š_e5OÛIñ¿¯µ¿âñžÓÀ2h„Nƒ –%` ëŽX$”j‚„°ÂÕ¸Çç åGngŽåPù0Ћ´ +h³ÈØ fU{¹öôËà_&‚qC'p:éÈ ]€ò‹âQµðiÀ&b›PáGŠXôd —ÁpÜó +ü*\¯ägŒƒØöxqßmà„ñ› hÿÈ¢[6‰gв¾o2à2‚R› ø,•ôºÉÀTâ¾ê›DÌ•õ=†È€Îë1m…Ô»ôå’Ò¬oÒ¥Ëêx8m“·m“ yë&-/Ž”Ç&a2–16)õ‚­›Ä3>Û7I_¼›´ù)vÛ&sÆmÅáC†õMZÔ0¼îÆeÇ­S ß7‰7¡@²ÿh—…sÊ8>“‚l2Ó¸a“ž¦ Ï|a„xŠÜd‚dÚ¯,Ž… -ÀxX¹®zĵT¬ÂèŒR:´‹ÿÒq€ïàbClXS˜7‡,1î®À"ÎC§Â²–¬A¿EZ­X8‘G!wŒ+üB:èÚèpÜ«…­ŒNuFÖ/ÇŠµV\cŒ¡Hõî‘·²~ç€B£°¤åsh *½–2yÅ^i 2ð_6Å:i9Ã_uäL.qCÑFÎP†®æ0ræp›WætÉ™£rÀ•ÚsæØÑXBÏ™£Á‰­9Ã;!1-g.à†b?¶œ9¬)ñ¬–œá[¡w{Î'üÕSæØsµn)sEí9C GÎ\F ðhfÒðe— õzaõP±'­¢‚]*[ÒX±ÀMWÁ¾Fö:¥$kÊD<ëzÄ*úÿpU5ê­–,teå:K±‚@X/4(l®ŒÅq¯˜ºÆÔßÁBÓ!^óB€99ªB°6‰¿ÇںϠ! çáR’l_/šˆ ±Ibx(Ý$ZÔ'ú@~’-ëh°¨‡hÃÊ +!t‚nóB© *L@ÿc;!ÑT$l™À±$ iØ’ö¹BueGÒvË[µ%Í;qM~’íIóÄŠ³#ižÚž»[’æ{Ò0o 0n=i>Бø-ižè¥9ѤyjoŸGÒ<û1ä-i,"âHš'V‘œž4ŸZ ,Iç-k ' [EÕ»‘4pÖÖ®&M8‹ÕCÅŽ¤V´fKšÞý{Ò¤f«&‹DÅÅlPVX§·®t¸,‰!‚¯|Ãn³HºóÚ2!Òùà‚€æç…Q=…îù¢ö°È1jׇ’ú+©p¸ñ¼†¨=pOã¢3BñÊgˆ^¼XI¡As[Ãå׊ýŠúH§MšêZßöhÏܹÚR"ꃼÁ_!ø©>?lDŒV.—'h (êöuOÛçzÀ”‘6aT*#mŽÌ¹¦Íó`1OObçñNÏõL±qËõL´uäMô j¿çz&íi9Cü´´‰ž/zÚ¨g¼õ[Ú¨g`bGÚBgkK›Öñ’64!Í´‰¢Ac´´±=:¹§M`kõŠèiƒìÐJji³sbÍkÚ vàfíHžÍïö´‰Bª~Kúy¤M¯æ–¶ÇŠz($`¤çM$kË›H$–Ð’7×äfÏ›h$à£ç©lY{öçF*2þåÛ7§ïàØ`Úž<ÑañWëgß¾å/E¹¾÷ûË>Ïù§ÓÍ뻇»83Çþ<öÿm†ê픘©'`©\ð¸¬Óÿ ¯’]ËŽ"øïîÉ^Õ”U•K»Åô’E#„,«»Í¤÷ÆÆâîy=`[w8ož¬"#.»1Å}[RÅzûÀïÁè»(ßsò ø8yöð¾~ˆð*n¨ÜÚâg«Z¿pkð6g€Ñ¨ãÇb3'ßûÏ÷Ѿ*Lž—Ýnö Ba$( ã÷€&çU~N÷y3_FaË…1ï‚`ß-#¶w|4ª7'ÜébÏÿöî 7öªÃM Î™Ž[ÇdhÜÓsÛ«fÀ°}¤ƒ¥Œ"´°GÖJ¿}4‚W6ikµ]>÷D¸4f¾XíM-Á¬‘%si°– ­5™iËL¦íG#`hØÔ:ÓŪ5³xfy›ô*uÛÙ8VžQçn„±:M§çÇv“üýŠï%|2´ šÓ­©á!çkžßÌ-]Ì>½# NÑ|Y8꿃>6Mš ÌL¯èÝ(7½5þ xlYVíiÛ¡ÅLu"bðó­0Þ ®ãrV:@ ¿Ó2/Ø»3*WNJÞHxHp–ã dXÀc.¶mGa+á°‚ç¥(ñHønøˆ†Vï$;R0‰Íâ­Xýê9ã?3Îê•]‡{0m}ͨĠPë"¤…m¹N·EVŒ)š³óUuMŽiæ +wçLhCSGakwZ€ RN¤ ,ìÅOéÝ(áð`”‰B(Ò)E3…€ï„W¬ Fƒ×ô»óp1YFD€Xâé ¶,-vjì¹’½mN×jüV¥S5K‹&ÇÁ¡›ˆÈ7nÒðî1y½·®ª§¶m°d¥¬ÍÚf¦†S[õeÃ3?zj»r< +N2N' rIïNÙ›ÅD˜¢%?’'Îj2ôNˆá@Ñ\Ü|Õæ…Hõ<àN“-+ÛV½ólë<;Ë¢m?+oìOÑlÀ÷ÅMœÞ/ß%ê +P+à­KÞÄÿ +†fkátBŒg>š¿ Ú¶¶ÂIžË³=li“ ½ö ÑqΗ´FÃÙ/#˜®ôÚóeŸÅ%+´=áʆ³°»C³ú‰6¶‡˜ä=#§NбKè îìn‡OÉlk{jfÁ¦¢ Ñ$×ÎM¿vŠa½\ÈÛýââÁ’¾KP2N¥±¶\l D'VáR„Ö —M¶ª<°…­dÜ{Þ÷»\öW—à©Ý/ƒ?¤f`[ló2åe‡­’1ìPJZ.ÂeÚYà¿WHm¬®¢J¡\`˜Z= œÁñ{ŒMþ^/{Í Y ÿ,ë±=³Zä"ߧI ®’A…sÁ vÍhtâøµ-"Ãð”ÅNž[ƒŽÔø{³“Ô[NA}8 ²Ü7ƒ>¹)œlâ÷Û>ÀÖ +Ô`Á+/ýe [Ø1Б;AÁè¾×1Mú¬«ô #,cïê.Œ)™RÑÅ 6™ðEö2Ø–µÅ8=&<‡ßýö¥GιztË~ZÎ,ý.Lò%½ÿÇë¾{ÝZl\6(fÏmÁGÔî g×Jœ*ÚþtM¬\»~ :û[èu–ZIìe¤ª£Íý|:½èÂË› +^Wå/=½ÐWÚUªÔÚ4A 9;AºØsLö˜é¡5·NÛÂí°¶‹4·~ jéþ2L<¤Œž¶CbL1$&þ¬=ñ0‘´ƒÚÇó®•àÔ;cb`¿T·~¬ôÓµŒ÷"À*ïKÒA«w¾~Vé§oÛ¬ƒoŸ0y¦qÛžPSP “l÷9+[cðk!†ºP¼è&ÚËp0/'¡jmÉ m(³PêLp,Þ;8|š¯M·­“¶m{jªmÈ„‚Ö©–®Sš~!0U&fî´Å1˜éöOpL‚Ö‰á``Æj>6øn·m:fO°lÍh¶ ùëRK'Õîž\[gËøÑ"Á~ s?h¹SP:\ ë=ÊLÛu>V“Mܲ¥¥³áÀî½> kD`$³3+½e_˜¢¬•ÍR²¹ ¨¶ÏÓq1^Ø+PF™ˆ¼8ª0×Æ‚á +c,¦¦øÂÓ‚%²+sÐA”…O ñzzз™÷¶m¼~2nŒgà8g˜Ñ!krÆïîëÑ3ûª—HAIu?—\­®úÏuæ¥Å>º~ +DÒØ×-râŸz²·³«O!øˆXû¢¬6KgH“2[‘ë£Ðrˆ·"%ƒ!•æ øðEÎê…’IíœâS}?¦bÙ…Ú ˆK£¹¶$謰]œç‚Ô}{ÇÉ 5î0`Ïh»’€£ô3G ‚57Óí·R809ÊÞ—ÎÛ'Á7i&¼ôã`®’/Ó=F:h"ÊvK- ½¡‡ô–P6ªI0²1Š¹LE kŠ!‡·pš!·ÒóÐuØ(bx=eùžWî‘Ï5;®æÍAŒ ù²ÅÃðÞ\UHs»ËçµÛÅÅ÷ñìšÁ°v®#wœààºÕ½2½ký|ÖÇ^ó .…¦©zil©dÎy¼ h»òµ'¯jžk}׈vËÙñJgÿ5Á:OP]äüž›Ý²ÞÀºÚ%‡ n’oÁ)‘hŸbÛf î’²˜ÂáRjÏ š,ÌÕiñ¹lGŸÃÆØžr¹äÔ$xœaH8uPt`'ò¨eÅ'–EßúB¢mA×,,óg£€Ö'æœÔ¹k¥í´äó6)‘/±Å,ÙS˜ÁçBt˜[ù;·w³@ãõS^‰K¹œËz2ü"»›—ܶsà1a2¦Ý]âñKÐO‡•]A¤¿pý´£eRß¹×~’ÐÝFð¡}ó ·Ï>¿½þÓ‹ûÕ—¿ÝßþæùíW?üôï¿ÿá»|÷Ãs€_½ûÛ?ž_ÀŸ}óüüÝÓ»·7Go€?(·/ñÏëŸ~ÂÿÕ[ñ¿_ÿøþ埀~¾Ûïoùk¹½5Ë?>|±Ðd7Œ´é$„‰ô,Q 6¿Ã/Ꭷ¬å<•¿ÿ¹ÿôÙãùú¡ =7hCü£ˆç_fŠm:€;˜¤Å÷–»N´`»0oði»]ÃÚ4 ô7Åïl ö dŠ½VÃÚ6VsSx˜u',ÕÔ C5^YǪà’0¡6Sùð(`ó] 4þyº­©0Cxè¶DbŸ`u¸[|ℬ¶?§ {å¿E6¬¥ý;e |A–i d †¤ÈZÛR‰¿&R‰iA©Ñ›‘PlDÕ™¡JŸ5RŠ¥êG§nGžL‘_ ‰ p…(ñ„‚®ßx°3´¹¤––=ä@ÖAþù!Þà”°,Oóÿø®–UKŽ#ø÷ÎÆ y1Ô+ë±”/lll Ï`Œ3ƒ »% ÿ½#òѧOwŸA®Buª+3###-4R¬l{}o1°[š°AÏoÏÊžÕSXï< 'FQŽ5ª¡:W-ƒC¶øk+—#±j°xu±O3ðÖWµìç£y€Ze€4‹˜ÕJ‹Že-í¤X5…®ò£Â!PÚ›Ž5ÌžÊÿô`I°º ô’Úr%¡Pp{ê)Ö/“¡­ª”‡ç®EnÔÎFîuò³Â³=YìÖp £&³hptج–wœwá`JŠ†L{;5d\58ï•K}ÙF¸£ óðLä‘=F/Ø,wb?G@Ã’´¬¹‘}ÃΊ-H‘Èfô†]Ë°˜ŒŠ†ÐÖPm' ‹iÿUbþ{M¯ž×Ö +0Ž Û$SØÞˆR8,ÏêéÞ +ƒ®Z3ØS7' ]‡Íl Uh z´øYHF.Þ¢ˆ‹´ÛË3h·jœE=0Õ¬Å;ÉFpš[W°¢Gõhîââ£.rô +‡ÞM¨@ÀÉq * ­¾=6Yl ¨ÙèŒàõ[˜ÒA%tK½Ì×) +fGpè>A4½Eìãú [÷4Š2Åæd9Ïpã%ú\WÚ1 +,Z`½ÐÔòV×"å²FüáÅâ@ÜJÜü¦tËŽ::Ñþ#:×}ôNy«¼<7Ã~¤iØí*¬'òÅ7Õ^ù&Hž“+cWí½Øƒ$-½ « Ub> þ­ÿ…õ8±<6{jmÙïxjÔÀ W¬­Œäè½Ü:–!Ô€¡¬?¾”+Ô2ùî3̪ƒî§‘ngÚ ;‡úe^ ´´*ó/l`¨n¤&í;JHÕ‰igu.CÀH-q¦jœ†M›©gP•»•*OïÖ[;ˆÑ\ŒCjSSµ &­-@;ve 2 {[€hv–NSÓè'Ï +w¥U"UÛhƒ„iX…O… fñ)Äf) ;XͽÝ.3Cíúù’](NƒO3ë=Ë"ŸY5`éf¯BöÓl>Ü uΉެ¤šª\7Â47“hxÅZ¹D*ºšïÕá‘t*е]ü«°2£Á{xÌ©“*csŸÊqàÞb0´"õê~ +f§ªã¶GŒøa¦ä¶c,Ø+ÕDFÖllC³Ð°NÑ–µo Žã²b Ý^°JY“^ç~Í[Z±4ölÖ]&¦Š:%íµHM¡°(¨9ˆÜ®®ÖSJ_ÎâøZÑ$)mQܲÄé%R‡KRÛÔ§¡øi§}YJûÎÉ…TúÏÒ½¼ž°ž‘»–šE@;þÎgHÖdH}ƛٺF¹üàšæu½`4L`$°T+öÉÛ%?¹-™'šP5üì¦ú]u.hƒ9‹ÂиaÎ¥ÒW¨DÃÍi\*(¯ñ£ñáÓiƒ€!ËâúaôU£:½ÚÒÌ}Ò¤ ôfêÖ‘dvÓækñ 7 Ž”¤®ÚØvóO4CÂbã ÉN:l@쌵Ä;ŠØu²n0Œ^5Wz+¬ì•à˜Ñ¤±7’& +O/ÀÕlàålO·ÌÈl.lÅ膇‘ïv– ³Eˆ²%64¸ñ4‹ ãCGUo;°T‹FÞ2ÛrÎ&a04Jq¤ –쯚üK«€pª¨«°éÓÌ9&½!¨ (^¬¶XÝtViÁÝ£^€)2œ‚ÏÏŠ¨Ó$9Ú°[áî›.^ͼv?\—n@zvTs°ôºÙ BwþÁ{-)ÆÄ4ûòm4ûyS0—o—œýh#®Ü8E©Íe³•§Òþðœ'Ê|xyF¯K"žIk\>Q\¹|Ñ —mó¬Å.Ûñ²q¯ö,¹—e8ì ¥=ÑàŸ)O{k²r»qaoHœ¬˜ ±3a`üèìð|«ÚȘ— hÙáy´<Cç‘e£åj¼]Â'CórÀ^Žâ'cûjÄ_šgÆáÒd\Ú‘«ÑðlŒ\œópzûtŽ]O<:Lå¬ÐžM»•k‘² Ý1AÄØLŠQŸ¢þèŸBA»5LÝò ÆTvåéFþ$tÇß½¬ÛW_ßÞÿåá¯_}óÛù÷ß¼~úöÇŸÿóÏ?}ÿÓOŸ|5ðÛÏÿø×ëüÕw¯¯ßÿðùÓMÑà[ùú%ݾÁ¿ï¡i¤éÖÞÿÿñ;üño@¿ÜÚí·¿þ-Ý>ñäŸY{m¾åƒ•ÎI­-ß?"àÖ6íWgDõúòk¾àÐÈND0:ªk$oÉÕ±f˜ñéF¬ù×—ö7An°Oþ屃‹éV0ÒôçöfC‰Ë_™û¬eèb éÔy©œíÏRÇÓ"E¬¿k0‹}ȥ˴ Òz{®VâlËË/X¤†2¤ìÀ~ï|vö\O·®Ü²¿€-Ò뛵Æ\AÌi?ÓšKdÚ‚%h We•ÃYšf4ú–Y`ãñßïQHúò£‘Êáè–jw0.H©Ï8qË7nM5;ŸÜ Ĩ"`¼d+âÎ*‘ˆjù8Um"0T/¢ƒ‘_ϦžäñR<¥¤qÇ¢‘ìûIØ¥–üK*Š¦¡&Ž´ièíŒ80Çr¦ep·Ø•êZwÌ~ŽU¦ÌÇ“h HÙ0lú8Û¨iþûŒQçò&jEýò0Í—í¸ÍéÎà¨?¥’ÂmÍ¡3’g—_@ç¸dw2Ò/^ÿ~g0jZUs‡›ëuX)·@A6¸ à2Ø°‰ $—;f~ù›åts¯?Ç„eŠŸïàj2F°V׶fn›ß m”PÀA>ôg¯ËÏîÈ›î¬fâcËÈ­‡uÉrÁ‰5£ô“ÒQµrÁµ ä0·zºýÓKö³nUÙ5PLwEÚ´ @!IópVé69؃¹…v0R³|;A®æù‚^ÊPÊ·Ö6· f.R·@DãµXF’í‚;œÒš‡•7cH\áíw ]à¾óñ¬Ž-ϬHïÞnm¶6PXðU½«$﬉Ú+ØáwÚ 0¹ +cKœuGÁ×V¯q_ðf5p‹O,xy©dà)5Ù[W¸Öá‚Ò‡S]ÕR“ ^Øk°?‹$O……Ûä#XwƒäVOg`Z*$Èp°€Ÿòg9/@zÎg»åk²RvP,[ñ©·~Ê7 ®t îÒ qÖ˜À’‰Žcö}¸˜\ëãI¬+Y§Ë$±Êƒ@ÌÛφþDüØ‚²ºo€n^õMK`$P™÷xV%õWw¬þ3E8[MüyBpö]÷YMâTé ððîlÆpŒ;WŸÓ‰ä‹'?„‘æ$³ + ~êg;—‹ ‚¥ÉŒ ¤õu:«ó˜ŸJ"ÞÜì1é‚bP2)c?øvùŽÖóös‡ñš¹?ž]ã>êÞÝÁ¨VÊãÙÊLêYfªõ…]1¼àìšµ*ö16Ô +Â׬ybï`JoDùÎÍf¢¾óáåcÇ zS\ Ü’×+9@:áè®ÕÇ4¸ÔáæÑûÛo0’¶pïp¸ö‚‰$Š#smýYu_˜\if¹‡ÛtÎPLj OXø\NàH-Ÿ?w—Bƒç&0êÅïà&q©äÃÙfÚ¬a#„TÝ?¿‡{°f¾çn5L!±-Ã˺¤n?Ï]EMk=>¥V@Á:Æ “ZÊý÷ܧóu©2Ë?¿‚n(Tw`Ì*NÍ-R“²å Ò#ˆ_Ô¹ ßa1'i`ë©YÅ”§„™ÓÇÚû6ĵ©ŸY¾}.n¾ºá‡+ÙŽD¾Õä`,3k[IŽÆ‹]ï|Þµïÿé.·UÝŽ" +?Áz‡u#¨ tõ¹/“/”-ŠˆˆH؉G²/bD|{GUêÙs.E‚+_jö_]]‡QÈLÉ´-¥-Bñöp7ŸB‡-2³ï^;aáÁú3E³Ù¶',Ó_ô€TñÏ°/%zØü•êî7˜d´xÑÑFØröôZÛÅâóVl¥ºYVI™G²Mæ1™&S/åߨÔÁD1Mí + 7é€Ü¨zàõò^ËŒ°ÅwÌ##t?iïÑŠwN[j*Cd­÷g¶óŒRÙxql‰Týña0\Åé^ …î,ïOC«¢Ì!ôÍZæÝyëë™&ʉë]O{Fh‡~‹U¥ØvçI Q¦OT9æÀâ}ì°$0/{]V ¶âÁ¯¦É"?ÍþTë+:)_CÈ“\_¥þÕN†®ñxú³è®TgÄ…"ïØSê kö²ÇS^ÂjXc«ŽûòJ¬ÍÔ`[e¾‡uvÃZÝ~×,®Ã±ë®—5}pÁˆ+DFygËlSÅÞ[~û^nØÆO-–Xº2Ôtñ5pKca.Ä! »ê¿`ñá}ƒÖâ€ÀXG|?–%X_¹†WÉ`™vƒ?¿Èë d^QÆúîß-dEƒ¹5jÀ¶ây¡Qä€Q6º–=l‘³°‹¬æÁ9`Õ­ ì.<Ä£kл`ûT-Lï7¹2qµp§7<`ñ–¥Y×;O-j{‡p«ÇþzÇ(ø°ÍƒÍqä§Þ#ø†èò}d²¿Êí ·¨”Ñܽ¼î*Ÿ#¼]ÒwE7Œ.ÿ*Š†Šcä¾k¯9`Ülfo)‡-\œ•¬é›Î!Ð^yÀ‰c‰¹ÃéBLaK¾ ÄÛþ^Ú6zÓµbƈ¬wpªlÝCŒ¨†ðkõáájW®e—™fg[îBÍÚÊ®l]ßa›!d¹ìÕ©ÐR> ;P£¢N[;‹Ä> qï¥7íKYk•p®Ò–óoá@‹BþÆoIòa™½–!eÛîÉ¥z+@¡ B½#8e¯É꨹p¸ÌÑVÐG?àGfñ*¥ßmÒ€kåd}ôúÒ“³uv !ýFñ ÓXY%oœ+ ¯Å3B%§Øb ~dA2È;ÛÊ•o± ˜ó¥¾{åXë#…·Ò3:'3²È~Àx…^KØb‰ºV¤æÚ퀘 =®p`ÕŠy‹L¹F(*8–žçýèö6çÃVµÊV ÿûÇ¥ ;¼ZuÈâ¬}Åòtâ\Z¥cØz;øða¡•WhÀ¸.v¿ü°8ËÕÕ¸Ã:23ÑZq¬o2n‹«þ¸I;“¼ƒ¥¶k}¼°Í ‡iø#$L«,ií€áÀH£=l;dmƒÎ „"(ÜVòÓ‹–XҢțÈx@8ˆÍs?B`¬,Ëç°Â1¤£>×0_ÆÜï¶Ëã­§Bàði%e9àã +wÛ"‰¶\o¬kQÅ÷ŽâP,gÚø çŒåæÀM§/M¡ÈÉb„jvŸJ„; ñoïl'2‚‡®&ˆo$4ê ÷hß; ô/ ™€}ŵ4ß^nÉ¡!)o¸^pþ| ´ƒ®-æðî2Ú¢çpŠ´F!`ÜÊÃvý<ÓYýŠ¹yÁ~l¨Öà}§ÎÑ ê®Á¿“‰1 ËšcÁÔcÆ¡¹Ž($D0“­]H­µ€r´ÎÆÁ(W!µý²ÐmÒ£uæjm ³pæ\vL×L:"ËŠBê®Ý‡ê¬%΀iÔ«|Rç«õÎdË,ZM’Îަ͙1LqùçˆFÉÝiöá¯1´¸¡M/óT±Ÿ·>ñjÜG°¯j0}§k#öÕ.ið#÷Õšò|ØBøô‰‘6|l•\7uÑÈC8*ûäº~r×HeܲùL)¥’õd +W›øÊW…Ú:èx潄m4ÀYO–ª‰ÎÃV}ï¡wR-d#y—Fèø|ã÷ìG:¨B §]w«^Š9ƒ1{>àG®Zmƨ[ôé.œ”»Í¿{‚7ëÏbÿ2«£å~ǺAegŽÐÎý³ úTjWåŸÐŠ) +/0Fsb⪜v$-U"kƒm?‹lô§Î]ãÞ€¨ŒHŽR·ØTÝÉÿïU÷ +Øùc(‹Z÷†µÙ€1jMëaÛ·<¬¾|ƒñ8-ù°Õðyù¢ò(Þu;mͶHôu™y?OËÂU#£]ÇÉäR€…°¼þÏ+èÝ~úåË—/?úì—óO¿øôÍçßÿëŸýí×?üðí÷Ÿ~þí_þöé†üå§O_÷í7¯F__ëO^Òëgøç«¿ü ÿ÷þ‘×dÿûê?ø—_á¿ýûµ¾þúõL¯ß¨õïôØtçô ¢¢AQ[dSAÉÛþøãIªXøð8äÿà}È'üòo^Ü9yýÍЩ4Þø+k6› Öû@Jõå¢óV‡­xÆ?Km}oñËÅýÔ#=åð§M^ƒ2½¼ÿq¦\‚M´®UèUjª"Í ]p¼W'$Ï´¾¬¶”I÷ƒÄ“¡T–»–0ü Ð‰›Ï,µ-•?ÆéCA8ë Žû÷È£dÓ Èòz +i³ ˜dô¦ÅiÕäCΚÀø@ oÇ=:‚÷ΓáÅÈ+„Å·…2Ø© —¥:`Ή§ÖÔã€<{Ä|ÍyÐÆ~`×ê€c…5Í€Ð2áÁ¬ü¾B!¤0…Ð|%¼‚¼{‰¾m t ÅN‹Mí1:ÓbÄÓ²n+è91ÃjnñîÂÆàhO®X¯î/¹J°\i*„è ¸Â±ëb¨žù¶)Tñ×#’ÑÞÉ¡3Û¿×YÝék(x…%œŠ"Wri]Ìý=¦{‹”•ø¾/ÿ?•}ä. ¼CLàÿ²]-Iì¨n+o7üÁ`Æ=­ ¼ýo¤E‚Ȭº=;¡pr0Id€¯¾KåÓóÅóe†|ßáO¬jÅèH>Al|ÏË,(¼íú;Y³à¶yÛ¬6@§²}ÕðÑ[Ó<îxãÖ‚q=ç ovÒÕ´zkAÀ'˜V',Rw›ÞŠ»pQ©|×I›Pä²fg[º7§|í«¿ÿ-ÃbJï‘OÑÜMù˜î'H¶KXŒO1æÕJÁxžìœœò¾äá}ƒøà†<‹¶—ã¼ØÈÛX|ó +Gùpm¢l±•[*@;ƒ/_à5ΘÚÀ7‰æîLê»n½1Ù Ê!w¸Ó&ÕίÈ+ P8º·±CÉoì*Kà_•ý è,6‹5ör·Ûûõvå$豈þÎKqßÆpï럜2àiΙ¶%ÀÅyŠúÿ: µáÔŠ‹‘[Ò { !sÞ‚iò¿ –`G†íïÏØþ[}öWAM§u›3áyÎâ³µÚȤå¸Íê¯b=ÍïÃà3,ù_øéëo ±8•³(*̦Π¢O†™Ï5Òd”ñK¿@×­Ûñ²¢ñw~*ÄöËĶ;+S bĵv,“Ãq'[–z¿µíðkNôíQ— Ï¡ÑĚų6¤ù¡J‹MPØr«Çg}_6ªÔÙ¥4sЯˋNÌö-õŒµªƒT¥÷¼F8D¿ÎZ¿ÙIßÐ[gà‚½ãºI ã>UFØ\eÛËLæ6¼ÔÕÅÚ¾v>‰%hÂÒœz0p츔¯Iã°mVƔ߿Ò>e1+ô'Õç}äÜàºÿW ´Áu­A=àmñáVΰش׎%v×ÜŸ"&¨‹O.!Ö {Æ<žgÖÏ,bÙ¼4Sóvä‰SÞ ïY b ¬¯ƒJ>¨§<çHNpMeZ©)Ç¥åy|<àE«´¹áRªÞVWùh®2¸á73à“5À´­œ€ÃIõg'M¸ÿjsn#XÁ è¼Âúh-5À#ä4Å^P`ÐŒbsn2pœ­5î´ prî¤lÊæêu¤UvP£AËÌÛ|öAùQÙmªq+p¹+Ksx_ÞSú7'DËàETç„hŒ àwx\W€ò!ëgoIðµJ÷ b|•@zÐöç'Ý-ÜEB¶°Ý?„w²iú$d;UHpÔÄèëD ®yç(Áƒ¦£v €`& :;€´Á}øô§àGæ¶4­íx?ƒ}Xúi0îSçšúªMh·v¼N&}[»oLJ^â!Ü/æàmÎar⪓2Àí(š»¨ò#NŽ¬M  ôÄ…5«þÙ!}\¸v…–?#W–—#Ÿòγs%8ÚÎÕÌ6Ù‰wq‰ï’¢Á5 ®›Ö ·âÞ^@ŽƒU7â¦QpKK[½9À»i|N«vC'KPŽðìž»À ;ù¯f„6êýÎiEõÔ°ÝÁC `ÿÓ¡à¶x§µHÖ6[¯°Å¹2±z/oCŽñ-Úhï[ ¿×·Ë´UøÑÆú§MCÞÖè¾Aì!?DÛæ=<Óð«¹ìmŸÿ)á ÿ¶!ÿìIñ*»uøYí”5œ~Ž·Œå±XbÆ’ ¼ys¥Ü¯esÓ“¬o{»çÅtF[ŸßpE2±`²¼|&´µfñ)¯OË0ì™Ü<]ƒÎ¯G«½ Ìx¦åBà ^,©'ýPr̃kõ¶Ñv0ÇóØÚuR×/Û– W2ÏüîÙþÔ»Þ¥âïà’}p +^'pù‹\œï¹nEÚÓÇÚ2ËQ†±Úa,Óþ€ µ] ßIµtÔÎÇ®RÇ].çð&Åj£Hþ²±yŽ–0{1@å+g®Weú®@Gw&ßú=—J;¬Ær1šQÀ²nõ;Šwµ¶Ó¿@ðÛü”*`¯ÚƸ±ÖÙµ +\õ®öÏ|ç¿_­•^ #Àc³…ÃÔ|öšAŠVKwßÿõSè§3Bí¾± ´7ð¬w€Î÷®)€t™1ªöt¾]·À¥´?{v§½’zGæ!”õ3s¹ø¹mLJX©z•®ðÜÌ6šÏ;o’7Îzs¶Ûçf8‰ dÜ Èÿ0€r¹Ü;M„Ū $¤˜† +°•q¬þ^´G´ -Às„ ~¿·Ø¼Þ0mjA”‚p«‘|ø¸º"ž-€NûSEZ¯0+w`ïÎåç:°Ëz;?^ù÷Öª¢Ù| ðNî#X?ƒ%ÿ i\öœª––Ó\ç(îZ‡Üç»<Àò{;.Lîs iN‹¸³ àæú\¯'xÏÐ7ÛŽ\7mX4 Ž^nåÍ*6ÁºAºçªÍgµDZ„§×XsóûӚʣڭÙí†?Zí-Ñ™ýÿ¾ +;A…M`€­o=«n¥D„½Ìµ+¤Â±L¢ïTU¸•{›ï}¦Ñ’i éd‡·Ußc¼ã÷GN.E‡ñûdÚí®¾|ƒY €BeŒÉµS(bîà‰‘fZ¦¾·N¡Ý6S«‘©1·>øø`™Œ"u²Ç÷‡£n,‹k³c'÷ÿlB© +_Âù4ç*x8‡†EÙ>Ñù-£‚òM¨à¾ý8×’{oÊüß~GtƒSÂcú‰´•H…iò §}ÞÆ!ïmûf+›Øœê\‹V ¶ +pË6êš/¡X¿v€|zeâí3€ÆÅQåíìÅ \µÐÕܹ2€õ§<WpÓUo¸tó7Ê%^Þ¹¿r¼bd ÷–.|C¶ž`Ÿ]Ú>×NÖlùòÀV}Y6uçÜðØZÏ5³‡Òf;¶O.ü°ε-¥àöhž*-±Ö¨*­á^yA(„>WÓhuôåwÒx¹€gc¾vã°¹lY&æ9¨Ö4}œçKUr'e¹ê“rؘ™£5bˆ¬Ò¶Ãñm·Hd×*døzåÃ>5[™¾P€[&°RKÖGuZ+ZQ'Ÿ0ÓÛG’¸Ýà¶jÔ¦5}6ë\;­(£Î³É¤NI ïÇšwÉÓîó;•ã^k~I»Nà+¢f߯C}A: ì3‹YFÄPºµúáÚFGP ƒ•ÕdŒU‡)|orÐÒøðn•XΫ€e²¦Æ8rZôê/¦ + Øff« +›Qø¸P'» ¿¯—¬Ü¾QJBÈòãv"Ùþ½± ï4—+«]fyóóu— ®ÂœþñT°w÷âÔw-ºÿD{´3Ý/…¥kån¨Æ^a9eù%ue½:%D4Þk@š*ìZ£¾_\»³¹\Â’ùJi#À}è}Ÿ÷è鎖:ÚÈÙ6Œ°w:EØÂÎN¾§ýIž)›(Ïîe°¿{Ì”ÀDÊB:ŒgÎÝö`æ§{§‘w¦=W·0¬·H^ ÓÒJÙåÕ„E xN>ÙíkÆíjÕaìêØ šñf˜‚òûKv<ÛŠ©ô¸±«œwåëÂÇ3@ë/™Ú¥Íßž15>*“ÖlVÛÉ$#ú ,X7h1¡ø 4Mþ“6_~ÏäÁ¤WS€ÜûŒ:•õZxÎû¾³î†…&xu¸y P¬'¨M3ê(al9 +HXÖ,Æ>32Ϊ +]y-wä°þl6ºå º ¶îiL÷çhÒÍÍTc>8GËmfTU3æx_÷fò)­‘Á<“9$GôÁ‚Ô$_»-·¿3m‚ƾ]£Î„¼-®ìóq"n“›¥;ωæZºV'Jm|PŒŸÔþ$<Ìò¾Å‰_Õ M8CùXÏ3a)¦ƒÔ€¨Öà•ÈŸå¼ØŠl[W¶טoütaÉ=ô!½ˆ£‰='ôyç±´Riÿgl€zø¶£Ígéõôù‰ûoL½€B9¨.Û<ë¦ðn$ÿ CRjTe7VÜÕ陟Yw±GÝ°2î뼮±ö½nª°ž*ŒslÈH£4tÌ=ÿØ{LP¸ØN°•ç˜áÛ"Ô}‡Œ`-J /†²Y-·“s#¬Ê BÀÒpßÀ†žïÞ¨×fg…{ÉöxhÛÚp`äÉë%"€œÉ¸=Þ º=ëhgìÝx¯iéì\·ó¼³ßì}‚m(_Ë{0hs¦ƒhnQã#Ï-К#ߪ±›Ó»Îpê÷e:ëZcÁ%¨$x  ¡K÷Ì+ø´aaÝÒ¹àÞ77¬Eq § 'Ÿ|„Á° ]†•™…ìï(^øîOÑÐéÄd–“MÍÜæfª!ÄÍWLÁ6¤Æ,ßëÑ:à‘çrG }@]· sÍyױɫæ“Ã&¬š¥ ø^=ïãÂwÆå\àqO ÂðHS—.ÈAýôeã¼­ÏEº6·ó,0N’æÔ°oPWôÐaCêÆÊ2˜^õ ;3ÂnåÚ5x€nm$hÖòmµYzcˆ_a¡²¼ô$ÈÁÀ¿9<Á¨Ùp}æ¶ô蟚Tu‘ç x±êÇÙûŸ¿FÅÊÉ,vL4 …ɽïE¶ø„…Õ¡àžÎ¾ó Þïu9yeXúµ°@ù†7±ê†‘}P®AÝÄÜiŒ–ü¯žÅ“YÜvN±È“Ȇ&(b lµÎe ¢ïÕýäæÒÍÏo6bÿu”8ci ž† ™Œ`fdV ‘g>lŽ¾ò0X£“¢N¸\ýÿìÚzWh š5‹[5ÞÕGFÒk`;Z­¬mŸ²†¥vv¾ëIG…ÇXD° ½¾¿F\ +«ƒcðésLŸî²n/€ÃµdP ‘éB²Œ¾܆b€©FŠÛÆ]$»Ë˜·)$^ñ¹+ø}:¯-#³·1AÜ’užwµú¼ÉäÒÎyQFpûÿlWËŽ-Gü‚ù‡³A¨žYYKûÀtÉÈBȺ÷òÔÌÂø!þžÈÊÈê>}F–환ìììªÌÈîpõ y; ºµ>À&!-öÔWlºÓ‘ªKžŠKœJ&ZâÍ@øpSPlerY×ä³æðôÖ(Ãe}¸]+m÷˜ÄÈt%(¸ü®±?k_·h`Š3,¶Ì&HÈñ¥ï%mŠs_bØц­Äó%I0ôVΖV¼5P—OgêÑZa ,5¨°Çæ( ˜áËÞœçôXðCÎAæÕG.cäè––hnÀ¡p‹w‘:cùQDYÝdºw`´YlZ,'ï ƒ›FÀ¶ö.)îa¶Ú˜9õ¨w˜àp0ÏFðÔ"Bá1Ù{‹½ ºY ¨sí”I6Èɘ:n,¹K ž6â#Lsí,;AŽ¥’‚"4Wí®WR7ñÍÃYj Í8s„'0ñ!ôh/5Í>ÂÕ4Tðç—¿/Û‰˜ma–‚{%¼¤ŠÇ'º\tn›ä”Å?Œ¬ƒàrd÷H ™Ö¦KêŒ-Zhc¢Ÿ+¸{i°““R·;‰¾LÏaÒ»`(IJ4)ÖCe ŠàêbÛ\.ÕŒøÅç@Dê˜GíŒM6ç¨@oï~A|šla•¡á^Ò/E¤bb¹EñÎâ Ϥ`Òç¼øÂv: +A¾WV +sž0no»ê8k‘csL]‡çÍ ð]:˜ìY/ÜweaÊðy¤]Y¨ ë ¾€X¾)ÕZQyfÍ**ëɧK\¥ñ]:˜ û=¿? +ÃRÑßf7ðpŽÂ»/£q3¿¸[Ÿ8˜Egœ-ËBdO¬6Œ­šº›¼ÌY}#<¿žu¹—^æE—CŒJÒÀ´AŠf€ +ÿÇãÚ0ôx|WøŠíQ,•‰e­™YE‹ìçªØv³­ø®ëÎq¯„¡[+a¦Äd›”úð~foÉmƒÿÁHê•pP;Òd÷K—ŽóÉ Ff>Ï1ô{¤•Vcê))‘T‚`T›ÞÞ-€wT±1š’ –0 xÆf·²ÇP±<ø1‰}÷W/?ûâ·ú·ß¼}úò»þûÏ?~ûý÷Ÿ¿{sðËÏÿø×Ûüó¯ÞÞ¾}ýüé¶Ðà[ÿÅKº}¿þéåü/ßÒúçëÿá—ßá‡úéÖn¿¿ýå¯éöÉ"ÿ´(}u?El©ú,ÒMÈÈ®é?짞‘õÔÛË/xÉ·_ãxª(7¸†ÿ ïÓí?ÌŠé^g8÷ýö¸¹dëÆ÷?ÅŒ‘övغóœiÂi~äóë‡eÆÎ[:h]L«ˆ8W£;HĆÅÓµQì¤)Q–„²šf7ïqZ’Ïðcl·{›»€„Uÿ8ñVU›\LuÎ¥3š”j‹e˜©JítwÃDcèÒ×çÜ‹ÀWø0)žõDî™@fÌLå§ÄˆÚ]ü ØÚ‚žà±YgÕç + 쌃•² 3 Œm{Mr¿“šOX<Žk.‘¦º‚+—¾¿‚ËFúó ~ p4°„B“îºàak ú”ñ»ìˆë¾Ê+À j¦Ö_›Ö ÆT(Ù´ñtk_œ`§¢À”åqîÅ˵öSG˜À…çÑz A5g' ¢D¡Hm†TĪõr0˜š„#†׉âÇ”]Ig§4¡´›o‚žÀ܉è%1]Qå­-N žéÂ+y€UrhØXßÍU½W Ýe AO¯IÍrŠÅ²æàBÃÒýœ@hdiºð®…‹lô(`µFŒƒ%Š_ÀÚë3XFo'0naê—ØA®²˜ º‚ºûú· $»PÚZô'm­a™íÆ-N­óË|Trün^0ÇÉuçöÞKôˆV§’rÐYšØ>óêBäˆÄné©]Rb£³àt2pøSԼÖ¿²7}•ØvkÎĘÕ¸ +P[ Ižb{uºÆ\Ìw@X¤¾ðâ?Ÿ²–¨ckÔ®¹êôÅéú16™Íµ¬f(z¾`¶º‚ÌØúTŠ¤5Gƒ5zXE³¦>NàGŠoô̼Äv|«I\Ñìã¦_5lm&Ø°Ö¤ókkYÄÈiB3Õ3è *ø©Ž§X(„êàœ¾åð]S„g€E^Ɉ8C)„1Ã;ê2Çdo”PïÆ%4'Ðs,ÔF‚œ¦†RkáŠËøǵå«á¬Zigð㶂‹%αŠ;O œœAl“"žà€ÛÖSÖP½¶Óòå±lÇxÚÈøD¨êÇX,¤–œšK†+¸ä…'8ÃÑ–æ+1µü.W³+Ø04CT0»djìÉR¶Ì›!ól§B,\b¡õ9Þi´òAkµ>öF°äêæ †¢*ÁLû0ü +ÁxwKµ^b8Ï×2^•µl-4òãøÅ)M^s©ÔžN[øX àÐ|ón)E;9#çU-Ô@Í5À%7W³­•·ÔÎ&.É5:{Ö%ѹ³ ¯Z€Ãʺs + BaEªœ-žXÀù¾®Ð&Bü®Öß—à°?£\æC†ð h ¬y£üxãrgZ]wŽÅöœKÆ#k7ƒòêcw<`‰m°'2Ìüó.Îø©Â‹Ý£‚Z”GS»TËã~Mvéþ‚@Á’á0 V#¸ÜPfH‚ÞÃc¦µLO¦7,jZçú#CcQ]¦q -ÆÌÄê(ÜPÚZ,´éTã¢îÔ†­¤—… ¡)zZÞ©Eå š¯Y»º÷9îâù]nÓß•¦»QɃ §`*ƒ ¸Á›ø‡?á JjÈaÓÁVM%¤^$À%¥¬é­E,”¹(Z:ÖqØ)Çò\k¥6•" ­-x˜·a'vj’Nh¾>¡ ïJ¦çØZ뛸®:šµÆ»Àè™]H}‡É‹¸íL%w~ª–ÊѨsØUç Œ³j}ö§XÜ Á¥Í¯àã{$ЮAeTeg°& *|€©Í +¹*¸ øa©}ì~Uï™LH 0IedçëÑ -‹cK>ÞÙ™¡‡ÛB€‚ $NÏ·î4pÑ®°ˆ~X'× ¹Ä—µ¹ª#`Ù´[ ü"‡ƒHªÏÖ9vÚY¨oCü5?‚kW¥CYp¸IÛîfW}qqì7 &Õýá‰+–¥µC(]Á2R?”醓úâØGÈ”µ6ø&-市áëÖ%MÞ¤›‡›«|ì‚3(5Íc™lx÷«‘½R%pÃö­`sÙÄÉŽÃ7_nÈ7HlXηwÀÿ¨0°„>¬1[m endstream endobj 13 0 obj <>stream +H‰|—ݪ^¹ †¯`ßÃwRhº°-Ë–gÒ´¤´SJ)ÃNúKr0aèÝ÷µ%yy­oSB’'Z¶¬Õƒ„ûCªÔC5•Ç/{ïí–dƒ¯/ëAIÛUVÒA¥ÎÊ!Ì|ƒùàÒªpâÞÔ ÍËøhò€ ²\V®mƒ¦¹ ßd¡ŒK’V¾¡’‹úÇ'ÌDâפ\ŠÁD)J¢ãõøI ËãßÍ©“áš8d{kóÔzp–vƒttÍK÷“<ãq=4‰áˆ ®•²A;`˜2óMv˜”\VDîWÕœ¸5›®F+Ù^•«¹ +d5Ú÷å€<ßd Öl~¨ÀáOµÇ .8¬µÃ’ÔhJõdqËü$ªÖƒr ý“Ö†{Q¾Éæ£W³•Æ©;œ¾Š—ƒXª_VŠ…*<Ј…)× F«Þd‡Ý²Ø©ÒEo1”óÃ¥p¬1n¯ëQ­4¶D£„¨‹àJóz9l9d›-µ©ª"ðƒ%ÖHY¨Õ-6<ÒEòã½ãªÖ¬ê°« þ+Í¿ïxNvQp}\í +¯J‘Ç›>ÿ1j‡i?$¡ |~½´:€šS1˜“Å“ *íÐlÃf»,ZJµ–"(¹8œ÷N8¼`OCS¨%¢<Ò`ä¹Zý4šß¤H8ÛB¯dÊ" ÐÔBÙÜ]¶´ÜŸ`nÑU7,èÊ.Ë2J‹S)%—,%²r˜«e—ÍÙewêtz‚Óáln#” ^ÎA¿s&L®„vkÁÈEæì¶íÍj ÃpNWz–E°ä€=ب†ÎÂ7p£XŽàþ¬áEå0Uæ‹wýßá]_}xùðò³¯#ýõ—ß|ÿãþñ‡ï~øáÓ÷_ ~óéïÿürÁ?ÿðåËwŸ?}|Lú~ô_ º|ßßþôò#þÊ4}û_üã·øá_@?=êãw?ÿ%=>É?ÝkE-­CwõP?Élß½YñÔ· ¾zÿtÎû—//_ 5~ÿ’Pà -(A‘Oiä€öp0¥ìW£ 0y;hª3B‘OÙ »#®0wϲ ÎÅD‘yja72/Í̃º½”²ÁWEmoÈæ¤YÜ^ ÆBYœåV]¶zÀœWs€%1ü‘®Äs…S,Ë•â !ý &ô»u@`kÑñ‡±YL]U!oÇ$(Þù÷¥“ËqÙÜCÖg-N`á5|Ø=Æ ñÑÄJrÄÛ\ÓSŒ$(3q*:N­ºÁp­7­]ÑÓ#V¹Xõ¨¨.Œ}xÉÄ®jl6¨.;gqƒj™Èhé|õzêϲB½>*¢1j‘FpèèzbO´8Ðùt¸â]Z78`d1]e%ØÆ/Áüµ“:†+ÿr¡Î6у²CI6ã2ÙrBûº¬ZºÉŽ)¥»l÷ç™ Î-`ÚžG“Õþð‰Þ@7vб²u´ñÂd‡:Œ‡—¦Ï²‰¸\…ášcaø°ñ)¹&Y‹'Êœ·b÷A)X±'¥áäs Ö²ÝñHúÌ|2û|,O6Ξ’gúë3Wof©+¥òâ×#iUë5ÏÏÇŽ9¢!·ªÞ‹BÌkíØG;»Isk»d˜¿öXg#pájŒ1¾‘©äê0*8`*´Â/‚ h žÙ®Œ‡+ÊÉìkØot›dómìm6ÇÇø|Ãd¦0…ùjsäÓã(ˆÜ¼ —ž<øµÕ¸*Qó8¯žö„ûjíëþÀH™l³#£§[éã¾mcgékÉ‚á"»AXMÞ ¶Õ±ƒÞ@úœÔ[&f(s*ÆÀröTÄÒêMÑVWV… v˜)é: 0Lцìɳ µ]2*¹ë:z]0KùÄLKÍÝaÑv‡&^œ8eu·ËÆÈÙã€ì¡áЪ ›ìhVÝ-Ë,íÑL݉D¿W +wiÉBËüºÁdž͇yT®gÙUfÛ¸Ešœ|»kpG‰rK],^PïF×´É®öÝ•k {â«Õ¾_0ó†ÿ/,;|]Ð[²oC}S­ žŠi ªÄ»ò|×SZD@#™ÐÏIÔŒ5õ­©ÑÕ®c&]Å1<0¶¤ÆÝ[X½þ*‹=ã±×±Žkop/<£Ø gLΦ-9oðÕW%t¢[ƒ‘„{©^“àËPËØ1•Z\VVÊV®¡A\5ÖAïù ´2¶ùÐ*#ŠÃ³&]aµxŒœ¶ð»Ì|É:¶4‡fÁ"Èr“EbKHšÛ zmŒ¯ž½95ÛÄkZF +:4åÇ›3ÝdÑÈ[A€õ}±N¼½Ke‹9JZŽM·Iñ°*…ÛÒ€»´­9ôBŽ¶GÔž¡ðù„ã{ŽËRtröQµ¬ß¯Œ ƒûMvÌñ6b¶•Ý€˜Úí .˜¯² +sPL3o@„P¤ñFÉzð€Ã\¶t”Q<&CEW¡“ÙçåÈ5b $óQD¼ÇWŸzYQÙ{.ðÏ1LUó áED0 –´Ã?½®c´¸Ë’rô‘a`‡¬™Q¤Þù…’?5W‹mÀÌÙãMÃÛbKì ÃÖ¤ñØ%ÛÍo–¾pÈ(ŽÝÕJ’¢àtÔ©½iÜ0`æ´f„ˆâfû 퀶:ÜEû@~:5 b©Æ†Šö®•²ñØê\;×v5\ãKî) '¦\|=‹Ñwx»žÁs¿.Åç.H}-ˆ¾5b†éŒ¦‘ï²\—Sµ7×¹ãŽètÈ¥GÀj.·A‘Ax‡vÿ˜ ô6hꨕê-¶Õff‰Ê(h£vµ×„ÔP?Þ{öiŸ»ó°¬ÁèVc®¶9HÇ™\’Æ,ûî–¾˜Ök‰L'$€7QíúµH_ù¡!ì¢Jµ ÿÎP¶e}xÌ~«ÒàÑåÉæ„aÏäcÄ&[Ì¢Z‹Ã”{Ûà«û³\eá†bÕí}w8KIpbn!ѽV`¤QésjĶK‘Õ:æ:1ZjÊaƒÂ³†cHœd…„Ì¥°m!1K­ˆ¨6,!+µÛø#Í¥Pô5tv'Ì@Cέˆå¡Uj«ÆõéaíLýá®ÝðˆH–¨ÈÃcÜBZ{E#ì0Ì)4çà]vÌWÕOEw¦DÄSÓÓœ g¢p>Ñì xB¯nlËæ —-Ë“,ö/‡ÃÉê|À¨1ÿã»\r,Mq(¼‚ÜCl $ ˜Ç8j˜¨qI=ëi¯¿± †ÊYè _~öñṉ̃ráü+ÿô{êɳϺ"Î2)팲­ô+hmÉ|ŒÅa +Ûª­ôþ@Ü!%ÞUåм¨LЦt߇yŸc.ýŠÃDÆá±^˜±§ý{ÇçªÂ¢8]5c'9ž,] Z·;uªËØIÑgÕsÉ̤) zmÛŠpÔ²xãÖ’ï`L"…lóånÖ>­>ÆÊ­Öª7‘¤7Do  G4 ¯¶J¾h0·§Ååðiµ!ß]† +PúA£JqññVŽ¡Øƒ½¯ó\Ï\ò×å!º[!õa1¶¹ʪ<ó¬Ù•õÂ>bb­¤£ë8ë#Àš°ê·íÀ°·Çc¸p³ñ'Äåç}„ƒ9 +aXÖšáRÁW*³{›[]Ç®®‹ q§îïÃd53Wû[âfäÁÌqisèR€¹bÞ±Ò$f¶Uun +H’¬Ÿþìí¢Yw,:@Ú0¦éòšúXƒ»;š?±¨‰¾$Íͧ– v8¡½ÀÁƒ“/ÐÕ^Ê+ãŸ_/f±Í¶­e™#¥ã˜œ«p8ôeÕ½ Ï©ã`*å…èÏ3H·a—ûVùœuvƒ„t"{¯©íWº~i}—-~Uûð#Ú'§YÆaúˆQ̲.´…“j}âH_ë¥2èÒΕ W覈fÑí±)Â0^l,o¥FJÈ* +;­^e¬\æÜ¡WËPÕŒ±ð«ÚÏ69"leëvÄæ!qî K¢¾-v: f41Û™B‡[qdyc» ž! êÎ`À +‹úíw@©†Üx/víö¯8ÞG@ß8¥å±ÝÝ*VM\úÅ2Œ±p ©¼_œÆÚVÓR¦Ž³-L™Ð:²;8±ødÑ?&úôB¤®Ùò.†w˜lfQS…ƒ«YxÌ7ãë;u‚Á3FB‹ø°Â%²-\ÇênѨóÈO,… +óA–²†O‡º€Èf¡X ¤»eüÿð'°ï{ÇhÄEEKî@/q™ou|€Ä+9¤‰´ff¥/Ãf8çÜín1«˜g/𰊇Ò"nsïà|Ì}¸äêXÂyyvÉÕ¶.<ÄÊ;!ƒõð=/=òÖqa¤ú´Ø>Û4¨íäbÚ¸prÓ‹K*T´{»@Ë9‡^¸~°‹þ]†ùÐLm|Àš÷+FÜJõX_U$yW¾C¸‚ê£Û…›Ï~5\lMìê×´ÄãÜ#ÞMg˜P ¹ÑZãœ[ñÑ.}ÃwuSOF'ð’SqýF£ŠÐOÐsmOì±íðpeËoé:!tù%í 1—4Ôƒµs±ðK¸û½”ÏúÄ.e°L›}"\ ã 8Æu¶ìyh‚†m5¨ú Gˆ­Ä“>µaÃãdGvÙ;ôpÏo¬t.ò¤ìÅì™l{ÕZ”‰¦ûôiy%¦A÷øÙ‘[¤ã_?»Ü@ªWd1Ò1 /¯_̆¼õó€aN' ‹YGz}>A¯uÄBGèž…Ày®„ïcý ›‚öÛäUÌÅ(ª•ê>ÀÁŽÓbW¥©Êwi»øÕ¡/˜Ê+ÛêËâßõ\k‡§õ¯Oüû`ø0Gøï^€~Šýþ¸­Ma0ËÇÄs‰Ù-ý¾Öu€xn xÍ‘o,—Ñî7¸ày¯€ÏÛ®m º³ ‰IHù͘¶g·]€øÕg:|rÖñIïí‚BÝ Ó²ZÍ+:4Z¾k°¥-¥±^S •§¶Y|*· +°X;š¯b'jôÄb/¥?:tÁ£Yv}c¹€Õå†Q5/ì + ˜w7p-–½þ ÛØ›ƒ ±âz“¹’‘©ÄŒÔ0ºt•+6¥9¬¡ÕªG&µìšaøì̳˜jš×Ë콯7uÍŒ;6-7¨¿—É¡ðGl.³Ü ñ‚+7|ƒ×¶ N²ÌtWˆl5Øf?W{2Ç8ŠŸ“ï0¯‚¾í×Ò²•Æ„ªíßc 1ÜÍE®Ÿ-ØÈî3]ã ?b¶Nb#ìü h>v‘ÑNr¢ÛXòž. +<- OlÛ®?XÛ |a·ÌK³UÝ\_Ÿ:FœEHúcÚOÕ{`>c˦Ãî0!€-@“4tJ +±5ã ›%;BÊÞ7ô#ä<ùŽ=#CT›Å •œ¸s_"º¯š¹°_¸IŸÌ*HÉÚôç ì“⑨0S"»mÌ»æO)MÏ»mM˜jħæðÆn¹xì$"…­O…pêZ3÷h5Š‹TëÅ.† +ùhï`ÝÁ·©QI+ésòb¶ìù6È­×½˜—¶Ý±»˜e£¶Ž1ʆ¾.‰@p™ä*™– :tA…ˆµ'w€KP¸ú«Á¬ÞÌà¾DÕ¾;ÖnVV¥Ü?à2#Ï+ws&€Èñ¡B¥õ˜ñ¢{ƒý0¢ŒþØ4ïbÛ¯ž1Äb/­Ú½,ãpA©¥Ê¾@Äk6S8õñ´õ,p Í°ÀÁ© ²X¨n×RP¿(Òæ£ÈlísÇU±elyéêp Œ)ÙoïÂèW‹•2Û•6ë³€¬ÉêP@¡öYŸØ-’iCø .32ÖAclU;¨«¦V{€©Ê8rËêdÕÒ½ÍÉU ŒcO;÷Ÿ·c;±RÜXñkÙXo36”Ã6F3²¥øÔÕðJ£c—"Ÿr%«Ôø‰DŽ›å‡¢tú„\ƒÜK®[ÕÂæƒ#yäÄs™’¡ÔWáÉ;ZkH +1jõ¸ŽêWèOÓ?æYëéê„+N}¨ˆŠ}ܱHN³kX5Oµ†ÂN¦Úw";ÆkPe{ðJ:‹ ôâMfØkÜXt‹OT{%~àå 7F­ÏZ› u™®î× z)Ó¤zÇâhŸ×4ôÂQóØ—xð$rD dXvBT ©‘Fòg¼Ç·í€‹4¸ƒ:=¶›½Åð)KÝå숇{«³êž½Db8Óa®,À¹®uE"ã†&w€6 éïF»ÒBš•Øì^ÏꮫŒ=9èbÖ'Ï'ZÝ’ÅΡ "@40$Ï΃]5.(}˾D+ßù÷[u#}"}«¸þF¾¢¡ßÚ\}bQ;<Ÿ)5@\ ײpÌ;‡ÃšfèÕ#z +šÆXäebñN÷¨´x˜ÓXhšŽÔm­Å¿©–&3áÒÙŠV@ ¨ø¨¯ß†G!³O‰Óð;¬ë]¥Ï"Ø_fvëÇ£$Ñ·ÿü¢¯¿e™ñ+Å'=yŠÿ>X´K»]Ͼ.Mg²^v"ñ>1`O¯8¥hV…Ò\o˜µÊôç§Þ²Ç¦éJ4œá–L´È3V¨^m=À\W7Äd».¦KÏÖwYw¥?ßnx%žh•ôz®VtD к¼Ÿþà$QW5cúñ9ÓÕ¹0}ä4ø‰í>1ˆÂkº6õ‘ô÷‹ævõºX´ë|†$îU³;YfOÑÿo۞ɋŠöð&¯e”vpá©ïZ#¨/Àµ×ýhoó‰E¥Ÿ‹ü6¢i.»‰îËÎÛY⦎WwŒÙŸØ]xx†zC¯ )/ü£A"Ì…B>šÖ oHðaŽðß½@ù)ög8üþõ#þm2À{@I¦ÞÜÈ–9ªÁÁÃàÜ“ðü?ûÕÖÛ6ÎDAþƒ_ +ì~X7’_²y²ã&Ð\P§Å÷Ðm±¥I•¢»¿~©[,9»Ñ¨m¨K„¢­0š±x83gÎ CúÈ+i†@Ûõb¹ïb£^Âè±- +ï%Ó'å©dÐ?Pò{£ÚxŒ±±Ïbb=)¨Q°Á™TîÕ'T^¢P-œº•Ø¨DF”«¡úÉ°iUíšJ²„ÑAKñwF‘IÙÒGÚ›SÇ ÓÏÞHð¡¡9>¸‘a BÌ^ÖS|/$ÌÔ%ƒ­Qé„(³šÃlò†ã÷ªÂÇÙ4Vêy\ÕFÐfé†[gÍ©æ›’afÛx¨¸¾?Ló@t0eŸŽŸùªeq˜%’¬ñ‰t2æ„ 2¿šPY`4ý²§=eî9 6¡ñ™îâM™Õ-"µ< ´ý°1Ñx`D]¨¤ØÈ4SÆèÔqÌ„ùŸ|ƒ)eBq¸iœ•T8&Õ‘6?@é¦q(âUÆT&OãêÆ5sšRËÊlö’‘Š°èX£þ –5½xUÆaÆ›kÈ°ç1„þé(¾¯þi¤×£9N–ƒ${£ºø@$w˜ûÜHN0ŒF±¢´m„khšý”1éáhÜ;ð«š +é…ÏÄÂ@ÿqtó…£ÿ;û¥†¾OÓɡ}áÞ5'„0h:¸‹’\€²Ð^fsî _ +ä:ĸœ«HO:VñÞ!,‡$3`z%vÍ9_»Ü#Ò4¯q€P‡å~ûx†—³vÛ«ŽvÛ«.¢vÛû”ÙöNÞò¶·H {zÉװ}Ïlöº®ÚvÝk×½vÝk×½vÝk×½æ®{Ý^S¾"H*¾ò4nå+€¨.+ŸÚƒ¦xƒéÜA6|Ëæ&nC`QíI{†7* :ÓAÚ½ ~Žnù ²³3¡®ƒ^Mᔤ .‘ïy±i”¶ºhi¾\zXNXkEú§ê2Lt6\,Ùe¨%8¸V²w%¶ýmØ?5jxÏÅÖ­ŸÓ¿m·—Ùí4–êT]‹S.þ~t¢M¨tv4Ç9#qBoýÂÜãQåêS$>l]Î0ƒWâó@ý(‹‚<çÌ“è'@€Aúb‰,<·P‘ÍÕi7é \åð©]ÆÐîšF,ðÁ½+K¯–`,?JÔ wœ0ù1ò*'˜Ç<ô1ž\5ÒCyb`“»X ÉsdFª&öÚ˼¡*¯ ¬ª ½9÷……'ÔuR¯¦^‰¬tÎ×.÷"éZ2@‚Hg%±šH +`=ú­†¸jžßßNà@N* $'wi f¥K N_ßz%¡DÞ!’·‚ÕorF£æR ד[Õçga9ð¦fgþÙ«2vùréa9¥~Îý…Ê,)‘×X¬pp“5AE)°‰¹x½”ôqB%¹ß>žáeçìèxþåò"Œ¨Itï ¹䶄©Õ=1ÀHv`$»2TŸ GòHlé€ÑÄÞúȃÉÊK¿Ä];&¾øŠ-9å>³Õù¦<§;IC¹JŒ2£êäêáœs:ÿÀàùeçòbè\U!K›P$Á7À],ä9ê+uûí)FÍÜÌ&ÔuÀЪ¾—™ïs­T–M|8½$îÕ-»M€¹’ÊëˆQ.\‡S¾ÚÕh5i.­-šFk-™U—ÌÀÅÖ’ÙkfÁlZÏÃëªâ=ß="±w`$«þ”€‹¬.½~»\zX֨ϗYÑN<°t‰‚éH¾¤—Œ äá ¿û˜Yð†9ˆÒp…Çüõ­JñÃg:D;0Æç’HËâóBï{B ÌÄè§=¸f«Ël…yŇ+<7uI÷¾XøT±®ÑXꃻßB>ŒboýUÆÓv|…;žå¸f ô*]g…ÙK3’ ÈîÜAŒa:Ç[’ 0¬çÚA~‚‚Üþ,Èç¥Í¡ñ\Š,¼ÆL^#·FÃhÔOmµu’?gfæŠ:|‚ ðÄ]?Cž4nÃ9¿ê“œ›ºhïsN¹¸Nè£.|–vž´gxCPp ¸Ïé/²AãÞ6U'pn^ªÒ}Kñ‘âB)ø*"gýj#GìÑ,ðª@ÆÞÚñ€ùq)øžœÐY;pËIWµ¼ à*³}øújÇ‚è#Úñ¨É#‘(4©"í¸à‚aîéó7! ƃ(íH)aå”Ù£…¨u­,`xûíÈgàé„,Ë_ûɉK…hÇ&°K‘‡gÛD’MpO%­$ˆ‘u”¼eäUö¡Kä{AlJC +®ËBd'~mv†¦¬ Ý1xêïàCWx‘½Âòó*Ý~»\zX֨ϗYÑN<𠈂¡Â8þ’îì½–ù äá ¿û˜YðŽ9ˆÒ¿’P|Ì_ߪ$opŽ¼NK‡” ·æ’HËâóBï{B ÌÄè'¾^Ó¦k*¯øx…ç¦.CéÞ Ÿ*š‚ˆàª &°ÈñÔbG±·þªê·߀Žg9®(½J×YaöÒŒd²;wc˜Î1Å–ä ëy vŸ  ·? òy`ishF<—" ¯1“×È­Ñ0Z#õSÛFíGäOüŸù_PÔá\€'îúò¤q“ÎùUŸÄàÜÔE{ŸsÊÅuBuá;°´ó¤=‚Á5x&H‘ Gð¶©:€sS¸D¾çĦÔ¿Âùǯ +yÈaÑj,)‡×X¬ppƒ5¢ðE›… +d½å$¼ÞJú8¡ç“øñ /;gGÇó/—aD…¸yrõ0#Iüßì\Äà [•k°Ã;€Þw±@’ p‡î´ ¨9÷……/rb5E‚kU ›ø\°Çîå±)®£6ÄÕ®e´_f´aËh-£Õ„ÑÀµÚ2Ú[d´BÞ¶DÃÁ´|V>kZËg¹|ö¦ZËguâ³VŸµ|öŸ]"ßóbSꋇ“·Lk`ðž´gxCPp p»dƒJkštºÛ¶ùMm3zËm߶Ío¾øŠ-9å>³‚)ßB+'®\%\ždb´‹.Åwd‹éE»ó÷Óƒf8£30Šeë.š«O¨&«Rϯù¼)´ÈÐòläälfWŸØ[;Èöá+jè« ¢hÆ#°'‘ðu.ñן¤ŽyÖ3:æ@ýktÔß3õ¬þ?S/:P¸áœqwí` ³ñ’¨©æ ]Œä¬@y¦"ðý3džæ endstream endobj 14 0 obj <>stream +H‰ì—QoÛ8Ç?A¿ƒ_ØÚµ&mq~Šë«Q IŠº×»·€&G[ŠÔQ¤ï§?ZŠ#9{³Jrý$öLÄŸf8óÿsj¸’Do1üz› ©¼"Æh¾´Š¿ ^¾žkMDД ¦An¿? ?JS·ýa69l¿û5á€þ²ýOÃJNÝw ÷Ïåj?~M„­îŸ}I2xòæ#—ä-¶…a3Xs²=º ûIÑûm)¬Æòi(¬0h²]xt¦…²šÂ¹ÈS‚Eãø‚ñ®—‰3< km$0'¶(8‘ÓŠ­/ãÀ×*5¢ÊA£<…k”£NˆÞ`Te¹*¸+ë¹áý›A—§ãƒ šmsMò”S<ÔIÕI‹3áý® {4ˆæ&ÍÀàëѧÁ0ÆBýxÆÙ†F¡ANñ §ñÔ® 2îtkáÇד–—'ª/Ûó)Ê [4Ð!üT;Ôö®¬ß']¾àm©– W°}™ˆZ>ßZz8üÎp8ƒd0y1\|›(3:Ô›jù¨™*+™#˜ª[lVì.ÔèFÝˉ>RÏ?^æ· > ²¹öŒÊÆ=dÞú–1ÑqÎFƒ3´ÐªÞü­Õ—ê ý™µ2µ†€×"BÞˆhCL{jVÓ,a°5î¢ãkj<³xZÆF‡!â†lÐyÞXŠ\ôK©‚£WzT`i–° +ØwÑñGψùVds¦Ú6®!7dƒæqƒÑ$Nªø–†èd¥äÄI˜pÉøJMÖ\ 0 l¢4‘+ôU+ó¼©ú.<:¹ã„„KngÄÌzµ‘/dáÕŒ÷¸,)œ²Z`•@ÐŒnÙÑ ÷ ¯NˆN&•D÷'¡ÔfV_G7 ×H‰Îæn‡pâÇ7Ü·ó›p÷ aó'ãGCω%Ϫj<~ÒGÞ|ìýn„.Çï<³&ÅWc½ÏÞ ™@¸?¶ÇAS52Zj²f¼t—Ÿ¶ÑkÏt†Òé~º[:=jù"ê& ¾²½¶'ïœï QwexôÆwz½ E+n Q,Jˆòi`=LkÁ0a w=ûÛÜé‰oÅ?&ƧDOãÝaß+Y¸`Ê:±Ëµ´:q*háD&^Úì'a6OWÆ­†Lù4\c±p!Ð/¥ +Ž^ê‘ÇìÖ4KXŒÙ»èZ Ä|û¥SÆF‡!â†lÐ#lžÁÎɳª}1±ãwhUM~ç™5)¾»øv&!Ú¿zìaÓ¿²`Þ  Âý±=¨‘ÑÒI`ÆK«õé®f¾‹ó,gXÜù¾Ow+´GØç jL•ƒ&Ƨ ÝQ'Doúâ® W}¤¶"H—á}@ ð~­¿…²šÂ¹ÈS‚G: `:iq>½WY® +Þ«­N47i†ÓCè…øã5gÚaS<Èi c4ˆ'² 2îtEðãëÇI+Ø|&\Üö¬VÍ\“<Å·®ïOÆ“ÄŸ{ÜÇÝsRwg×@:¾;ñ­Õ“݉êËò ]ß›EÔ +¢¯lÒOµ;ýgïÊÚ ¾tÁÍÙR/@¯`û"u|¾´ôp. hﳇ3H“ÃÅ·ù‡2£C}©–ßš©²’9‚©ºÅögÅîBMnÔ½œèãôüãõ"%LÝ\ã/!ó·ŒÁܾ®Tܳ jø°¬9Ù]âý¤è5^ +ëñ)ý“5 e5…s‘§äЖǜآàDN«²=YÄšIR€ÙZ *L_z.äu]FŸbA˜gù5@¶¡ÙàA6-^ÿ«òýàé3£/Ã<Ä t¥*Õnk’§œþÌ9Z³ÎY³Wã³_°-‰ß\m,®üêÚ´@2>¡In83)šæ.:>ÑkAãÕ -èF‰nOB©Í¬ ¾†n®‘Í]A(1n¸oƒ7áî02åOÆ †žKžUÕxü¤/Ÿãá |µzi…»ý˜tE¥ù¤t ¨rÐÄ(ü0®º,À;n%Êj +ç"O ÉC¿ÇtÒÚuMà½ÊrUðÖ®,´÷ÙÃ$ƒÉ‹áâÛüC™Ñ!“5ú۷螸mË 9<ÒªÁ±ic<“ÜpfR4Í]t§‰Rà«ïEváÑ™Ôò;P3UV2w¾©ò܇Ã…AkPîåDG=ÿx]ë’ë×è¼c±ŒÁ¬Ž®ÌJÓÔg¨·P¥"ë¼{NìÉùÛè Ë`Q`²¢¾ÃâI›]QCÖ€¿œÍ”ÌæÂpC=; æ+Êè¯\îåD'Ü^9,†Â +üòØ…·&^vOuÅî…´ÙÑð ßÑð ßÑð ßá>´58D×hâ´½¸T¼80ËwzØŽwt|GÇwt|GÇwt|GÇw¨ŽïÕøP<_IÇ]ßÙÁ¹¾¢£ëû”­»¾)ÅVšà÷ý!Z>´s( ›Áš“í¤u3)z‘—ÂzÄÙÿA[·"~æšä)§‡¢èB +UÞB$ ò2>‹›[œÈiÅÖû£’¤³=´¤±ûrƒ‚š­ã=¾^é²6¤./tYJ·º?‡6Ì®vT}cT up»?N²¾ú/ûe·œ6 DáWÉ $ƒ éEz•’†›¤t†¼À"äXƒl¹úIBŸ¾þ`Bg¼bYr<¹! }ÞÕ9GÚí’Y¿¨ß9¦7·£+‹<·áÔ"ÈU«[µ1”ÃôPö, ³„¿!‰œJÐmkôa_БP÷9èØÏ“ÎéíbÙØmÜ›J`%Þ¦}Ôº©Hs¡˜¦é݊űQ-æžôéÞcÙ"ëÓâ'Bß~jî©È”† ?tGuþ"ÖçcAÀ"žÔ8G{Kj}Cáqö’èâóK„;7?—9‡¼A_,ŒŒP»aÜʳˆ®ÑOþ²ÔèýP>Ö;ov5m-©)uý6Â’P^üSn Ô¨è,cݳJÔ˦!bÖ™öPÌãV¦J{*§ÄpsÓb§Câë2ñ} “ÐJµ›ÚŸï¹È¨E<.ô7 +îöjw ‡¼ä‰Üö7/ qiˆKØ=,¶`^j +{L‘S ÚÂ÷·¾Ç)ð4ßÃàBIèÏÀ#µÜg˜ÆêÓT¤¹PuÚ E@2¤T3ÒGq@âúS.õd‚™ø|…X·¬l‚D^w/_ëq'XÿÖvG Ï=k«™Iȼ¸ùîŸ+ÇFµ\7ï¼Ó¥RÞéˆçÞ‰­Á;;ñN[.ßmSå”rn,¾”u¶ïÝ×q¬¨þÁ”tuÇóÎ6¥5ô‰ÊZ>Ñ€²ÐÙ#€^œo B,hà¿Ã_Sêbä8ì~É1×Õ苳E$ýchF6h®OUîC:^f&=~- ×,qN–‰…fš$X>U­~fÜð Æ9᳑KËÁA¹P’èBIh/mþ i¡È{ +ÅW½÷IØ£‹Ñöï¯>ÞÁ"W¯ðº±[î{V”*¼wŠa¯…¡ÆTp!Ÿv0ŘàmwuO_”²ðÝf‘óa;åÊAê±õÎÀ(Å +ÛÐQšÇ±¢ºÏ3w +›õÐ9fºD+à?<©÷ endstream endobj 15 0 obj <>stream +H‰ìWínTÉ}¿Ãý³HñuwU†_`ÚÄ°ìn¢ ö&k­ñx¿}NuU_›, 8ÞÑLMwu}ªSoï;?,û'óçëÕbùz¸óöøh‰¯Ûòßrv<¿;üiëÞpçà­|Øyñdï)~þ¼µóp±¿^œ,g« |»wyq¶†¦WçëùY»qµšý׉ý7‹£ƒÕ|)¿Ó°óírýî³ë‹Óöìíøaû~žëɃ‹Í=¹°Ÿg°ÿ€Á÷­<œwxx6_ß}ÿÁK#?鯜‚ÅrÓœœÎW³õÉjãD\^øê…µ{r|úàè|õýùêÕ¦þ­ægçGë½ëÇ¿ºoqõlv4ßÔ±¥øŠS‹å 8¤˜‘t­æ÷NßÌ6÷Œ>Á5ºÁ¾ yr¶XÏoQkðãÆíùl}ðpþób&&mœ‘w/ÝhøHݾ&ñ9 ð“ûÄ aéñìüìl1[Š·N’’‡óÕ£ÅjãÖ}[ÊíËϤÛý¿’çõìÕæqÐ[F¢?ñ•¢ÁmìŽøþ£|=ÛÜ««wn¬©<:_î?»EÝäwWecÜðǨ³Çÿ¯³¬3ÿGig6(³/·ì,çËõ÷«Ùò ”çÕûLjɥŸ'§óÕl}²Ú¸2./ÜÈ:ðüä|µ?ÿ=²ÏÏaÖŸL@¿²Oš-ðêÓ7‹ýÍúÈ"ûŽWtƒý©µˆ³Åz~‹ZÓùêõüþÑчÏíÙu>¯¯|åï‹iK¸Dó¦i3à†_­7Ü;ç‡Ã½­oîëÝË¿,žÿøøQ»)¢(’—OO–‡Š5´lo«øÁüõbyõ‡­§§¦£ýôüâøÕÉÑÖ°fx°XÜÝrÃ}ü½øeëüÊ?7<Äßw[nÌ9ÅÁ.0èÝè]-òÍq~ ÈÚ¿øòW|ø7D¿ ax2üó_n8Í϶¶sqÄT½½^B‘·\3"hBª*bW“ÝöcÉ,qryô¬¹+Hhñb¼Kcªš¿Gb⛟ìÏ”ÚSTŽÅÎ)9ásI-w¥ÂýØÊÄ#ÎeØ}Mím¢dY*”ä?Ù|jGÓˆ’õ¢@¢{ùÕ1ÅV(ÐëRËu‘¨5Š¨¨(BafƘ4ÅãºÓÄfd„øªRô ùPÄRÔk‰Ùî§5Ã"cnñ/ˆef¹PÚ-<êi¡@"Ìc`g `Å—àÛÙ8²x¥±Ê`…0Œ‰4~ŒŸ½×ûÄÎîç„X»&F}kõe±šZPS²’ò¹&uŠ2ÂÓî§2ú@MŒTD}?Y„ZPr2T«V@‹¿*cIZz¨I\l ’X ã4!"H Põ—XËÙšôlA×h¡¬[,Ô…‚ÕzžúG¦‘œ3dd¶VAÀxŽ]«Þg|Ò—PÃ>—)„½‚´ƒÕÔÞ’Δ ÂÕy+ŒT½yP'´£2I|ÑÑ$ €Cb´†Km­®×ðîÖ#ôêo^ÊŸžÉÿHchýZƒ¬¡´~ {ÿ†ßUMª• KÒ]õ€½V>!Ý­ŠŸâƒ{ì8д£œ“IJ© ¸4Kè Ôm.¢+ÕlˆP˜‡*ØnýP­ÈH/·"\ÏMªƒ«¨h…êZ«i³£CJô!™ +¤ R;¾àƒÎx—)VzQµ2© +F̺ +´ëÈíÜ«Q‘ÑY£×¤d.leUœ³TsäØ5 ‹„K'›a3%¤8•ƒ`ÐÈU#Výd€ ˆ´ çƬ°]õEáÒú¨"£”`H°pR‹Y1[Ðì´+ufdC\ö2纰æ:…&yÖÐÖ£!!fí¥¢×{k!Ö 2¡^‰$¥)ý4ê»Äª¨‰Á†>&ou‚.p™OôæXz÷QAK,ÖR«tAmž ùJmïBÅsüý€¿:ܹ;¼ø‡0žÓ«´çShÌ{Àsmè\8׆͵As]È\ 0׆˵Árm¨\(׆ÉÓ‰ù ëÿí)¢t +SÉåé)66w`£Ie–î™8†:y¨—âwuy3ˆ9s^éóq3¤\u-\iñ ]êF•ÕGŸ9VSÀ¤ûõc»“¼ùâPú“Þàëtƹ dJ±ÆŒ0U‹±ÃÎaœg‹ZÙècßC!ô콨i’þd¬ øÁzÓL1h ™Ø[˜ßsZã,$ÅûbõÖŽw1©-`‚B¯÷¦E¥í/½¥¯LVÀ`_@˜²W! Ôvm[ ä³ŠCVó‚dªZ-ÄÐJPZˆH}ÈT-ÈhZªœÅ(rò¦´Š5ÞÎfN¦ ºb ±U Ê¥b¥ÉŸv“a?,¿:ÚÚ[ÓiËT–^ð–÷ë‰É¶ +ìbz¿uÓé°ÙR‹ôÊ)ý¾´ÅÔ3Ϻ‚AˆzIÛN+'}êBÙ]p›8RhÝضÒ%—¼Í•¬;m–ª@ø'™O®Ÿ…V+]Ÿ}íL5WS€ÒVæ ¸¾zR5Ö¾Äxg;aºL°…™°rŠv?NÉB3*mšˆ0Q'ÀÒGì~Èv0ËfÒï“Ïæ@ˆ}ð%׃«éD5™N®”ì>ÊÒ»ßV§=Û8S7½2 îl 9t©N++Ûv ÓÑä¦ ú~]òfaõÝ-B´“=å]ÖNƒ"kcNb=ÅÊM<ÃM•%R•,-}È#¯¹W– ¹ç•‰ºR™‡½0\ÖÅ +¤â•:v=Ù2NTˆoÕêaˆ;A«´À^©mï#©0aÉià‚VJ¹Ï‡l`åØ~õºŸ‚„Ðw«ê¸/½)d“‚âÝßê‡*v‰ìpŒÜW¶ÎàÐœ“õ7åO•M˜+24ŠØu« SßlM´‰“¶Ù-7A *ù”,˜ª6KeV²PX\yË h(/aßÉ + T¨*&NÎ¥N‚B£ +Â\c &¬U{AEÀ£b0Ó¨&{ +U” +u `FÌ(s˜®;ŸÙ¶À)3QàN¸³5ÖbǽÅ'´-§ì¡À×"˜¤=xv®”F”ëÌ0O +@¥ +@ødÄï™Øe2.¥,Y¬ç”oá(z¹¤S¢ª ˜¶az-Fê^Lc; {­R1/©ŠÜa¼Ý.LÁ>RDúÆ«nWΣ>Áy‡½±”HˆÌôü6\'\ÀD –²Bˆ!¶‘±eåíSÝ]=»NlÀÒ­kçë¯g¦»«jÇ&æ)p€ ‚[CrÚJ#VîÚÏ1¢Ã(¤]Ä‘‹D²½°óëݺ“ žJ¸Û!D}£pbÑ ®R¢>µC9ú@h(ãºO‰ ýÚ¹3È^ø_ ]ÐØû¨ÊîôaT¾Ùx=&¡û¬oCû˜t#µQ¯`LÈð &t†^Q›ÏÒä¤DõýrÕÁ•»I·ŒÔ`ûF¶ÁxíÖ¾FòšìÌ é>´¸\²Òaøí“í[ŒdSÇI`C¤ÄLEøлâàĪÀLí”Â^J…?ç‘™ÇÖ[KÕka3W­l‘¤*DŽ( ¦EÁ†ÊAÜÎBÔázÁ<-8r³b–*¥!–V08 +ɶpO¸Ö°8ù¶b ³¥™xÚy4+j §a`Ä5–ª©ÍDnV»†•èÎ"@H†(-6a­Ý[ú'’]dÕÎçI Û‡Þ¦f*¸v$1àìÁ‰~&…È‘" ™¥oÉc”Qk‚P +dtI=º„²B åMsÝE&tîÜZL¶&en.-˜_MÐEúBÕç<÷Þ©\GŒ7N#·§‹Y&‰âé~¤œ7%a»N°t<:$A ”ý·ÜŽÆþ­JÒã•ÒrµòQªkðª)ÐÜkã:æ¾À i˜bJ9¸§æ%FuÚ(¯‡½äæ`´¦ÀXÃ"+FjZPúÎ ¯“!´j΃©•þÉc^øj°Qz†]«=òYÍ‹°(ê½è»¤3/–}SÛ¢ƒ¿ÚìÝJ¯»§Ùä1h­ÉGÐÞ³õgmÑ·¦Dâ¸æ¹ì$¨ótU±69ŒÚC&{y†|çä©zx­Ì3ÕHÖAµJ¼m…Òð:Ó\L¨#"Ï„8”ô^r(#‡$Eù×Alu‚‹ç— ‘äæp›jª!Ú=—P”yf½žág¥Í fG6/,'uoŽTœ]…`Â×#5èÒÀÅ[4ù;‰AAlå}û|Ô¡ÙȘqx× ÒÔ‘ ¨í »hœ0–ìZ=à*B»„! jrn6³³„…°¶+A¸ ÁƒÚS2$+à· +Áp ~3½+eïs3eŒ|9WIº+®=»ö˜óQU]„¦Ê›Ü· .«Yž.yœš[–ì˜Csב=StúR‰‚ŒölÅ¥ôH+ó Ï⻂ˆXPßgöAÈR[8YÞnÓÝà3½ÑlU ®Ò+A ã^ÉÝØ{× ÐermS«ë” à5VæÎP® îK"œ‚w³f:ƒŽ!™t¯vž‡€Âœ‘ïÆ zb’œØÀT­ì#"ÀÚ:AôRÀ¯Â ÛacX㇥äìk”Îlþ^aQŠÁ&ÑŒÎMPZ‚SIe^ò©&qï5Df…› íÝÎáþuÛËÂö¢~©QÍ…ʵþˆSD'(ëÀTnƒ°êª†óàÛ4(¦.Š<;W,¼ÙÓñ|çì•J»”rîŒs +ƒEX»‹³¶6åùîÆÂV‚ š¿ÌETðmm7³$Y¥,ulší°kšƒ{Jm{ãZÙq•¨û飖+³laÄ „ƾbíBïÕ63â.ã‚M_2QßáV0<Ä&­M¨w+ e¯^·6¼vZ Á˜ûÊa7߀Á½mº(1òæ\Ÿ1]“4N +1§#&x¿J<·w%Ù‚Á¬ ¶I+Xt-šN÷ŸA6Ry0‹$ÏÅš™8F©$ש.^Ý·j0‹‚£˜§ù¸â ¤µNb¨Û°Uã‚ wˆu“ÇMiG×ЖàLpÈd\¡Ç·Cdíq ­šzÄ/•kGè!½ÆEêÛB-!Óæ5Ê0ØJüžh™¡º£îð®²ôDT›SBºW¼wãM·CaæÚâ‡[°ö•¹Ù8 uYüÁ(ð­%F™g;˜.¼Àz±×®«^ën-sÃ=a|x¸ºkg9ãB¨R6Þæ T­†]2´Áµø¤¤º'ÜÆÒT4ó Ý­n8Ê"„1—tR?•SÁÚo‡0„”kq‘hT¸¹Åï¬õðîïÞÖh•Ýépº‰‚g û +¸ç ÜRõ029€}µË;ãÆ MŒÇ³ˆ!þì8‚·ŒÂý»Wß#ÌWwzùàÃË£?ÿo|üäÅß¾½|üòÕgÏ?ÿúÕw—ß¼|ýŸï.zñâû/õR.¸ÀGŸß=(½´"¥btpáoËÛÿjü‚¹ŽI3?ÅÐùPŠ|"eô»2~~ZÊ݃ÇáUŸ<ýæÕÓÏ¿~ùÃåW€.}ùêåÓçÿ¸D2=…oŸ<¶|>¼ü+~ÏMòy=zr7 ”ËgwÅÜݸ k:êf¤ â‹™8½ü‹,‹áXcäš|xF´U—œà*j°…é«î{ú‚eë'ˆ?’€½|sG2Ûþü2Í<”5îE°¦bö~œ¥ÇóvÙß°÷@‹6²Q~8^a›Aù ‚‹åÛœä‚홀õˆáºÛd€åJÜ6 ™h³§ÍŠsÑk€¥S¾á§6‡¯Å5n =?Ç&ˆÏk¨ GE+[)æ²a›¬ëùÍAÐø9ž7‰Õ|mƒÖë3×–¸d˜Ôì!ŸÀ‰p›'¡†ò¸á}iB¡ƒ³\TîLý²!N„wXhœŒÍ‘v4ç4njÁåG>¡§®×ÏzäÞÛ¥e ø%kŸrùÂþþ—òü½ÿŽ_ßàKø7«>ˆ|Å3þ`ã®Ôèô10¾ç°X]8ŠÄnÓ~øŸ¯ðÑÓ&ÿÛþàóè ò+ÞßÞK?à?¿Ã—zsé—?\þò×ryb+¿¸{OJïIè=éü/|í/ý?zjϺ`*ùº·à>i   Ñ2¯ðÛAîc¼Zzf§gtÁd#l~ÂXL†™7Ñw#¼›WÝO"côý´×=¼æ±'CL¼äš‡næ1Ã}k Kâ'h/pTº/ƒëЛ°K“̬žÌÞN 3kfXB5o»™zäm—"t5?[|‚ dˆ_YÌà•é­Nž6xŽÁ]ïhXAô™ º4°Ïž¼‡ € –qS„,ö’ ë!Čǩ˜9ÂiŽ0âîƒG?µ Á®XùºuÐU4c˜;‹O±„µ1,µ”؉ɹúJùl0&‹üü]P}yÏ <ÂmqÏ{®|¶òÊ^+È:…ç2¯µ# kR†í§þúÈšÒvS"-³=l2¶íŒËtM"••!t&˜ëŠÔµÏ–çݯÍÙU²öʼ M³j䜗=OKå„…ÀÍdg¿v÷КI„ë‹“ÉI€ÂÊÆòßY´r“p<œ†uWÆ=œS´ÿså(óær”ûuJP³Dw+@«êM®Y4¹v-VG_9FÎÐBÍŽ¬d-[Îó8Ù,šÐÈÖMW­I×`q[ø2NCþt.°GNCÞwÃîJC1;ˆ¢Í§E®]>ÒËÚ ÝŸ™²˜V]-š©Û×F°I‚p¢ ˜IŒ°Ý6»)E^ðC$¼ÚvÄ*Žàì?œKXŠ`¦»ç–<9†<ä‚ XBÈ"1 NX;rlYü}jÎtõœÝÅÄ’‘é­ÓÓÓ]]娵4ñ`s„01(®¸ÎM㉔YN6bˆ7ùÇê܆<Ù ;E_¯<*Åa!N’‰'aÙá—åºß:Yc2צ![áêë BJš§P§:l^/㱯ۇ딽¼n‰tGRe}w %òYð&rI–ï¢òÏõY†´²x’¹9^×V#Ì>Ëý1di¥c/5ߨºxÌwK_æ­†±Ø]Û#éRÖ> IU‡q_Ô,Žz¶çF¼ÞI«_a,‘{òI_¦PŸ%ùê97w©wÁ\{sç \Þ[=›'Ða6·`ñ¯qß ¦‹#¦!ÕÃÄ9j¯}g•Šz]zi\ƒ )2YЬ¹XÌ–Öeσ• +rìeÏ8<\¯°MyzäNˆ¾ÞvBø0!XÔ¯5à$a3½[Ä:'ÏŠÅï÷^X2%‚uq.¦®QLQ˜6ƆYƒfÃ{¶uoN{òçµÔø}î][ÔƒÉXmé…A-ÆTË%ÌÊæi×XÙîB÷-r +~Vemй¬ÍÚj{œ,¶ ŽâçKØ‹+Ý &ÍwÈ|†â'Aµ»â`?«)ô–$\y)¯B"XdDÛe«îMœjÊpæ¸Fªšgï–béî“Ò"_¨Œdî[æ@Õ)ksÇ(œˆÀ5mžo’¦7%×Ä‘"YÙ2;Î9ÚqŠ2ÔÌ|s%;¥ßªC©Ô_ÕŒ9LÿEí"þ1ɺy¥IŒÏäJKó늖>'­m€ŽetáêØ3Ëû +•^8Ø!oq€eƒ#( D° ‡åPJiL¿_S÷v@5qAœÄ±µ<·¨ƒ°PÊ>í§†ZÁž§µµ­æ{ Üíö†¹ŒBØüŽ½T•°µÔ‹ +Aûa D$ö†Ê”`ÂØÊÊg›Ñ”G›.S…`éÕW/§ò vwÊÝóª`OÏkÖ¹:€îp¬™s¥F¸nÚš•»¾&¿öcpÊøÔP“%rœ½d!šõéË¿¿}ùúÕwo~:~ÐñÉWoß¼|õÏc%óüLáûÏg>_áÄoð÷’|B^Ï^ÜT¼ÃOçßt|9ÿ ù€K“îp¼6¦--Áö‡óçøé{üÃŽ®Å +Á6YæÖâÕe7ÓÇXqýøÑxŸîí¹â_áï×øË Óñä?ó?øûì=ÒKg%ϬÂÿüÿøBï;þt|óm:^Ì“_Þ| £äóóÙüß{w~óó›_*Å¿ùû²z]¤p¼·^36ã¶=ôNÓðõKø.Ænã†å°~p2O€÷Âð k[LkÇ!Iºm`+~í–hµî}íi¤ÑÓ2EøÁ@˜FÏkO“fáOÅxvN8ƒÌ¬‰´ã¾i¸ôpmÔHL˜í½žYÀ#ô ðm„ù¨M“<8½9‡KM0µÃm„™‹n¢CƒùZuo4Å/—“nzñ¬OÍIA®Qf¤#;¹”yš‚ë@ a6 ºz°Ìm~§/1ùŒF ½ïE.[„MêNÞ §PÛÖ;“YÒ¡åµßpvhØÁ>¼8Ø[ÛuáJPçš .,ÃZyEx°º5Éx¤ÂG*£.¬ðl{îÖÕƒµÊAšOÑWŒue³¶ª—4aOiÌ™éñtJØ\ý[ÍŒ­9 +½k¸¦fúd›Ø4=9À!¦ÁL!&¶%ìدßËÓÕíÜ,Í0~+³º¹f~lä(.–¿_·ïW±8ªŒû¼ï–F¸Ë¥àÌÊzg²¶Û¦÷Â]+8R-\Ð8üZQwgPS7À31ÙJÖУö`à=ÜÔ+sª KaçC÷‰á¶@äcà‰nT5ÇÚˆ¨jS°PÚÔÌ.»<Úcw0g×ÌT6‡‡éû n#ÛÛ4Z«0è)ð„ݯµê.»«Ýa™,PR•ØÐy4…O AyiQfldXç ´\‡â:·÷0Â2gOÁýir‹ËHÕV–Pð{®°4*c0So2°Æ>=…¹«x¬Ìy>½‰Ûź¤Ð’»#Ø2uyY]»Ù’ÖT.YEÄ¥£ÀZy¸5N1NsXx¬e¦9, :™9ìW[‡®÷º `ÆlÈ­ÌÇ©[˜ezÑÄ0X,i©“€a¾^æ^6Cg,0ŒM0“0xB 4á£ø˜e¿:ÜœQ¹j–k q§Ë’V¦hãâo5»)à'+ùÌ™bWÉ8Kfðb¤Py²]ÑäLmÊFa‡ú*7§8vŠÏK9§„ý†æq³`Ê3ß”Q">‚¨M®(õiud@if«'‚Ãeg–h÷2t~`H>Ød ¤Ð µÇe´Ðrºa©Ü Ýá€F‘!0êhÊr^îqcÌÓ涆qI¢Û¹Æ~Aš0êéÙ›ë³Ë‹×W·/Ú}w}uvñó¶'sÒSx{zbùm_芯ôµ$Oš×ñéAÀ/TŠÙ´©‰TûDªî¿ØAqÕ +ÞËÚzI/ð|À)îLÝûG?wªíçY­ÝÿÝ—-²ýÍòîöÀÀAý6oîîWÁÊ!a¥vå› û§çÁ-œü\¿}<ØÁ.Ü=Ñn?v°ûàöNEZ˜°tS¡` <ϸO,÷nÿÖÉáɾhûÖþÿ‡ükÿ^=·VÚ}ÿ;ç‰/>{¢ûÂ?u¦ñGNu_ûGÏÕ#ÿï'{xò^_¯ô…²¥íñ¿ìŸ¾Ž?ôšµöê¥üQ?üEßüS¡›lÛþþÚNmå·ÿ¯Ü{Ï÷BOðæ¿è|Î»Ñ Týü8…˜†_0¸ Ñ NøvA×öE¥‘ž”5ú66ýüž‡–4a¦qr‚ûÐ ¥†#Ú§ŒOŸöd¦‘Ò.1Ï4ôR‘FYÒ¸½»)Sbä‡  +‚¤áÓï.blc+)µCÍ6ãfßàRF"iŸ5ë²Ú15=  ´€ß—a:,xZƒ¸*¬î¬auœhØç »‹Åá–Õ±Å{2.‚[É"ÈNe±,!ÖŒ}dñ”Õÿ"®6_ ‘Èr³³Kž°„„: +y5ñâا´~Ղµéñ * ¢WûÉzÜ–BïwÂ5µ{Ê–j•»…¡hñˆqI9T¯—¥ŒúL»Ãæï³,«)!mó¡ãDõ9à’ Š+ƒmß9úé«ëöÓ/X›m<š=?ê[šçÛŠ¸Sö£ u²Xg¶’Ð}ÉÉ‹> ŸbkÈKôQ –Õˆ1×¥šÏÝÑœg©Ž¥A¥z€“e;‹"@Žè°àw¯(âB“6"$A߆B›JÂï3Ï°ÚkX›ü&i€KEf"c-—ݵè´Dö¹æåjUƒ#…½(KA1gè`ö^ £™m\À᪩a«ËB)AÜ2<ØìBU#Æ’[š–¡­*œN©\Õ;˜‚Ú?ñÒ¯£Èy_©¤AÜC…@HØ)ÍÖ©b¯-AE%àÊ퇲:ùcFüŒjÁ¯u÷W§0”aõâª6?¢ºkºª 6c áa’=ª“˜xßì +i÷ãQ»,#j[Ûn–¡„!̸%N‘mw‘«[p†9P_²È8 iQˆÈ ×éá"¤§ˆ€‰’àY"¡_àIÄy1 þÛV”£,g3á†]MT/ƒØ©-pö£ž±s,;q¯©6ä‘sªÆóâ|JCÜà¥VŒmp¦éŠØx˜°pà%ÈPç\qû¦ÓÐñ¶®ÈBa¬5Eq3ÐnqAKø¤\³*lƒEN~µÕIJÁ\\AÙLÚ 8°ÓH¸Ú WÔ‚ €I›»­Æp/Ä35=ÌÁuÄCóÎÅûÚÚ¢ƒ $Ýåb´ˆê ›F’3¢ÊÈ׸Ç7!¶¹Óç´v#jtàa3„Pš9}\”äæ®L—à$õjÁ÷É=˜Â9`â¬Ëá”ê:kcŒ¯ÏSÔpA• ¾ äyuð”=n¹ãêRMQØ`‡—|[ŠnôfVvF¾”>©³[3Á¾i³"^¬sÔ+­AÎ ®‘ÝÓ'æ²ôE‚HJ™}Øu^fÏr=O ý„pN|ºåÙ\16èjä'ÐUÄ䔪.9w¸Ö~~ÿj„¹%ûVfçŸq.·W[˜Wm{p´ÿð™wŸ^þøv{|uýââåëëwÛ×W7ïßmß_^þz´ñF›ÒÏñ˃CRÏKLúW)~C·ÿxÿFe€rÊÏ”;[~(< * ‡Êq-?#:8GÛºâ+}-É“æu|zÿ ð»ÑOnî¼h{aÿµaU‡é!‰ö=dë8ýd6ûý^¯¢Û+dõÛzWúÅùøBôã^¢,Ò¹´´¦äÓyŒy×MûmrÒâÓwozˆRÕi¤^ýV7¥w¥.WG VB!¨§ëL¦ƒ ¸É\kŒ%=ì¿7J+­/µI® =f•42P¢nþóª4ÔZëó™Eˆ¾šSb¯T6¢[!vMáЫmŒÔùÖq iZÃÃníÑy¦ú˜#«eÿ‚©RÚJhüûÕ²ªÛQ„_e?Á¦ëÒ—Êztq&„€³·÷«[¯uHŽ/`Ivªû¯Õ]]õ]2i$ûdÑAæ[|m§,Á‚»†ŒM¨1àìä +¦$2ø/Œxÿž€˜é€;’/y4±A'7ÑbN)ƒÛýäs‹]`ƒ’íyj÷ñö +¦ uïACRçЧ¯~§Ï¾bá_Ѥâ=Éþ¯¾îÆ\ŸhÓ@ï‘®Ó Ÿ¢ð…± +ë©¿w`9ˉ¤‰óh¢¹º‡`d=f1Æ‹hŠ9T ×ÿÈP´ÇÏ|’H;HaxÔôsÇ‹@ù=ƒ¢í[?ãÖp ìÁ$X5ü¯Rìž@æiU;¨0ò Boe=ArPñ™eÜ Ûypĺ?ûrûi\êz‘תvÇÜKª-úd×U•aA†k3)X*ΰªVQœÝAèN×±ò-:…~ž• õ¿s˜—_5ÑݵðȨ%§÷q²Þ§¨îöÝð9TŠ¢å£&y]Zs°_8³ 9jrJáy°eß®7ÂùðW~k¥êŠÙ-á–QiÎK êæ]c0ª QxïaЯì7J¼×n Ë’m^CKQPí3òûoÐû‡Œ³5ÐÛg ¯Ìôƹÿx‡$LUN›ãq +!Ôþ†¶Ý:áw2dטægˆjrÆ5V¾˜ë ë̦¡5ó0v»æ$@ÍRhõœ¸ÝX˜¸þŽè1O;R¤;Û­é™.¥»¬hR7$Žvè MÆÁíÖ✞nLŠ™p¦Ôì>.N!±€ßÙÌI™@`×Óþ0P:R uZ.Sæ«G—+¹°¦u-<ó:=Zƒ’Šuã5î|¦Ê—™çÖjñaikObmNÉ9Zðp›?Ü98 ¢vô~€¿]Ó·©%§mÛkJ`”$ý³&bîDYÏKT0¢Óî訽*ÉëÖ¡w£ÇÏ´V­Ef®ë¨M×ÚÏ{³çi4Êq?>1óКãb*ÄÃ~õ¶É_ðÏ_ñ]é3>~÷ŸüOüûo?Ý!r-Jéçøß?àÏ#øÓ‡~üéãïÿÿÌ_üðÛ0ý6LÿÃôç︆ïóN‹3°ÙØ÷Ãrµ®ˆSWÜ™øë]˜:µEˆ½¾ÍÔ²°r$QR +õŸ‘™¬˜Â'›Ø¨l´Ö+N(sR®Ñ,ÞÓ*‰pŽä/?ùå›ó(SÝn`T^ç¡û‘š¡¬r¥,ùPœËêÜÎÕÇØ}ò4H±K>âƒB]¿zŠçˆÞ*«4¦öªŽè L\ tàW~ñ¼€K ¢S=Û~|8OÍéÂþëË.…Âæÿ´1®®Æ· Þ”AñAýrý +/Ú¹ª¼ªþŠVsP# +$;špÊmB6[•ð׉!üeTå¦,º‰ÍOEµ˪$6N=U?`ÞX§ÚX&þúrÕ\ëùÅæ€ÊÈZ¶ IOH‚'Á.Wo˜‰*ë€;Œ ®yv÷Õy28Ì®î IKˆ 2wãk–ÌwÒêà˜rSà‘)P “5°˜>ŽTsóXTÂVwÚ„‘ÐÛ)\ñr¥ ½y«¹i“µ6ÞV)0zGšŒ’mZ)åÌYÃ7XA“5o†yŸ`v‚¥<¸¸e·#PeÐ]·{£'Ó®kèlZ]£Ë6­²¢Ë™+«¯›ÍJZÂÂ5aàv“©r;ߎ +btúÒS·vŠe77&4B¤¸›×¸þcR'ðW¬Ós=FíW}ŽÆNlBã‰zÝoÕÆU4ãÕmÇÑV=RK ¼óînºû…¹Ó:Û>­2vš#¤ðYH‰rdôã;Qe²¢ú"çN4” ¶Šüêy¢É\°Z1xÈËk7Ïì{ŠʘMy»‘WÛåÙùh¤ÜEÃ>ã_þ‚æÂX\?˜SÚ@¶nˆ©paa*¶P‚¯‰<Î'YQù$úÂÙ€ÝXX îbë­ÔÇe–8£µK’*Lo§ÞŽº")LŽ,+ ƒÁñRAB-Ä•g§@Wñ©…5ÆjÙ¥!J\7OÓ +š%^Ê?où±]”M>sybRN¿,ÍÔ3ÐÁç¡_XV êåèÞÇ´ z9-¡€²sM¡W8ms‰š,hh½7„ø†—ŠÝÐR~Õ §FS9à ñ¼½ïq«ÉU@žqH[©€ÌQ5ƒ…­èËÏM³vy!ãvR1_ð¿Š«flšAÒ=›ùô)Ä΋ÀvÝÏnå–'ƒÍ«þW¸ëò‡ p3‡ñ‰T/‡PDö¾p ZFÏ «Ž¡é7Ö¼ØåöVT¤5îfð,QPòr l_)ˆ~Bn­µÚã ’Ó+¢‹6gä5¤š…ýhÂRC¡½ü"_ïQT¼ƒì¼ èP{ÄÙé"íì7TŃÛYŽ +„ £v9ÞüY(;4üM!‚*ͼŒ¿J{»Û^þXk§3\R9“§ôîrSS»g“ +*)\˜ù¢~‰OѼ¤‡;Åœãt^ê tGñfˆ^)”SaùN.“ •_DŠ®‚ÛJ–žužbùXï©Íh—±ªiïjÖÁµ^ TµÔñLˆ 7›¹µ]Ôv™•¿×—LtiEý¡:2(³=—Mî n™ŒoS€.«”Ã;çqœcHÿªœhâ=ëEÈ5nßYª”A§ÏýÀ)mÙÑ"x–õ£f×:Tá7S"£–,›)O1µ•‚<8Ý´*‘5j,£ònacŸØ"µàòàI½·VüJusT“R&"B¹ {Ág®Hv9@¸»SNå`9­¡ ÁÙ¡`>¥C\µ!b‘äl»¶ÉÒ­û$&5gð£ËÝÍ¥±G=ï+íBaÙåű^à…³ÍfQ¾½A hÎoîÔõ|00äIm¶fÁ°<©N+©0FÙSøb4hVl]‚ì Ûá¦9>=o­UVÎn.=1Ö‹#ì‚¥AŽ +» :õ$`I® S Œº]'Ã`: ÂI¹Üõª.´WÅ Ìfûù6Cö§›ÙÄC|ï€FËsÐì ›ßCÃyÈE"?òÞš Ä|­.ˆƒXR[p¼µ#m‹ãÝ¡ +‡XÑ#·Dˆk æôW0}O¼“RÅÆ8çõNcÎœ»öh¡oÕ4Ù!›…,4‚PjÏ;aA帺mCDÁK·ò\Ó2^癧a+6s¶iŽ¯ŒP72ÖÛwƒ ÙPú/¤„`.¥tûõ.È +ˆ…‹ùF² +î¡TA”¨r¿e‚ë€CByÕn1ï˜ÒâNœ;Ÿ¢/C_Ä‚nÎS 1ïT ºh¦ ³³å•†k¡õ×cÚ1~8UÀs"¹@’A­ æ,Sú‰ÿËzµõêuÔпò½ 5¢3ÏÅðÒò¡­Z‘ªJRh IPHõßwù6{¾¦ªD¥“žcÏöööØËkEXÆGíº¥u—˜ƒÊ…÷Ø7…#ë†OIPÍF>®j+ŒÔ¼9l ÕÈĆáb«è…ê4¶BÁ·k1.À W©/Tj¬U\Á‚U–%—ÖéñšbJd÷†réÖ{8Df=P<ñNÁºN;°CØž™ÈÀÒ²ÌÇ༾”²ïq£@zh½Ÿ}º6‰Š¯ærFOb©†Q•ît+4C›Kì4m¥ÆRW›ä¸¬z up{ŒB‹Ó´üz1Ãà9NfKCîŽóýé§'º{9…W”5Ž¦jjh ÷S%fHÑY±LÉTÒÄPˆõ´BÝJñÕ#ê:w 9š}†:¸­­áHi@à9˜A)(s!ˆõ£JRº#?ŸÒ$fÉMÀ¡‚Ÿ‚´Èt#ÈäÄÆØ%RQ.ô1w •j¬è!F‘õôßK‚­ê>TI3*|Íy¤‡žîОر±O—‹%âͺcàL—–Æ‘ÆN#bìfls9}[Ne²¬X|59Ž]tœîγäŠOXÆ7jq"ž±qFtá¨CÛÿñ¶û2T»,ïH¼¯,]ô¬ Y5‚†±ƒ¸ínÐ lbGà±yºX©ë¦ñvt:Œ{°&z¶XM±‘VŸÍ4R¿d ±ÆãÒý›Yf=;rÎ(†m–ÇÛŸ(¶¬ÊÑ츠à> o4#ø[ÀãµíÕQ‹U#éÚãíh}J2¥qØ’ú¶ÐvÇd;+ËDÇt9ýYl~¹h¨Êéãa»O“ùG@jqh,Vªñârt¹ƒ/…£%Ïb—?fŒ4aäÙn?ûz±ŠG—¯X/Er'UÿÔ#ÜŸñ#·ÜžüågýõÁÃg¯þþõíáë7Ÿ¾üì«7ßÜ~ûúí¿¹ýéÕ«oÜê­ÜpÙO>ÃÁ_.­P©X÷wåþ¿êl ÖøP#ãúˆJg}~@VŒKÁïâu=úæù«—_½þîök˜n~ñæõó—ÿ¼yB_Z_?ûRszpû%Nü?ï}HA†Ožá—±Í¡YEââ¤â~^lÍÔjOÁÙæL}T +yµÒ’'Žµô 7Ÿn ´IQ1ƒº's0QHâ!D»8¨4 ÑÓÍVûhÝ7Æ“BBº/¢å›Oi¯Ü‘½âµOƒN ¡ ÅÎçêûWæ0Še>rS7ʺv JoæÀs²‡–»1?¬­ÙVÂו‡ÒèHüé°Z.›ªÖºÄn¡:r4ÇV Fè%¥!ù¸î•›¢N!&hÒŠ…°f 6vä‹îxpJ@ü +cYªusP¬ÔÓ(Sv0HÆØÔaä"ñOjÆ­5a®'ÌMa9*Ic×!O‹n{ñ5 1—dü9úÄViÞg…€Y50ÐjÔYl¹:Ãĵ™ÔÙ½­ˆò ~*fçå«—7¨“…rûTÿý•Öú†%Äh‰‚ÆA2ø«`©þ{\0êÜF9ü‹moÕÛ¬§(šhZܧVZ’µ.²@eýÚø0evÔhߨö…ÞCò¾ç†RÁè•Jã>”f*€É¡ÀîÛ-·ØÛø)·ÏÿGãþÁüð¾Û°<§#;û‰haR.Qª\ïإ߆ƒÆDÇÈ©ã‹÷6d¹=üÿÏþ}ònç¨ëÑÐwöçïñë¿`|wãÛoý[¹=ó'>ÿÿ¦øÉQ¨·?gÊ®rÒ–Ieîî<’’™c:˜c +‘è鸕WiOfwM¾ÞCØáP”8Âiô7îÞ³Z8V³~ýA|ð‹÷úÒG÷­a^uf$+2¦ŽÔÖÈÉo0ó0êëÝØ1©vP +9B‡±4Ê ëÎð>‰3Cðs‡€u¹z£©ð‹ÞK‘]™>ðø.-|'±TÇÌd'o*­ŽÑ{Ôa9¨[0“rjä!ó¸ªG;°1c_šAÄ%Š1€àÁ˜dõÝY£‡qRv•36ÜãJFßRN&<ã¥ó°Ï" î|`7TçÌ%eYi£ÝÐv¥Â’W{ÿFȶì1â£Æ²Ú™†÷‘> \éj¥5g† y žÍüæj¥ëˆGhÞ]Ê=ûNÚ]—µÌú¸xU¤meìH[UM™FFó\¡zDæÑòøiÊ>q"çÆ™E»6"Q9¡#4Ôj&=øD…$Q5Çä7¡‚Z¢ù£[i±§î®–=ä‹w1x×TÿŽ“`µw×%ñåÁÍ(Ù«• !UäÈ÷j¦Šã´Ëm°™ð³!ýܳϥ,:" ÊÙL.wuâ.™XjÍi®2ããðK?÷}$‰¹í‰¨Ñ (HñuâÂ׌hè|žèD„>òt]G`å-¡Â 3, +ë¯-¥¥±“§†ñZëê:ÖÏžîÀÂóì:–ˆ;ºë*'q(èï=øŸm]0®DYké(,F¨eì/ºmóäºB+u6¯‡©ïÐ*0™"™`ô0Ú@™q­ørp‚Â3“ 9bfCš˜pc¥ÔŸ#À´»Îq­!]Î_ï3J~}xßiKi-‹Û’À[ù'Ñ‘òu/€¶xéÒ’¸ÑU¥J!â€Ü7{shÿU>ºFÈ9GÓ2fÍØÓÍq/öÓšìƆ¡9;z9~5gþ1)…"nÎ3*E\k™k®(zŽ+Di#>Dœ€î,n4Ü9à+¹Ù\Ý4¾³FLf$qwS` •#‰Å|n|V¼37lͱD5W.‰ÒKD^c¹•Þ¢B«Ïk¡¤šIr2 +Ï ®•‹¶Éfå\5¥8Ç€c܇ž-Ÿº2ÌòÛÖ01l0öŸ'Ð<;6Ìí¾¤?6\“è|DX«gE]aŽ`øÀìƲK4EÎÔ(YÉÝ3ÅŒ0ÊŠ!¾#W£ä‚19UDn4M†7\0ŽæÆUKNõœ§ŽXsǾ”ȵ£ŽÒ-Îí€Ý6Z†}ÜIþ×ÆsD´7OŒ’R¼^Næ3Ëí{Ö«nW«Ûˆ¾Ê¾‰R<öxl·WÈE›&(? +RUE¤B œŠ€PÞ¾Ë{ÏòžÍiZEê‘>8Ção<ϬY‹›[9k7ãìPÌIŽÀ’t΀¹+º™ws7Ð%Ö°Út0ôÁÉmRÈ,ù}À)§€8Kà¸Sdí¾D“KòøÕÙÔ„,©¼¨ñ¢ÀLOÔQ®‹v¯¿`ž®‹ Ï€æÑ¢Áu 6zâ»kI¼dikqŠ¤óP¼¸|ä¢~*"ž'1râ*œb2©èÉ!ñ¸YC¶a0^4ÕfŸ?¬³ÚeVúß Së‰A¾(!`g<p­S78¯“C:»Î²˜££SvZ<,E¾|(Øz â$]ùPgÇbßpúêzÌäÖÐ99xn¹:]îÃYwÃ!Éw—R饹gC=Íi„èTY>ßÒÀž“³î}¾ñâê;s·˜Ó¦â»³i|®Vb¹ð­îª4ÁÕ..0ÇÒYE=K;hŸGf£,HMJð¼Ô/ŠåjLv"§¯Ó£»°†œ„‡ÖØM£òžàNR:¥g«lá¶Â»ô ]üÒÔT§o > Ý(~¬S8JZš±R÷® Š¨ yD¸C%è"$U‘cÜ£ä…GG¾±Xûyš?ÍÎP¨Ù¡Sæ›e†-míÖ¾û +¢ÅŒì :ÖcWîn˜Ê/Ì<ÈžÂûL¡Ë0Ž:üiMÁ–†Ÿc³^Ÿ¿€µP¢·öGgS’ˆ³®RP,¯OÕ~ñkß »ïðÛ»Û“ïÓ_wî?»ùñùvÿí»¯Þ<~úîÅöÙÛ÷?¿Ø¾½¹yuw“-mà +Ocã'€‰’PÑ ?`쟧ë 'ä‹=2¼=/~˜SÕù}TÕ°G)á÷{8îáË¿½{yóæéÛ_¶ßci»÷Í»·/ßü};úaãù³fLw·ßaÇð¹u‘„Ÿ<Ã/¶–]…4J‡/2±†¢aSî“òÕ©\5{±åº˜†/ñDôææ͆Í~À{ÿ¤íëù/ŽB¤‰ÕÈÊÀŽSÿ´Ûaý°ªaŸ”û)mB#ãÂÜ(ºCsúS»¯¸àÞ(tëßÜzÅ´Ýÿ×ñßþï“+Äù„{’~Ùÿü#~ý?lºýyûË_ÓöìøÆ×ÿ׿ i:âøê¤êŸ§·|8J½d -Ñò0Ô¨)8!öz¿®ž®×0ã¥á˜^i˜Žƒ;mn‘Á`ìðT÷ô¤´D•úÑ™.õt¨IòrÔå 1Såj£àÍÊݬ¸È[Î-HX€Z¦ïÂ`†­3£þ(ˆ!fo±Ü¿@ËAÊÝúÆHÉRP6“B½FV"’£4ýÕ\^NÅÔkØëúªW¿ÅNËOSÖ °MÂoP`ú..Æ:'‘ùbÃ1Æšy ѧžEïxòkd§sN%/è1†Ö;ÃÄ Ø•c°c÷pB‚‘Ò‡§ +ˆå+®ç.lÎÅÓw-¬@4/UP ¾h®ìƒ¥òñêèÁ7ž²rwÁ÷N.ŽJ5Ë'YZµÇµ^XÒÍÊ¥P5³Ö§Ž8{¬¯'xhºóóš*KxTÂD/A|NSÒ'~…‘Ú“»Ãåe¡€%6Ýè i0-¢ë> +]ø’I¿1 ŽiÊó„Ù8$$È/ÞÏZÈY×æÂ5e'ôXô0ôàŒM{gÐzÔP¾ÌD¿cq¤µ¨AHNÆÇ3kq™ Š™×Mð,ÛB‚B´)Zò?@„Zñ,í„Ê8³3 *dßyúVI¾»v ®I‡ge/$¶Îˇ炜 Óoy.G'{šzñÝàk¥Åa]ÌF”Ȳ¥¬žS±H…¡óØè‹õ‚zÑM[¾1‡é"G×û|>³tº®å¨Mš{骶êZ¡AS ;ƒFë%(…(˜‚AË™m›¢É³­RÕå¿4ײnV ]ˆF•¶F¼ÎIîòJÙF˜…**-КñÁä4|æD)0š}rKêê‹cò žøb¶ºn¢–‹!]«a!Ì~Ú× + 1–Èä˜Ü „ p`It}oÇb•nÁƒ”âÇ¡SÙq#‰ëíXìÉ2]yC~`µ§u+Yéf5ÙÐhlÃ&ÉwÖa-ø†]›6O´öJL€¸õKöÚ¹hƒ7Gÿëŵ­@¢çÑÛ±¨e©ÉÍrŽÅ¡Çnr‰Pߤû¡#§åÛ”¾­ÿ‰_Xœ‰ +ýâ4}§XÔ½…‘7ËÞÝ´¼®ž™À•Wª»„vN¥[ qÃ醬 kƒuÚAd¹XvË=b fŸÒµž¸Â&êVéW«aç°ÁÅ<0×r š?CY^"S%¸•Ôy¿‘#Žåªì¤² ° V´pQÙ[{wž.ÊH¬<©1w–Y’;dECº µÈÝ$·åêØ9ÉÚyh«M‹“&à}ccÔÑ/˜ù F–\ŒÄé`Òr̵<9Œör>ÖT9÷!Vt (‚Â. +÷áiRÃ_êqt)ÏI÷Yx=>³_ÚßE Á<|yk_d‰çŽÂœúðŽÄ¶À¸õÊõøs*¬gE ëoѺ}d‘iÿ]ííÓÚ€Á?ØM$KU#°Ú~›bþvcµJå×êÞ€õ×Rä4ºN¯vûÝF¿ÙÚHêßy°X! hµñ CM8™ÕI›€ k¤Œ“Æ"7ËDŸ»B&+FyV`À,󬌥;Ü®!ì;Avf9ÃjáQA(}cé¾Ï]MÄÍDîÈç‹÷äó›®·ÄÐÿð1·Û.j=_ý=þõÏ—+û­Ð{)=öŸ_Û×ßMøpðñÝñÓÏá¸O¼ø˜ùÈüýÌë»Oþ?[D³›yd]Þå5fJP„55EWla{ÅnÊ5ž¦¬aºÍe +ÞÓ6Œo¯:P²WÄ03˸3»ö HàØÈ.ßù|wHa”$Þ!»~8T6‡ö'||åYÎޞƚ˜a,¯}æéñÝRJ:ÃËY½-F¢[Žš¢”éX;šÉú(Ñe6üeY Ê”º‚ñN=‡ Sؼªx"y9Mòò^ŠBÛIÓUï ãæ„žZÃ/Î „U“°sßÚ)l[!¹+ Ù!À)â‚Kdñ +¦ Ô‘L 失ñv80à^Óé)*¿çÛÙV»¯¨Y¯B=ÔÊ—ð1yYðIisŸêÂÕ8m>E›‘¯*„·'BFm„™™¶÷§$Z&å¶LHr÷b›Êº—‚ÓÒ–Ì“?f=°.¿µ8…2[ +’/Õ{ÍÕ›ѹ"!TbR…/±¤c(ShЛ¶S9£1òn[Á-Fu&¦ðdóÎÖ§ I¨JrÈ0yu’²N+™QïTLç’aA¢7mõ‰ÓÙeÄŠRá!ó<ËØušp\ho9/Á¼[]²’Kj \Sз]XVýÒ¤€¶j!Ù¶ +MÓÕ`´QñT¨€2a)¡OŒ¦\D³GŸ ·[Û^Ä’uzlûÁ +ÁlKÿsWuâ#¬®b$¤0¨¢ý0â ž (WĬ=œ 0héœ}}‘85n=-€™©>stream +H‰¬WÛn]Ç ý‚óû%€ Ô29ÎLûä+š6E\»n\aKJìÚÒI©†ÿ¾‹—Ùg RŠ° çpöæpH®Å55-½R>J¹çåÑî«×÷Zç~Tz*s¡”ÅURu#wja,-K‹Êe.Hâ囃ï&¬ éh´V· $9d¬þónìµÔ0fáx²÷Ú6›fæê p]6¾3·pÓ¤Ëf!u!‰„ÿT¸…‘ó˜Fò“†‹Ã¦‰KñØgˆÆÜ®a¨ÆŽµ´-†ÏÁUËé‰úôT¯QXRÖ÷…,×ÞH³µñŽ9;¼ÔÕpW?ˆ/Y­ŽÇÎ6 ½o¾ù­7¶îtVÒ Bòì° ë9·ÑÝxcº{‰¿±Ü¹»¼úîw}»óàdÿötypqùíù³7—ï–‡W¿¾[þ±ß¼»ðB ¨íÕ3<øÊ”ˆ ?òSºþþ‚ÁCRå zÈ£DéqÂ(Ò÷Á¡Cžáó}l÷øýñåûýù›‹ÏËaZ¼xþÓ⽶0NO^kLw—?à‰?áïÆA¾:ÁYÍ¡hm ó‹æ÷rÌ©Ÿ¦èøºüg­Û`Þò{³ÔÉ‚)¯s5­¢µ>>Œ¶^çÀª2¹j²ñ€‡åó +¥)‚¿¹O|`›àäœÔæQ˜)zÆæ­‰ÐàÎñÇHôùþ|Aýïtþ^óÁq7ÛËÛëmÕ\kùàk¦×OÎO^|>{»ÿßžþôþ<,w~¼:]ž~Ü:½¸»£åþ^}Ú]á—–ow™·$þ!O´|€ùùMó_aÖ¦¶È?ãË_ðáßX͘uŸ–²ümùþZNÔûóH(LC¡{©­Òµ“î5ð8F$̨7Õ­¡¶%ƒùK†SñO}vt”\¥dÂ'“Š7Ýþk÷ööDݹGÚV€ÕÝF>|p||uö|ùFŸÝäåšTdæHôô]ÿUK +/Ÿ4¤ ”LHKÆüáî§ÊhJ—?­£cI›ì–¶.ì*z­ÙZ©ÐeÖWm€ÖtâªUocÁŽÌ»md®ÊÖŒ­Q5#pWM¶vF&sHæ°4Ò“M¬Viz€°ý‘õ +D¸Mµü±¾ŽÃp±)+¹nlB箸|䊡5ýdóÛ7öÍ›j! Ÿ){ M‡Ž2Q)šÜØk@^ˆ@Ô#¶zo2z@É{Íöì(h?ƒ€@v#².v!èÑg„¢CÍPBÙ ¬ÆœLi"£5¤øÒ4 ˜âéƒXK¦JdBû¹î‡'ÒÙ¨¼PÀ m&¥¦jàNx¤íµ.*óÔˆ¸»Õª©.'wÀŽ6X…A…Õ;@)‹ÅîsÑÄl̔Ң²¶J8K6$HíÊš‚Vô¶2<‚¼ÞsZs…cAˆ±´ n=®>‰X‡eÍP”Œ-‚Ræå•ÔMÁi ‘‰ÀK­a”a„YQyå\‹æ!¦jÊ-mVT.Øu·dtVèÚ0žR‰>Òh™ì®“QÏî·—¬€2 ™ð–£7¿“ˆsujQv³22:jTTÕ⃫âêq¥“Ã$c}„Ôa 5ehæL”+d¹Ú³ºÚÙÀ­BC´9p˜¢·Ns€ÀàÛŽ[|έZb”órrã &y´ûñv"£åþ×ç—˜àöåxöóþêüdùõÝ›ŸO—³ýÉé†Ä®vwží?~x³<Þ_.P—§à8%m^é[¿çiD3j‰NË©¤–Ÿw1'!võà ì,¬)wµVñ¬;IGÎѸ9 Ë`.»ÍJ?”¼…ТhýꉴŠÊs»í44jq©€MXe€3êÁVÅ’ý:¤åÂYÃ$o^Ý¥L¥‡vFQP“ds€*Ææ¸B—ÌŽ Ú4©¨š[¡uZo.• ÎÖÞ}½Vu±”°_”B“ו'@ñx÷ô7féÿ_#ê—$c©•Š—aCJ·û`Ç\9Ÿb>Ø€YCe%kxc°ßñJpaÄ;8 Ez¿ hQ^€È¬w]ÁAZÈdh¬à¤ÈPBÅâB©Œ=œrF³ëÐ2¸x?êB±4Ö¤ ÌØ`$וc5âYUÅk¿Ug2,ùäÓª8¶¡ÆgQá Ð°–D²J³+ŽM[éëñ‰¦~›²¢v›FеY²¸V£`È7ã7FÛòð9©2À3öÀSb$J2½\ƒ‰|xÎséæ”Ó ŽS+@'’/Õ{CÁ):Ží® Êe%<Û„¦‹µ1¡”Ê0›`JxcŽ}‡J@ô,¹7ßãÐàÔ,âGunCÃd?h£Ä«ƒèMíŒ2 Ã:B†nLFÁÑ硆I‚Ñ{¬Ó3ZOÔAA‰Z¶)‹‚l¢«¦Pê¸Ç»?ï¾ ÿ¾Ê¼—»árÐ}&}Ñœ®z6“¾à-ÕŸ¬‘!†µÒØt´Åô¤1$*Ý¥ùw•1SÞÑ쵕º™žèo»Ø<°œ9ãæjjd˜ŒjDÆp‘tα:<Ú½øôr«ïÉü¦j­ô…MŸ*P§¼A;+Ľ)uv%fý`§—TW'èwÔÜÚd1ÅèCÅ"·mÆÙHƒ&º~ÉátÄM¼tU»ÊK×üH†@ГM†ª +ÄâPb¶¯ ÉQ'¤);j™ÜXxbÀÔP3%½%™] AHz‘Óp|±Ó)j«ŠT›®ec~Ý’µ:{39j««JÌ«°ÙEU¯ôóŽ« ä³cÄPæLj* ùBU5æäU:Yp¸+0@Š„Ø™&!OõiæIHÀ\°H„¼šz= +Y&z­íÊ/qmrAT¿:dÆ”Ó`¸© À{e{&FéÖj*Áö1ÙÐ*ÞU¨”cñn5¢ÏêE±D« tÙ¢ë ÈÏ"_ðl4ü.Â%v6ȦêÔú_ÚËlUãŠÂO w87ùF®y¸´"8!881„lI N4 vôöYî>22c!ñ«è®®ÚÃÚßjA5Ìo-~Ê>˜?ªð`ZÜ|±j0¼ÏëY,&äÎs6з¥£0dw7Îa‘–¤ ô ‡ÄªK »Žca…ùv{ä c B;Ôxvn‡ÙêæƒF/6^j­fåØ6ËZö¶ë€î-åòb4<"4Üð1žÊÛΠ†È-ŠTä$ÐV è™/(IpPpº˜aðe˜£^&½˜ÃT½èûh8ñžždpvÙ8€­ú]çàòP„zûœvLh·A°%BÏu`yX±åQÅ}e–»ÖK`i¤¹b’û ³#ç&œü°ß{ý±„i]ôC< Y n:13qî3`(åª0¬˜ÿŽx²€!ÍŸöÝ:ÍMÏ^žÆ¬t@`¡tj5ùPG³«ä½äþ«ø?î/ ÓX¯}§²Œ,ÀßúÖ¦7U¹·Àdp¿'}ND$(>8©‚“~¥.R’-’aä§7‚k"Å‘-NÄŒ²M„‹pz@ç0sÐÂâ…_ûµDx`™Ub ¥{:‡Ë6-™º8œ»‹4[réÓFÙqr&‚ …šŽÏFY†ÅIèžÊq;Š¸NV4ô¡ì±ýèñrЇä©Nrzo~æ¶,¦w}¨ëÐa9r8Ï?ÓÇú?T­Ü•Bp{6P#qßÝӑÑò8l¸îÙ[4ÑPÔ¡ž‹¡ës„ L£tµŽµæ +éW°Î Ò¶ž¢?FÝ!‡“&ά±VRÏ‘-öäy“«u Ê‘È—çÀZ`"µþ1+¬…Él¢/\S¶Ù©yXô>£ +UŒ2æÎzh˜ÔqcêŒÛ¼ËñvJDm>qïç,i£À”ÑeLÒ™—+­˜Þù„òÛÉO±˜ë4œÈ'Î/è?{Š ßÚÑPÚy)%c‹SåùÑwÇðÅ Àåx‚î¢S;úûÇ ûõCœè½c„RÝꪑÉâI¶OAÛCÃQ0P;"P‹&Žõ`­°?€‡\¶i|Õ<ÌÝÂ~@Fr|,#)R‹§jžÔÿW`”Š\NÎÙ¿Ï€³ü%o×j° ©kŒ0ý§ýîžDS׃ H™Ât&Û;SÚªbG$è¸ ÕƒnÐ5’îmlp,[q­òcõªÐ}a$h—jÙ¯YŽ^ÅûÕ§í=DßT×m[¦‰zªÆaÄðn×±¾7&”öy ŽÀ'ïs ŽÑ J +\>&ˆâýËqBH·Éc8:¸Çv¯yyèk…`s±¸S‡K‚­ö^ŽJãD©Åb{°ÐÁ7!„þ§ÞWtÛÐûm^½>5ß‚“I½Îp=Q5™¶ôc÷ë{ûæþ'7þò“_?»q¾ùô‹onîöE®üæÍ›§¯^|õôŸß½~óJ›þùÝ÷O½qºùôó—/Ÿýäɇ/¾ùöÙÓß¾ýîÉÓ×~ª|¸ÝÃÿ|ÿòÕ›»tÂsªßàÁô÷‡/žüéÝóo_>‹ÿŽ¼ˆ•û·Ï^¾}ò `è3ü}ô£¸èK@ª!éÏ¿±ã÷£wøÏïðã_Xúñ¦Ýüáæ¯K7OøêWÌ à· aIÁ¯T’XIRgäiCrZEÁjø§™Ö:#q«(š=)C¨^`¢lKŠŒ¤EÕÎáÊ’éÆТ’j`ù"Ȫ2©¼ „MêˆÅ“: +E®FT∘$>kc)ŽÊZOÅ$½ŠÌëh@y€aØ1ŽŸ¢šFˆ• '!ëÓRŒ­Lˆ {‘„HDÒô;uå°¯[b5`74±ˆìâNà) ÂЮP=ö‚7¨á•[Ä ì¶cFÕï/†O•¡¡ÌÔ²çFib(dù8Œ»%j£ 1…9à‡ð˯g2œŽŠwÒîºkEðQü’Wr +‹Z)ÇùgX4¸tsñˆŠfR·D@)J]ŸBÞ‘µÈ!k£ëcx`¹ˆyœ%÷|Ö\50D¦(¿ÓæW€Ü¯5´ÌÁš•‚É؇Uj÷®¨cç¥ÁçÖ}N0廙ÑÆuquê^áŠ:Æk=G`dÒ|¸»¡ed>dnÈfêSŨÅv踑Lr$‘Fe(7pŸ%ùQ¼ÕÉÁŒai®,T)̬‚EöØ.¢‰ŠXeŒ†Ãì3†fT²àggY¢J.þõ]ÇZaÑöš;sÙæw,+iÈö©²`&'.:hÞ 2¹s8‹)@‡%?qR§'ªHܼ(¾×d¦ée >Ûl t¯¨Ñ"è§)3ƒ8 v£²0hûuž…iU¼Ø¿}Ů˨ÓVR°ó½:^˜Š“g­q‚œt¬A#3]“ÇÒ æEá;¥3¯á¸H¯±DhÞiûÉrm@ŠjÒŒæÃÛ™>Q½Œêk]C|•¾CM5Ð 7ë`cwH«YÄZG{LW瞪l8Äà#±.Ù¼Œxõª¢[”é­¢h&@/pÅQ}‚U†+éÀr&/([=‹„ü—í2ÛÕ«8¢ðœwøo@Š ‡½{H® !ˆ(‘ +È€9yøíó­Uµ+Y>˜>½»«kXÃrOÉtcœvòŒ±vɈ8€„¬z ¹@D‹gXç«LCºaqNí¬ù=qÏvur¸ƒ5UÅ¥CÇŽÁ×[)×ôЀŒ_|/¥T¬>»)de"Ù‘^_8…]&5íê-@KiG<¶5·‘ë +D%”i‘§ Ì΄£]ÔË@ ä›ÚÍ0ã 1ŒÒá6Øá¬~ÄœšÀ$Ôj@$éÞ:݆<Öö%K•$Ïatü=wõ¹2.ë~-"ëFà!Rðð€ïÚ£¥&˜€šÈ å4÷Í´¼[MaèrZ¶xª´–ËÌ€ˆ^öˆ‰j°y· œ£*Í5„Ñ1†3¤‡DQ¢)”˱Lšh„=[,–€F³r˜âÊL¿¥I\1‰[ˆyìt£MéÒU4$úü2h9H[9WÒr/‘.¥ÓÊW—!Yò{«xÖfðc¼`H¥(X’i"ʼTâlÏÈ!×´°®@pË&˜ôY,Óݧ­Õ@ù3£Mìgî‚Šûê ¿{÷pT`qd•ƒ\VòÑ4”Ñ:P3H)XÂm•e„âõícÍ8µß)•Ì¬^kdÛÔ¦ÅÂ~ÉN¥õpZ¬ã<ë=€=·}C •«Efà׫îºè—fäÛ‡?âTánxÞyCÕpÝMj¸)!>ÓoA5ìÍWP@-ÿc££c!“S©]Ùä[mjŽÔ$³hŒš“’è{œ­‘•Š;ÀŸ~„>žj£FK }/:)ö"’ÿbÉ—ÜU݆Êfí!…™¹*ˆ?¤í¶‘@¶£5¡æRU~¤À'ÄnGÆQU!†8¬˜æPa¡ºÛr2Ý ÛNûn¿L}&Œ Õ@4D6×ýþÈèbL»>ï2QÀ&âàÈ«¥MíexS°RÃÞ-ì ÷{¬}ƒ@«í™ ËÏÀÔœÛr±Èa\O¾Ê1Z±j±dµ„¤])‡CYj¯qÓywX2>Ã4«9 E +öYCÞåW@¤r—÷rÝŒ.T¾eN<–é( ±C—ZgEÐRï,C;;üì<¸ÞúM{Ó6¢R`ç•úq¤ P·]ö„kkäjÒ,Ã2‰˜F +¤ÑÂ51)ëê –(¬ag°&æÔÀkæ÷g°é!~9.ƒ5Bf³L»·êÐwÉiQ!»EÖæ]‘ šÙj6'>²p‹ÅÑCð<;g°5…¬Ü'WïZNŒ»(µ‘9jf/h‘Æ!áéë>b5ïÇpçyñä­ ·°r™0D«„ BâOi·ÀA*˸…&á;!«ŸQnšYn4lgBÒ[°ÇJà;ur¯ã‚ÁYr±!5›Gº¢q$ Ò43Ô¤(¥ƒ,äÙ{èüëTó(â7í,’ájxÙYÏXÞаöÊ«™¡ýÜÈD}úØB8ÐjVäoIÏœàòžyJŒ#=æá4?[²Cö¥ç .gϪ”³‡Ïásø9nŸw"•BDG²Í:àÚáJ3SYíÎêQ‚"*0† CϧA-”òC¿â=¾}øü዇w¾þàïþðôÛWOŸ?{üâÍí·,ÕÛŸ>{u{÷ÃOkÿÛãW¯ž¼xöèÉ¿ž¾|õⱶýýÍOÞ»ý†åöÁGÏŸÿ;?~öø›ïŸ|òúéwO^Æ®öËã>þùÇç/^½=èwü}Ø·wß»}ù‡wØX¾þøÙwŸ¿ùá›çßçÿDÏråÝO^<~ùò=<Ù‡üýò§‡×÷?åöWŒÜ ´5~0@åö–ñkþóåþçOüãß,ýt;n¹}e‡Æ¯U¥Ïœ*Àv ÞmÓvQHI`»Þ—ç1]íÑ 0H5­Û0ï3#œO])\ײkØOzÓÜ®–¥½‹ªræ"H¥iEWM°‘h(»¬°YbŒÀÔLBÌï@ `S&»QfÀ½ôI«önÄö•»ÄÂÝó.1Ãå·`«ÉDž +†Ðæ+ÀÚZ¨U£}•«„»VI +p|1®¨Ô®­†ÕÌy?×PkÆk#TûþèðW©:Ž»ðÂt¥{6Œá¥Ùz^p/ø®¦®•‘¯vKÆvµ\lÜ"µžóJ¸ÈËÒ•¦F²EVOÝ€ª‹TMgE²Ê™Ú}È Hù*%«@n +È\U3ö” ¨® eïRñq”Ô[æeõÀ&iósç÷Ëu•D®Ue9$©À¶ø^*–éÍÂYm^Ô¯ªàwóT¸¯øu¿¯÷|¬øú +pb(T¢¾2À©G3¶)T:2€!±ÂÚ:[õÈ‘Wü€Ÿ|^ÌûHrˆgDÃö>¤4Ñ ʼÀékX`6®:¯"ÒC-ÄNl-›MjaçtÙ*ÉÀ1/ÃÐïõR¾MéÌÚ *J×—‰7Ûi~_QcùÊ•R +à +é§ð“3µ,z Š–æY×̹ƒVè2²µÏ•áØL†¤–&Ÿiw+ cTÙ{(Û«¾u€C¨ãÌËÊ4Ék1å{Ý$Z”›4ÁPZO™ Ë¹½¬Ì1CIZ’£zÕô êÇRƒ|òsÐgî¬Ñ†ƒI2ft|òÞ\Í&8Ãþþ×1úÏÿ„Î ðß=Œd’/îœò‚ymÆz¤ŸTÄ!‚è|ó‹h¨Šhü‰êÆ M®¨§|Æñ‹"þ4gÛ¡´Ë¡Ðtg`€uÞ½{-´%úÿH/›ÕË’"ˆ?A¿ÃÙ¸*³¾—Ò;iEq#³PÜè0ooDFÖ¹ÿÁðrz˜É®›§>2#~ ˜ M•¾ÙC°8òMË)i¦`p·qÒH'°m”7¼ !~κ´ò06.xã($¤ÆÑ£DÅ.(¤OU!^>'ÀELâɃ䰧ƒ` ÐUFJÔ´†¾Šƒ4î~IÏùÃ!5 Ia¿y…ÙgEáýëØq;h‡ŠÞûÏ bò8ì¢àØA\´05 ¨š&[Ãjh%pâØgøß]ÃÑòÙf â!›ìjJ–ÃDAú º(+7Ñ8W-ûé›B­²PшqÙ N9bÈ}Rk¤@­CRÇOÏA¯US±­Ù,?j5<¾ˆžÑãtjʘÀƒ ”õ†CEMÆ@²ø’4ÓšðÈbÙ-Ƽ…½0G0ÌN;ÞfšI%µæù`ßµï¦XL÷ ÊV±9”@l>^$î8χ¡D¶… µj1>G>ˆêŠÓ+ÃD@EÇkõLxvÌEáêػޔõ—¡IXr´þ'ù röY¬éÖ‚²Ä„6Âj·K|3Q³‰¾ê~¢’G+qŽdɼc´—[ôðÆtçS¹ËÔ‚à°Ÿ?Þ¹sVäÐ×gœ œu +=R$ Ïëu׳½B3]!Gt‹Љáª{HÑ×%`”Ivé^ï4ès€’ŸÉÄ@̳m™ÐwØvA£Ù‘>æžÚ2ý „ `°²ô|“æâR*X9Ý^wГÎaésf`W­æ.J´Ü÷»²t4Y]9Ü„_æ&`\1­…›]3H'­MAO±ç1ô¼ƒgøŽ§d¢rm)·Žîu}s_Ÿ?}P krgÐ.ƒ]~Íhż…ÁAýŒ¥¥DÃF°ï©•ñÒŸ3ÊJ\ZÓZˆÁ×A½EUçzé´ %@¶YºšÐØæù1‹@Ö^K½¾u‚×Éú𮞭ه£…rÏy¬tMêk7U0裞ºb9Ÿ£­¦:y½î'ÙÔÇÜ溾¹mÊUëÌþ¬‡ûaxY-–ÎL™ý’« ƨpŸRÑ4#[Ìs}3o~°²þ&÷ ½ˆbþšá¤j¨Ë ûÉ°xŃ¯Hîa2Àí*Q +îÐ! ‘€„•ßãÂ2ù¹{-áu¦‰8œ8ƒþ£÷à“ @ƒÙ¦Sw¢àé]tC5ËXÅk÷tr†×ÌËÇ}•Cž‹ª*ž,^LÝ$¡Düå+ËrçãÓ0¶‡iÑßP¶º0ŠMp9Ê6®î„½HÙè»]º\TÂÎ๤møÝ×°ù´‰>»DÃ0hãd-«žÝÒV#æ;jâŽ~™¡”{(3ÏÅ»ã Óµ´øÏém­¶!­Ä4q)f;Ôž†–wEO.>dbF”¿Gg·…kǤ”k+§‰v÷ ÿŽ þ¶Þ +"òzß2WÜšËÁ-=—…Ìú¦ãBÏóX©¶Mám{çZ=!ƒžž_HžT½€­w‚j³Ú,Û 3PÞïbÿЄá$àû4w–ÞLÊÅ›%„E F1‹ SôóAŸ7-ÝV<1dúTè­‚[W°P¸• Ñ*vŽ +M*ôpRÁ3c+¼ŽãË(åõû5öuåcÇïCð‰ˆè˜“ ù>ÓõåÞ@ Y%™[T /«îhM”xäý×V²WéóU[–ƒu•0Ù¼î&)p•¤? På'wë9(œnãÚRS;ÓÈÄ.çÑÎÝî&ÓcÃÏÒs·1÷D0„5Khz>•5È"˜[­‰&¯yá`¯== ƒÂ%ÄæSÊÜÇyˆÝ ùGx˜j ¥ +o*–¯ÀÖ(ãƒè)ot±I4×,¤…slë3Q,‘~Ñ᳑n-(™ü¥¹ŠAM®M5& bòê¢Æ[ÖÂø®[@‹`2梭•Ã}³h[´ÌrŒ¬/†Ž&~ +J\¹:8ü„­‡3L*:ahð /¼ÉK 4C´Î°R~„'am0´»¦2“b¬jú}Y3ßWUv^ZßõäÅ{· h$ó}2Ø 8[&mgÐ}õdåqÀŠz⃀K­‰ÛhÿœXV2<_ÀF2ç&!)îµì~µݸ-š n«Dâ:JÂå@áÜoá&³Céõ9N· Âzz1¬N[¿«OhQ9´ !]C0æUµ‰ûÎ ^,ѧ¸úQ„wm#ßvVuieõ„–Ðk& T •Æˆ­¢hnD]U5Ò—=ÎVkOEÇ–ìÊ=!¹bŽºÀšÂ“uÔßwM¯)UvÂÕk:“™ÕW¸†Ÿ3Ì¢y…[&1Àa†q”Å€qy2YtOìúåGì­è÷CðqtÝl /´ãí±MiÇǯm‚bZïh>@&«ÇÄ”áɆÝLÏÞGÏÏa„²t•È*ípÑêѼèXoî‰ÇY¤¨±V„¬àÖ] />ƒÞ8=«Ô€Š/ª,o¼ÞˆhWæN „˜ ŒîÅåx·P2`í §ãÌW¸‘\›^N夣•èÇ“ô&K=-R] éÅ褫¶¼¦$Ôy~_(_M7ª©k±6^—)´„„·@Kp_ŒbºB”ŒfQÜqû€–ó }Õ›ï '‰ÿ²]e×v­80GðˆAƒó§kûø£ÿ®Ël©†Þ-ÞÏßo5?£PÖç ÚëЃ먟ӆ“ÌŒy¾€„ çܱ†„°ä0(eü`êàÛ^#^‡]°&ÍÞlõmïã>ƒ\ƒ?%%¬q¼Ê§1§É¾ ¥Ïú[MíÔâžÏZ>-*Á¡“<;| ±£î½ZT‡Â¸;ÀÕÓÎUÚùlšú0ÓÀÌ£;%ñ†dËJLãq™lg{ÍÁñÐØiÒÔ[ ±Ntúáà™ú‘Íò7ü½¢h–GºW1„c]ŽN¦€ Üýys¤o´A¤ €á +œMéŸ8Å•C^[Ñ«âg:¹ËÑm¦¸íãP8Áð5èhÏTïáè pÆ+ hìŠHÙŸ.¾yENƒô+ÖOºéñ1–{4i(>›¯ÉqŸÆ$"¸Íyk-¾vW-_æžþ©Ÿ–°žá¡Ó¿çy¬ÙØðñÁSa”04kfäŒX”¹•Øòö禔‹4þ°gðxÊêèãk)Ü}0©tv™Q¢Zr.cðKÀ¨Brð÷xÖi3¸gÔZ{©Tß-jÓ K‡k®ï·k*ùþU¶Û +i¾°öߺš>ŽÑxï}ÅCUfH˜ªÜ^W€ª=JÖYÓâ¬A…k`ÃúÇÖ’™wÝt*yöìA€+Û?ƒ +¸§¨™þ$ï[k³$‰ ÷§óål N©NóO¬EŠ=!gj/\¦|:ÁUdÍçî’>l‹«W#a^T»¥H…1ŒÒä‰R¹iÇÆpw”€‚¡¥iè\Z îï­ÖŠ’î{úÔ» ½CÈØWÈ|%‰åcÁHU· ñ´+«¸Ÿð™à'Ø„Õ{H¼AÊ_g¡õÒBæðÕnù‚Ât»ÒÂl­å ßêb6¼±+Ùæ¦J€þmªtÃjÚ½1÷Y%Ö½6uþƒü=¦hµ2P˜õý§Õýý'¼À¦‰ ê¤Ú€êÛEëg3 é·‚]¼£*k=ÌO]£’ÃŽª,Î+ÜÆ°W½^¡Áø³[9Sa[]ÜóׄxKf¤›»l4×JæØz 1‡^ƒ-*mò¹R–{á¯í18Ÿcç®öþïÚæ;ÁxƒÈÊÀE,§ØÞ›ß …-šF1Ú5¸!U*,JÐákô³ Lª&˜t jDf†#FÆ+›•*t”íÀwôÿjZZ…1mm3 @F¼ØÏvÜY'`0"Ý +®dy­FöŠH¯7æ†òuqV@9pNÙµïA ŽëµpÝÓ_ì åB $P±g5XC¿_ñÑùñæø_ !± +s§x€OO;è+Ü2tË“„?GŸvܱE‡ײ¹Þîi¿thÈ{‚vÚp'Ïò¤Ýÿü²pÊ}†èéØòú÷ñy“7Ç+·%AlœÞ³Úp}º;sT#ö nqÛ ÅÚQÅ:}T¢öã•uÏjÃuÿM9Ú¥Ê%'`Æ—º0•yŽ†:Ÿp[mŒ±Û^ Ý~-HÓj§œØE <ÔÖeÒÙ«œÉ9üìÜÒÉî€ùqªrÄH¬Ô‚Zó‘,à‘VT†TçagQ§q +WÁ‘ Ÿop°=å¹ûªÃ¢1»Ì ʧ)cÒTŽ×Å(ÒhT– î|T Ê?@˜Ø~ªX%¾Í†Rà'nÚìÐ;™9óeÌÕ¡ Û»Òò­S´µWýŠl wÞóä[ v+ Öt¢D³ŠawüX>MQŽšðØ»\ÝOp°;ÔcaDåb¯m0HFÞ—B³_¼Kk +® Ù¬, ¿Õ^ãh•]óµú{ÍôØ&¤ªÞQHeÄNjZ¸×€¯tŒîŽÅê§a˜$ûKX §M1$šûBÑjEz{wÝ%›QqñѶ÷%<¢àl»t—iÈ͈(¡-ŽäpÄTÕ+—àÑiÿµ/à1Vžòz_ô€Ö(JÍ–o ìVq‹=¹N°|kÿkÛIs3êû»ÖÊXaºÑ%ÃÞÀžfÁ’‹fƒròûÆXŠ5!þ:iƒ2Q]èø½»àë¬qÁ¤0C<••àDW©éðšÿÁåÎÚ7èŠ`Ü€‘àµJj{‚8Œ/À[púUœÙDkA° !È!‘^šÂ!¦ø¹.Âð× ³¤Ÿ&èýFU±ôÝl„ê͵L'^Éú”{¥ÁH.Ÿ\êK xµŽº)oñî«¡&¨Þ&ù·àÌ?… ÷\ûõúÁQï9Ð)GÔ¡w‘43ÜÝÜÁvƒ­¬ÖŠuê×á­6òs¾I`‘ï1 á AÒÈÛc6>Á6¹¬k¼r8JUœÆRáË$±ÊÐs>>à`‚kùq?eAËËO¦=œÉzMÇPlU–lßfØåòGì±Ók«åÄHÚ•¾ ™GäÅÓT•U =*§ïŒ·°‰Q{Ñ(AdUy\‚«í*Då.o²J?Z>× +C–E§Ù¢,ÀÎïkÌV¶íœ½XEÛJ4hf<Ì)îcWxæuû¤¯]ÃÜ ÿTiç8¶&hÀY %œàÅ^Ÿ¨ŒugÁ÷þ4…éDJ¡øðx0Y´‡÷Èòù¸Åm>ŠØ,T ¡7¤¥õpH±ïï@›|6QÞ÷T ŒFüÐÞ +=íVŠ£æÛC~íÁ Èû*ëƒ{œ*ÆèŸÆÿðj[µì¨¢_Ðÿ°_PR×YUú“ø0 bÞš˜D0ih[Á¿w\fíÓBÀ†}ÆY«VÕ¬9Çe—š%Vá Ž3R3ãydº|I¢·,ZÑÎù3!dÖÌ»$°tÓ œ–Ö>Ómj³ë'ÈP+—i¥p3c]Ò±sòÔ´nGŽ‘º¯ÅEŽIÌñâ°¶ ®¼n ¿NçÅ^ƺOC/và9iCª^GèB;¼k-ƪiREç^€SÝ3“È0|yáá\ÄÓÝ«Ü.Œ†xÂ-Ê2ÚO‚íZØ­b9‚ðÝ`ìç.‘ðšÙ^„C —pÌ´­‹6Ì÷…+hÕ0ÒE{î Ó?L¡¸¹—-WZ3«ÃŠ©ü$ß¡Ö7Xœ%™äž;Öu^3Ò[šXЉUd®4+¶- +¦ã§Ÿò"¾™å9 (Õj¹…ÕOÏ¡Æh,;'ÁS3]Ö´aû,Y!Š”ˆÝ:òÙB;î½ÖA@½ÉBØVOf­–‚"•I¢Ý¹ `éEO—Í @mF ÃÇÞ‹ Mì7­ÿûÍ£7äzL‹„Ûp# šÎ™`ÔSlCIp®ö\qkÖÈ~ya3á¸ÓN¸évs> 7Vj%Líâ ÞiÐÉ%Odzág¬ã2#aµ‚`Ú·/^ã$\êŽ'Œ–ŸC)îž -éÕJ[éÝÑEM+Ó9†Íö ‚löb³÷º%q‚‡¦‡)ˆñŒ ‡O ìMé‘· GŠ!N˜nLÛŠâñÒLu9ˤG1Ó¢Â$H•½§¨ ¶ž’N±hF¯FTè­Ôû~)íéÀ˜5"zzÀŒ˜ÓZ孙ȶSi°…zÒÇl1žƒŽª ÎÚòB~0™†ó¼ÀSkîm×8"9W ðÂ7àp9+G•Ò€q'«›ls’pë‡ qÑöò­+€™óžp<äóÖh ׂpg¥ÎËæó>r—Á­Þ;ž–DcÝÞ‘‡SHÌëÄX,¬Ôó–ÁPvú²ß î“©à˜]°(ü^Ï÷™Üz¨jõ·@$ÈEóãÚ%ŸMŸ&œÎ4¸Ì³žT +/\•Ëpõ%j ©LkA±2vy¹¦¡Û4Œ‘ý´j¹ ‹ŒV×J×ÑÚL2Õt +„ý0Șé†Ä—˜1µé`iÍ·Ìo²±TžYpy£ +Ä$sG¦øÓïÁcòeÚúÓl# Œr–í~›fØ +5˜mdˆžCgÁbÝpi6aCŽWgDÿm‹Ês®ÌA²Ôm"›¹Šh… ¥rI«{ç±³N{r/€NÑ–0:#wëÞ˜Œ~Ë×ê3˜Qû.Ç7®& +þ‚ìMŒ%+”l:TA+óbôë|#ü3K‰Kû'„¦4š×Bôy£,Ú£fä¡8•˜I¼|¯ š,²Ätb©tý&F‚gÔq˜&±ÂwÖsÊMëXÍÎ%¥Ýc_°Ž‘³„Ã×—“…§4ÙºwÝ&+Ô#ܵhºØ'Á^/96µÇ6|#bÃÇìÕɳgêÊÙ>qéöö—Ú.™qóβ¨k¶ðÕY¡b3SÔTÛÎ}ÀЦ §ø¢@SÛ¡Nó“°ák<˜¾\¦¼ïíNU5„âÓiîvêÀŠÁ>¢yœ9£…“)š:õÙI£ÉI׎۽ \{8Kqœ!XynŽs9®ð¯á76•O¢jð‚žqx8x˜š“[dáä¢pEº ‚ =Ó"ìr@ŒÚ3†->` ŒÐ° ¯ÈCXS(˜bd°å‰îTÁˆ†vx¥A.àælX~ Z"è›áÓÑžnŽ•¡R¢užÍA³Z#Ÿ-Úþº¨'‹®e-N›æ`âýzsó +F©9©Hut3 HÖ«Ò‹«ÔçØ»zhiìL²Èh;÷ZC{ë¥Yë3E³¶;€Áõ6gKR!Ik¾è~ÒO¸ØóÚ/¦;‚ðtk§2Àcœ\÷8MÉëŽ}‰éy•«ü$˜òzÍÓ_nVh)l«æ!ÐÍ LJÍ»]í¶1ÎÝŸµ…‚¤¼ ›Õ…Í`î(7þÞ¶‚Ú8"70y‚Ú¬CŠù¾{×(Qõ¨áà0X¼žy ´ì¨Å_p“ ‰ØiË Š¥¹Úr½ç¬ì$4ß݃åPÏáËÕ´ÑMNÿU–:¹öˆmW9i:íÌ0 š«4 óˆföK;ô‡l1^GámÈkêæÓ’l´w9WB{•kôä8‚§îÄÊ ÙÚtC~¿¥ 'ÜG÷³èúè Î~j‚ ¤_Š}7€Á;÷ºŽ;`K»Œ®ïrk ShhDôe½,€$˜ºó *,ž^JdtÆp ¹Â”§d]õ’Ay9`Tpä¢å–µRBǾ°P\yÛÚ•¶ZéSÓ·èS«m7~k÷^¦¨^ƒç6„…i4p'6nÔp‰"7Ÿǹ_ÀCQÒE5w¡µ TÉÝrÙ†uçxô‘®–ðn§zÂå³ ž:3ás›0Öj™úʽ/ôaŸå©>žeta?Whý‰O[zIÚؠԬ̺n›³\“¹/d嶺9 +;öI>ÒÇ j´#@¡n¯—>®;a5”–í"øð°ª,µèá‘­T0FÂp E§…ák{ð´ã +døÚº½¢”ù§ ïšråå0ù“’žÆ`ë+†üâªþnS¢3Sñà|Œòt 1`D]®Lp_%¸c(çÄ”hô0¤ês” mFÏ>óÙ²-\ò÷øÅpk~u}üøú/ÿç}úÝÛ¿~ÿøôÝû?üüÕ7ïxüîÝ¿þùÃãOoßþããG}”.ãë¯^½.½+øGóûò¿ÿªÿRàN0Ä_ˆ)>k¥}M¯^+¾(åÕë7ŸàSŸÿøíûßþüÍ»ÿ<~èñÉß¿ûñç¿?¼™7ÚÂ÷ß½á~>~ü +Oü?l¾`__÷*xýeIíPJ‚x´2ÿä?”éiP}ྕ^ h³iA¯ÐêÇ¿õ4.4TH…½¶äx·üK[fµ=AÌ[á +¶¡Z ]ödc³©™Ì~bH2ÀÃOÎ*ÎCÆï›KLs;ÚC/o²aØ ÏÓŠ Ô.`=z8†6†#—€pŽÁÌDwrJ×Ah²íã¸û}ä§ùb¬E¹ô<Èþ/ÛÕ¶c×QD¿Àÿp^àeÒ÷Ë#‰d@ +!„{@{9ÁϺTïñËk¦Ï>µ«ª«Ö…^©o~Ý6ƒÌbK²:V±A»•f¿ùXH¶ŠXôtÛw9Ù3õÀŽº·ªÅË7’ÀŸ€n_’‡€™‡¸È$Ô%:Vź¦",TÜã$ÆQç§w +"ߥ-[‹ RX[;>‚}ûcˆAH^ãih€%(1pƒÞûé>‚ûÔÍÚ¶-@¼"z耀cÞP”f7ݨ‡uÓàËQåý0,» ž Žòm2SõêP¦é;uãC…*T´ÉÔ‰íT“‰5¶¥„6h·Sb š­‹¨HOÛ´QÜ4¬dÎëHÃ=!ÊôíØüd_‰² †ða8[6K×e´©{\D/ž^!CF‘Q¢¥Ñ ÊÅâŽW4YʵkÿЀfʧ‡KR ¼™Ëè3 +n»ö’m·ÄRþ'^ÑÌqMJ:¢:ö’¸†&~Ô“5§~JÁ@Hlu€ ÐEÉ­ŽcoÉêù†rŽ†ÒàLÅæÄOrd‘ºLR®€ŒÓ zÓZׇõÇîJGÕ*LéOúc‡MËÙÞ úÁ€!½7ûÑ€[=.Û=Žú(—RßÖvµeUŠäÙ4•£ wÕõÑ‘q¨£¼F'Z¥Ï1HùûN9/j†34päJûH„Ó ê‚m¿¡®•-Šœ˜Ø·æ |#ªÂ‡íÉÜÃœöíA“MôëU²vrÕ±å‘O€|þtÅùðÎìv +™¨{»1Ý£L&¡( j¤‡>‚צ…Ä8![‹ç>…‚2ÅŽ $¹qEÖˆyn¦ü„H  0O#æªhè=lžoêâ| ±§S&„Mt›=9#3;Þžôå©#ϵi­FI+ö½#Aó KV-Œ·ù2’qP—Ù ËÇ$Æ"“J,âÑo‰§Q‰ ƒŸªæ{0û2Ö(Jb kQÜ.×x2Ð#{¸öÔ: ®{!X3=ýÆC‰/Ûë"ñ1®©C0 `‹§sÉ¡›7‰OwE/“˜eˆß?ËPDon}´+Ÿ|TaåÛ'_>¼C´/s…ªzwÿöá«û~÷û·ßð±¯ßo¡•nŸ| +é÷äÉϾùöÕýo~üîåý~ªü<ÜçÿýþÍÛwž(¶_àÁô÷Ï^þñýëoß¼Š¿?Eqò˯î_Þ¾xõæ§û·¿BM¿ÆÏŸzö#þ¥ÛžÑÞ€| ÿa;Òíß8¦8UõïñÇoñË¿pôÓ­Ý~wûëßÒí%C|õÌD‘ ªØ¹ÙÀ7¯ãx/ +o°ñ€~¾‚¥0ì O¬†h ¬ép™°^0À"ä“kê&X™Ü¨&BÝ6V²˜ÓĽXÍßÇ°Cp ‰ +–%Бw0è`®°aIÒ(ÅÛ7Ÿ±l­ÕÔÚ·X•’чN¿g²’ÆÇ…UÞ9ÏÌ5£ØªyHJÕ¨0Ð)ú>ÙM#´@î l‘ž­x-õ^ŸDÃQÏ"VG'RAªŸ) 8ÏuJCJ)©39,¬ëR®xWO{Å»6rt^Úç…‹#c‚ô׳]9‡Frcx¨¶«1ÚÊŠÍ™˜~íÕ.­¶Û]0§³U궟Æu5ücóYjýmVÙapÛd_S&¥D“'ÿˆš4rÖd¡B HŒlk÷h¦zÀEòZ¥Ú¤¦â!e Üw1n‹bÝænä!_û¶Ž‚ø ÄâÀJ_|­c4M×Þ@p} ùWÞG«@;  j$͹•6,9Ú ùWãB:@bŸÁõà5„J ½hæXQÁtÐ\šurI1Î#¶õy̳†SOk‰ý2ø2Ai¦z'‘Ù‚}m,¢Òº¨24Y’¯W¨ ±‘ÔçÓ7 ÉÌÁõÆjiÃòö’31CSc2»ÓKUãð{êÝ l‘Å}à1$7]Ad¶màtÜàÿ”Ú8ê²ãªÏ0’Y=Ù¡)³Á]À´áæ»ÐžÝ”Þ:K w}é~4h¾2›¸‚Ok +^Çqìa¯Wf03K“ýÑð™> ³T€@ðÈC§¡ãV¤üã°h‰ÐŸQ½Ý ÚaÃØÉ¿mD€ä•TúbÅÈAyÊ -ÝsÝ^(a6WÎ͇*½ +fgDò€ˆ”²îà-F°Ê×q(¹N·fs +õ4ð>¼S‹ œQwÓ>‡‚@Nm…°#§#бfýê+RC´!ÔC>9îîrkÈð,«å†2Q’“Hù"ôbÙRk%,}˜"-l5•ÿ¼„SÛ¥íÀöˆ=j˧í¨mõÙ"ÈnÒò¸cë{>8® )(éØCìÍ_2pyë”nKëÈO8#݈H~eÔ3ÓØé:B,UƒIÞœz+€å(+Ƭîb­äuEƒu˜ç»”爑,kêÙ ¹m©ø(§<®‰žy,;ŸîëANPu%äi½.=ÝÅo<+%… ª¯+æ6§ðΤâû†§ùzí:o½¥°dœ¯cJÝ]5¦ä£6¤S9æêeG?¥B)±S=«†¸2h:†0¸Ù§ˆ •Dõ£yB[† ieOÓƒé˜Wni„{Q;ux²¢Ï!=*U˜²vØ êj‰™©®²ÒÊuçÃîÈ檳m«E©s[ükXî†ëPo %RÛgÓä"qa‡ÒX›A€‡GtÈïÉ-UÐ#j9†“œ§ ó¸HÖÀ²³íòh‹ô5 èQìþ,!g5áá8Åì˜ó’S£N 4«"öÒÆÒ•Ð&lyôšÍÙóœF bÇ«-xÜ“&„&ÌÈÄ*æ©—ìØãUÜgS“ß aI^ìgSIK@ 4¦(‰Æ! èÇEÔ’ˆ! +\Ú5ÍD…$V@X¤KM¡ç;€›“Ÿä~&5ðáeÂU4Q¼ø~Úcˆ{ ^Qÿ_€nçT†×™” ãvïüá¢N0³ãÉÈ]bhk±~+Ùzƒöi;†¾ÙBµ®¨<œ¡9ŠÝ”S+´"m.çígõ.ߺ ©sÈÑWpH¿. é +—Wãl–#*ñSD¡‘m +‚Ã&íÖsòP[á€M3³ÛëÒ°¡=éÙò¯ @®ùÂ'¨"^Qe à°|àšÊ4Ðà˜zMB‘ ]¬ÍäÍ­^ܺ6ãÞQÙÿx¯–U»Ž#ú÷Î$ A"÷»»’‘|3Ë…€E3c8sí H„ü}VÕªê½Ï¹×3% «Ü§vw=ÖrNë nvy[zĈ®…Å…ÜÊÔ +Ñ+ävÆðµ©C«6b*⨠0·âoùæçâ´Q3©8¤…`gžl üâä$Î^½ö3„à©;ÎÂEuŸ9TT˜Õ™v÷Kô×Mÿ×R ¡ØUq¼t ±îmŽÍ~jÐìÅf¤È~HhCŽÉhîÑQò@9)E8e™.Sø M)'qM`¨›ÚmE öÙ'‡h%„ÙÅ +H£µ\¼a,ƤƒDypÜém¯´{(Îü“S +äzçiS‚ž|²úL-o¡ødv¯ªþ&”j³ƒ«Ér½ZÂM4éÑUKiÖR †iÛ'’JY¨´ÐfhE B#îÕ¼u°ï¥±hó/:¿v1%‚Ø “Ò|¨l|&l/m+ Ôy±—óðôðãÃ_äòæíåﳡ™¿ÈÙkØèAgøÊ¡G!PÔ,¬ËR\ÖW•™&³¨TÁÞƒÏSØ +,ä˜ö)×n j²KXÉÐvy¨Ø½BÚ¢âhBƒî…†8½D·n¤Eô¥MXM÷YƒzÓG7`îÉOõÜFpߊ^`]%P¢ aQVâÚH¹³.sèÔ§5…¤QTDX_u•WgÖEºâÚ*8·Â +›‡Ÿ…½èl‹ÔJÜH(Aóg™È¿ºöGÉ3gÀYI¯ÕJnœÃ1`@«Î-XÉÂ9 †O •W·¬U;ß<˜Ò°úcýµr¶×o ¬Nc +ϔ֡+‰É7õ¥5¯¾[Uh°¿¤xp]½EÙm€Ç‚e7 Ovûm[ócv;UݬšŸ²þ駡'ãîJÍ6»íòcÉùxý½ ʨ͓*òG‚jÜqáÀG䢩ãÅ%W óVTš`»1õ+Vœß•àa×h…t¬݆¦Ïa“¤ØMÏjB/áßþ-Í윥ɷçÄi#;¡È´»)1ͪ~užtYØÞTBÊ5ÉÞG²‰Å¦ØS!ž¿×pów9shP¤G¿(¤µ[³ÕÉ ®=øð/ïÿ‰Îÿùeçò§ÙýGúºç»h9èð¹Rc0,aÖb÷s0ø6×g±¡•X}Êzyøým. Žœ +l¤ËÓ÷=sw½9{¼µm½qû~•\„T ‰¬uÔê…9‡S¯Óá¦Ì.Ü6Œ÷ð !Ÿ‚WO°@·wgÁµ¬»¬ç Ý ^v„uÇÏI5Eœo.°ƒq^öîìñ¬#é]¼4Ø_(&>…áç»(Ê’òTJgÐl"ƒôv;ô2õèé,DTîý.ëäâ÷GÔÑý”S ²Þ߃±èq×ÓÙó[#éýû£.+hû?ÿõìáî0¶Î6Á×+râ,ⶵå EƼöïGÔEêêŽ`J˜¨By2oê2 ±Fy¹ÜÖ}t,ÿ>.ÕódÐî|Ýh3´èl—ù·æVÔwÐÒüæûoÞúüÇ×Ïùù‡Oÿ¹ü¡7¿Ó9%æ·—o>|þôñç\Þ|ûíûëõËów¿|þAϾ½ü'ÿ€?¿N]ËHÈd ª&Hçì2ûJÙÅùp]±hxÙÎ:¥9§ŽYÜOÃÃ[î,×ö±ýhtu/ïÇïSnûVçŒæv«BYš7mJjñ[q¡¥îk›œ_¾ýúð¯¯Tï_# ÝÃJ+§rJÜb*¢7†SÍÕicØîj€EJ­¥C^J)‘ øš@4áõÚ£ÒoÌ +K±öžh‚ýºÅ?F¢5ühûV¶kÇž£¬Ð*8‘)Œ2íè«%øZcîÉSõ¹3Oõ=–pé ¹W¬Ó—pæ$»èAh]ž—S”g­Xˆ¼rì/BЩfIyŒ,õ1úØ×Jtˆ'Ìq—@‰>”ƺ^^­ÁÿG0j« ƒµY­½+ƒ¢þ­Æá.bæê쾋cî´Î½Ê=SÈœô-Z*Ð_µý,”º`­SÚ¬2”$-WàcÄæòƒãå¿ÖûÏ¿§2Õ<ìþýòÚÓ¿ŠPþÖÞ\ÕBÔvw ëÕHè\ÉtÊå©„­Õu”n0ß•Þã÷Ó!o´H¼+Q£ÝeîßKÊ6n%ñ¥<l\³Úà;±‹“ªÝßq¿—>ýl)ÉeF¹^^-Àÿ£¡”kmA~£ù0ß„±tÝUu‡nŠ]ÛAX‘v +R;ªD–zwV­lœÝYÏA»A$8Â˵Ð)«vî¸Áv°··½9{¼kg}QƒiMAmÒçIékoú>mJ„6%ulTϧ`¬iXÊÓYµ²5ßå<íûLp„ÇyÓ£ºJ¹¿cÕy×ÓÉãM§œ/Þ¯…I´‰†Ží¢ÄŒ¿Ö4t„‘¸üÛ½hÎúÓ¡¶K÷Ë<|‡Y´ÉePhtª^nIõ`O|®†“žw‘˜×ë¤v6S‚ûm›A¿™H ÜÏB,ÿØZæ”`eDL2a¦°£ÆݸAQÂzô'€p&Ï™¸ûPÁ.40Wuˆ—Pè ª&€ÞœÄP¬Fá Ú;ÀjŽ¤¶SPf@5¼>|Ðþ•=NªŸ¸²§­m0y)丫`08œHÍÜ¢«ö{䈯=@R\hý—íjYÕò†¢Oà;ü“‚N$ÙIv’a+´Ð‰R*”RèQJKjß¾kíµóýX‹(¿“ìëºdXº §ëåcÔ’ŠJò„oaŠ÷‰ªJ¯ydû«#1zXÑ…ì3'S¸Âñ{Ï,ÊõÚBE—õÔPS]9ØÄš¼È lGŠw¤÷–6]óîÊ,wØ°€Bª¯q?‚U¨Â[!ÀÏ›¤q’¹~¾¢Ðê.@ѽ§œ˜­ö  ø¶iqwÖº.è£L}›6®ÏÒŽÛùJpŽ{?é“ÐKVåªù^ÍÒ &Tg\8§ÎLf£–Þ¨‘„ÖF¡^eÂ4&TmMûl‰Ÿµkí€y}¨Ä§¹ 8nÔ&„Kzv ŸßÇqD¯^äûyùþðDE ŒD²57T˜:–%ÙçP`²æJi0ëõæ6aßÑì‰ï­­ËÃY‚ó­h1nµžß¦?bX úùÈD»()ýVt¥‰igt‹Ï7/Ç‘öz­ ?©4ØÒâUv|3`ôÏ.—µøÛ¢äàâ"W‚ O\q5 µX&ÀJì“œ½_‡Í=ŸÚ×ÀnCŽkëÛuÜ.i”dÄCl”ô/áé(@&»ê`Ù’Íæ‰ †Ã‰ÂàÁ¨ÍO*”Fêxýæ‰ßž>»½þ ‡_ü¯|Q¼ó›¯¾þ®–_¿}÷æÕ§ÇßÞÿ•ÿþæíï¼Ë“§/ßÿýöǷϨ}ðûõÇ'ÿàWj ƒü TMèA¿^Â?¾Ç_þÄÑÇ[¿ýpûù—r{Ã+^¦1fÉ•‘yL7 ÓŽã²£“%u®(eÃÞv°žÒv ÔfÌÒîÚ·ÅŽ€>«¥½T[ß6°`Ô ·¹¦“9¦¥Vêàüɲƒb€Ë©¼Œk…HÉùÝòDZî³Výx'‡åæ÷@k×æOL‡…Ä$û×’¦ÉƒE™,¦®-™¦ úzÐÈ RyAI‹ ˆ®S•ÝRÚF3Î2+´[¦{Ò•ÂBE3…’Æo{‹LŸÒ§8$öHÁ·ÄÝÂ…Ï ÃŠ &Á:.˜ˆ6uÞ*'‚I½ž)€Š*aL œy±£Ó°H¨Fé 3f=R(‡°£ Í›Æè^ĸ1¼÷b*r½5>æ:³ÆÖâF¬š!h¬1kü4J\ÚNé× åôŸXÂ`Ýñ|•ø}8е=ZЙÉñ½m5¥ždø3¤ ;Î@¼J¥ ÁÒUíµÓ›ëÐq0l2¥No¤; +¥SXÈ[*|I%ä-;ƒŽG¶ƒòs–Ó©‰ëL0’V|SS3‚ÐúÉ +έŒoAä1œC.gê„uV.2@c5œÈ½¦†Gêj·ÓŽDiÊHbµAaÂÎÎÖNTðú–Ku &Ì)ß‚pqé3@…êVÈ.2|:H@Õh!/ gª‹ê;w¹­u¬•œQ Ëò‚i—–—‹s°CÓ,T»X©®¥t‰AQ(ô«S†ç[°µK=gS´Ø “F{VÕ¿v?_îGѯÿÅ'&zíUÒÀfwÌ}™(`Ǩ› Yc”R´íÅañrœ¤´­_±½˜krUÆuó|¯TˆlC¨gcê>?Û†a¨Q-§åÄ"G §—c 73¦¢¨æá”ÊƦÆ}~n©Ä€?&L•±òØš4–eh]´ºÓcÅ)lmÈ–pT#ájM¨\|Ç6áÖjálh{¨w®HÊW1¯EÚ¥1¥­‰çXx¬fJ{®nÖì%ü´ VEL‚L‘ÑF3Kú"#þ±:Ás&æpµ,Jq¡ñ-h ‹V4ùLÉ1wß’ÄhÒ Ù7ÃÂ7˜›v¶ÓÈ­Ý ++åcŽÃö<É avÇ–X_±búp[Ƽ}*–ou;Ò«VOX×z†ÂÚ)¬û¢ÀÝw#HÏ|ËL®Ÿ!FЦ9/¾("[x¯  9NM´žÈ¬ €†Âå³õ!·qM¨ï@¦0M™ªçªÔpªE&Ãòpõ–€µKb>è ÌZbJ¡Þ»Š†[·%IŠ‹‡àm'dpjl}…Öt;¹KÁRE;é ¶d·aÕÚŸ4Ö¹ õC•š<ËÔ{è +6y¶¬PˆùÖ!³ì®/CÓü·h©hPŒè`¹ŽŠ¬n®6áeRß3ö–õ! k°ç³)†zÀ‡á Ó ìAF†H-€Ú+0NWø/\V—{É‚ž™]¯'ŠnA?d²„Øš}Mð]ŒV©JS`“KhäFà²Q +Ú Ü‹ +ȧ׉‹1 ’ÒÝ"ÂG¶b€ô:Sл\bø3át"C…æn +y,Àäcn}6¢ß#£ð \6ÃR`#ÖU,)´ûB]fCéñ×€¯¶é©d ,%Æèý¹ÔL€¶¦Îö%tAy+¤~ÇÉQ -ß|ªWvÊg: …NOwà„èX›š1h'“cm‹7»dc–=­ ÍO“:‹¶]2‹È3ÄnmÞÅ"F´ÕÈ–Þ(U–Ž‹ÊÞ‹ò°\#Øøü‡Hâ~F%]‡Œè"v'¦¸µX—9’˜Q!0O™ÉÆ`t­€b% UëÙ8®f¤K]š8 bë}¦»I¡ã¬¸h£¹æNÊ+|e”×ùe^0[’¹»LX¤ ÎÃzÖ; *l+Ð[_rž3¢¹¨.O.‡´&¡2o¦‹ +€1¹<Œä13~™ò/ØûÄLS“,ü¡S›äÕƒH9Ûwz¯piÈqÙ¬G|ü;ƒ'» endstream endobj 17 0 obj <>stream +H‰lW˵e¹Š´ |¸´–$@ q›Ð&Ô¸üŸ¾@;3ëõ,“»â ½²B5Í~þÇÝî¸6ÛxüDÏpÛJc ßÛþþ f½#nÐì6¶­üT†_»°…Ô_aÝhBì_[þçý"K +"JlFB¤B²¤Z³(Ù‹•ÑFÎm‰VH‰’Уh9CDÎœQ¨vϽ)‘̨þ³4¿¼ªwŸÿÒY¿¯\ê™™ïåm“s˜Í„ÏÍ4ÞŠù—.ý%9€\ h1Ö¾d(n§H:ÂSLmòY‰Žk™1þ´4pÇú»e¹ð‚ØíC§Ò(DÂߤ×’<ùŵ¿iÀ°èêu ¼¡‰D¿Rã~øEµ“õS<ð¤ÓO~ ü§,Äb$ÄÂŒ7¬ÀaB&¤:‰OË|£§É *o[È7)õžÈ©mAf¥,nÚ¬|÷º¥y——þ.#ê4Ÿß°/¾CÀ³ÏÕyN*oGJËs +¯­yR9#VB„çí¾i„dÙÚkˆI,3n¬œš>+%#’/O}wÙæ0ñN …2S/,AMÐÓ +î­q‡7˜ò°Ýúe†›êì<>gêí’ÍÔ^™½kpò¸H¹Ê½àÑUA /…¥ù»¥xß0O›uÀÜâ1ÔD£ºÂZ3qاG$È)¸GTݹ9‡€¯ù ƧÎ('yN*`eIª¢ïN!¤§îFevðj;;5(?m} ÷Ks‘étyJãÓõ„ ©@);e®&ŒÇ¨L' 22Àά‹ºè¿ïûªæRlWõºO¯Æ!4Á(žÝ=mÄaä)+¹/»´8î \V²;7ŽFþ3ߺTuwbì’}r·ûõ‡wE#È–WÔÑòÈŒ 4“RSSZ} §‰à©—ž6â ¹Yš ¹Ó$‰:|¢459 øæWoÊpHäÕ#°sc7©où°9Wv ù8é·2&B®&R´^§Kò”VEswžP­"߉‰¡ÉJ¥^Á8÷!·Ðå:a2ê÷S\½æMÛùl$øÛüÿ5úáÓ'VÀ¢³ 8ØDki„ ´Ÿ?À+Ù ¯{KDd·Š2| {«²ÝðÅ䲧b¹RƒM€hmB1ü”@b¾ëãIÒ}Ä;i2Ϻ<)6+=ë+ ÃQTbÜ…YC »M^ Íe¸ . 3û“qSbÍoÖñìšõtÀý_À顤‘z:ßB‡º»ŒO_9Ûô—ŽôÖw v¥{PÐézÝâ»UO. +ú9Ž½«wÕÖ|Ô©>ìÅ‘½‰&ìXãš +âxË(pȲbqïéKfÊ-w’õ‹!ªä‰¢ÉœvðN‹ ¢šZï¹Wóîý}ŠUå„<*íkêÊA×"ÚNƒ½!APøzäŠÛêŸç.‡‰4vät% Eœ¥gCSÓR‰€½ ÝwSË#P»+ØW.Qˆ(u¯ ö‘à;¾V 6jJ¬Ñ¼œÿ° T›µÛoœ Érh—;W=øoc=yy5k}´ [ê÷š&ë´ÍfsúaŽŠ+bh‘¶è¬ÞÍ­eMK¨Ö,×ߢg¿¤·¶ï.ÌÛÏ?Ô¯\7´zžäðÎ þ +EÇϬb‡z:„ž¸Û L¿¶Ìü®…Ég®=¦¯¢Ùïó ªÏ¼føÝ>;®4s½Ä Œ‘!ƒH,ØJ_ÕhÌ ,O‡\‹âdk§c¤CºQJûß¼™Ä1Â8“&±ª‡ckþ–ËØÂç#Ù&1\ŽõDm›žDeÜöJ^H5²¨$ÁÖX,Þþ$/1ÂîCB:¨Ìþ¬JƯð&V=K›dW” Y”܆OkF›ãkØ;«²’ᦔ'ù¢GßÉœiôø¼µ&|î¯ÃéçïÒ+e<Å,Î@ôA=jùi¢wQãšÚÃè]¡éëXf“t C,72€;ÂiÃËuÄiQOjŒ-J"!l{ç"·A˜¨¦‡ië=v1™_áÔW¯¡¿"|xQiÖ-,–ócPu…Hˆm=¡ÈüoúÊ“Õ°,_@ò”òât°ÞèÒµ´t€C6â³ÚJ“Yg͈Çtóu‚¿’ƒöe^~8*FqŠSq±¬_QÆÀ&%°'£[§€<ë³Ü/­Š°¬éT¿ú·dë³Åb¬É"CZoÿQ²WÊoÛmÇ1Hk”¸±ØÏs(aû¦­Z$ûŸŽœ*ãjgΙNz¨sÆ ×2üHññÖt™©ðVfè(3u>NéÛ1y­&w©æN5Bœ?†?²d?üíÇïï°à"CüI ‹sÏÐVì>… e‡¢ó­e÷œÔ =›D¸ o£›rïº ?X¤Ï]žq$î¦ò<‹uîþ7Ÿ+ѽ?‡p2 |õ…9D9‹%¿*ûãåQ,_£"™iòÒbâýÊîìÅi%x6JT*øÎ÷:·e‚Ri>Ö‘F[`øÝs«EÉ‚kÕVq*£©×:ãâ7˜¬–±HP÷;]Çnê<Òë Ò 9¯mýîjé9Ón=× +ÍÍ´{ÿ¨ÄïgÑ8¼³­Glo™é¥‹¨Z±CoÜ•w]t¥YÆÓß‚»ÐC7âƒþ#tb⟗`Téêú#¨ø`¬¾3I¾Rò×,‰Ý‰·{(:eè ßi&Õ,ÍÇ}æ«ÀÚi íB5Ftá56F¬éµ•F;‡_éOø±3z‘å‘T°Ló ö®úýXGº&¿]6Ôú|5IvtW°Ýw@uˆŽÖ5õ‡Ž,”+ËgàòÈ¢4ÿë‹„¶Ü™çXÁûDôÎEænùyËÿGÃ+˜X~N ’Ô"gpF7ÅBWÆ´hçI^tÔ{³<v,ÊÆc.DèÝÛ[¤&Îó¾} 2B)¢Ù¿ó•‡¨ Rງx}føW\x†º³øJ©î˜vU„IΜϹô6¾2¿0ûþããÞÏ®ÐÔ´þ êÑïý×–'][åæá~œ¢«Q€Y†Ã¶/'Mž é‰7üŽTêóÚ¤°srŸ7¨‰gÃJÝ¿EÏöª³ i-梗ñqOiþ·ȺÖ2cÍÄ»°OPÕ‹XÉ6 +ÿ¸ÖÏ“'b¹Î QªRÂ0ûQ»æËÜk­ä®{-¶t½mæ¹'ñ7ô5¼xÓ>åán9åžqH± èQ~¸|ÀIbo¥òD‡±ËbâÛýÆ[tqäyÁQ\㦈ÞöRüHsˆùÓ'Ù|Xám,“Þ@‘†IóI8ðb•É·°¹*WöÑDË®¡žn*ÇÝ9v|‹*¼T:†TŽ©zh +g‡±ö|’í¾;\fóG>¥ì6[ñÎ S¹Ñ1Ãî,em +h½¼> Þ¡Im”Åþ­4yvÿiôy®:rˆÐH”>—ânø=\Z7Ž<ÏÛ^ùµG Oî#Ž%¿ûoQ‚1 íL¥ÞÅéu™Úª­Q{D)w¼¨ÏA]Ê£÷F=õ’·VoV¸…ù!‰´iâ¼HËb[žÍ¿ïL?ÜȺ¨Å¹ß;÷¦&ºü'jóýšJ)íq¼ +7c±Ÿ"W$[ãóÊh¹FÎ6âaç²ý‘þZßJ§§JLº´‘ p»[O^­¬ÍûâýA‹ZôD°¨7ºQÊAÂxìÔ÷6˜q0kï3ÿ¾éBþ”C½MU¹X¬rœTq”Êa²%ªYÃéÏ}Íøœ e¬;ÏŸðÑg³KhY‚=ß¹äïæB-sZë=ö^†]Ò&Z½Èت2ÆŠ@¯êä?FML StÀ¨¼µ÷î¬$¨ßŸ ²•W$ +é£MxÂÚÂÎt &äOÕ’ìåuÝH îÚ¹Ùºþ9Îð°‹¢:j²Ã ÃIܾCWè4ŠUZp +¹ƒÔÓòPÆROlŠ­µÑ˜ÃíŒ'`?J2â_yo÷äNþ,-‘_é×ÎãEQ9w)~ýÜ.qZgÜÄ¥p÷v¦Ë×” ±õm„kO~~ì8Ký|æÓÙ*Öá©ã˜ÆŒ`NN™ZŒ$ †X—;çê´k«žï\iíìLë•Ä ‰\±ê=‚H½¸öÈE!¹çƒV¶±ÀÕe2"¼ñgÃÀvc|¢íéö|Ìâîås‡«[ähwç·1è¬í1óCêö¬l&’E%XшËÎ.Ý-t¥ œOA¾S³Œg¼œ:ßÅj’뤉F»;¿mšZƒä\ë`qyÒ¹7ló p_Tk³ß àµXüÛx1Käjƒaç&/{Xo™û&éΨåŠF-A™iº\ÀC)óf¾À¾‰Nœ>½ hƒÊÂõJ‡&íÄõ©ÔU8Ô¬yóiÞÿ2³€c™³:o–¼Ÿî¨Å|âo쑘÷rÜͶã3|"à¨ÁݬÞOŽD׺ÕE·àN;saC­ô%cqnj,4P`N-s½R~; ’ zJk`Î5ªrçûœMÇq8 +$&­>w6ÅÚhFóÄ&ã„?âÍÄqëÉцkɉý€/B¢@-nâÜT´:v¾PæÑâvNnéæù_f‹jUÖ¼žb×ÖVÖ½qÓŒi1°ÿMÄ}_%¥øùòFFsšë$`(H˜.µ (!šï•‚¦„„ œƒ… +‡ƒ·ŸåʯeI† æøÔoÚC|“ê^»Èbž{“zKû€'ªí +>t)ˆìC¤äôdT·Þ™sBžîÌŒl‚ºÓÈÈ’ÁðÙ, +Ø(—§VõÛƒ +Ð Óƒ¡,ËþhI/ÌÝó4ÀÍtu0¬“ßFd½‘ZÊòÏÑ^ À çš2ìèö Òøð%ÙÍNÅç(¾ÓÚˆL£úbåi‘YͨÙÄ£)5Gç”PK18U44¶Õs‚÷F‡w÷[¥K +HI˜).—®ÕH +м‰¡º6,É(³ûæ +¬zèZÅ<hQ£ß#Œr4ÃíÙºæ ò=AÌf\-MzÁ ”fˆ+Þ«MGIIʳ1iÒ”fäÕ¼¼VH-9]0aÊpÀ ÈP˜ˆÊ”•pÙõ᧻_:»S²Ëºê,ìbûŽçIÕ3%VÊp’/ò æ„N!¨T1¿÷ö™Ánƒ ê]ï䬜‡"{8̨‡3C>pYR79O/‚å  }L½Mɧ=ììßhàŽÔ@Õw +R’Fg᪬ÂËGø•²> +îá}ÁÀ`Û{Ûq[hžJóæ Ճ˶&/€ÆTÆÕ>ÊiµžÁÒvV 2½Ð\9‘x„0þ­ôUÁÒ‹ÂPSw[N•i…`7 §K‘¥êÝê)57ö”¾a”—™x<àŒá‰1“pƒó}'56‰Dª+^x_ã>|ç«Ç—`»ìÞ–Bdß<ªቆ¬®{±è§§ç£™fh>5îìçb0 +uÓˆY£zW¸›T¸‹QsÐ6ŠcÇ—‚€¢y¦·à^»Ñ.±fa%.ƒAlñÌ €–ƒ9–^\…½É¾*eLÚ&Môëšá:]€yè‡Êxvo{l#Þ,>«¡£ßãÅKö…#x®¹ieÇ+¤…{äÚG30æMè +ƒöפf´ª¹ìT\*¯c<ÅH bê¸ëâ_¤l¿¸„÷èôx¡dʪ8;œäÏV•T ¦§ ¼ê;;P:l ’ñ„ïÖÆ?£(ø3ÿgçL_òmgÇÈ5ê•ÏáYœÐÜt:—í!1IQ¦%îØÎ{5cB¸(Þj÷’›ð0¨¦£—TRÜD4lôŠ^b³ŒÌ:Á ƒ*Ÿ‚¸íDÐN”Á +÷œ¶ RáõœMsX€s†åá·FUküE¾˜H_@·HÕDà[´Ûðahо!?Æðdo·Ûoƒœ°óx0Xos—VfœÇTñ’µ(ù¹ÕÎU{É«;NmG0_‡…Sj”6’A%èÓ8{Lú&5¸8;0œ !Fb!±ƒp« bz¶+&(}V·ÝW J9 (%/sßNÎ:†¶õ´Xµ‰´u¦¦¾!µ¾FÿÃÍ¿CD|ôc‚v¿ÏDT/AË•?dÙ9w€I oÇ6µé¦»ôhŒÐXäÒîRÙ9]Mõ-*^þ¦Á,PK›®f«ñŽIÈZÝóWa£ŸÌÒ=ÓN('¹í:Í´*0npîv ^s¯¼ü6—e–` +#>ùÎ@ž³(<Ïðµª1ƒ` ¨?$¢ØÊ6ç‹ýûK{£å$®K¼ñ[v¢ô(-ñª ʼYúáÅØ=ñS1»Kn@Í÷"XnórB¿8HUª©‹õxXNaÅT„A-ºƒc§m w$ë—øÔÌŒ ÛÔµ™AÆôÚBƒ•5Pu{ÐLP Tû(6”Ù¦Âo\ËKDZ©8!wÏ¡.®;ìPÌ(‚`Ú°›r*ü9Ö›$£WeéØÄYìKÑplJ> ›Ö F~WµÀy…D¯ï 4Á:ˆŠ¼&,Ã3 +£ §;¿P=J°]LJÚæЕqŽí¼Ï³Ød,Ô†9¾Ã‡é6HR¨H—1¯uÝ 1žËyÉnyð?E. %‘¡ð,¡º|viÄú]ú ÑkÐÔS²'eü®ÞDYŒÉ˜JX[(d!S…Ä¥·”îÆ@ærlÊ`óƒ`~@«Ë Yu´«‡p4³h·Ðyr\ü&SÛšÑ7µPfrºh-Ó;ˆ_–29g\šæ¥ n’øyòÜM$!;Ômÿžx¯°ê¤áŒ‚xƒß %¿™™Ÿ4Ío±‡h…Ó€Bñ Ë@aÊÄf}q +æ©Î»ĺÝÐ:Øêižå¾ÃëDbU®g S$¬€pù˜ò‚Š/š€IüŸŸ@ú¼K& o”‡‹à#ð:È)z…‚ >\ЬgëQ̪Ó@çùå›'Õ<‰wΕYÔ™ÝÝmÆÏ \—ƒ(YÌÉèVJ•)¬¨+èüêdIázFT +÷H&À-2Œ:TÜ”PðÐaŽxé€t±<ëhLcŽͺ=öá= ,Ut¹{ó×Þ]ûr†§×Wã,Áï¿~y›¥' J ;;bx]Äis¢óvŠ3®˜9Ifðÿñøk”î^˜#6ÌgÅ„ù©åì8£¼wó“û~’Ĺ?ÄD¾Ë²dÀ 6-ûìø›/B{l¶hâùøÔ$(mñ*ñËŸâÅÖK ´ÅªýGÙ, º7má,.€t§'~9i¡O¢4ˆ¶ÃÞMâTAi‹åøý‰ŸY’娡O[¤i”+ `Ê(O—€óÔ¿ŠÒÒOØ0Ã\z¬/ªÊ©U®±1„Zªk¾èP:ð±\‚ àŸÆtøÑn²óNÇšU]б&|<ø¨Cµ{<”j¥:U×4úi|6~.o;Ë™pÑ£÷{h•X®»ÓIsi‘œuÓ’[òÛ“Üñ^cˆ5zòJÌØl‰m€N6‘ô'¬¦žpΟ?9¾ô‹I„~`ç>†|¾uÿ¼}Gè莧Öa×dò×úrp“ìh;PÁìtâ‡Ù÷GØtÏIa &ñ–Ù7ËìO‰ ÇæØN–‰]¥kž‹ß4Ó{x¾aÌØ~–§Q^,ð5Í®$ë¹µë1¶®˜/_Wñ>å·›¾ðvëëºíοߴGû~ëÖ¦‚Âx8öV9·êëþ $/ù¾ƒùþúñežMÙéijï éÂ91µ2]­åë¥Ë­ä[èç—¢GݪCÇÉ,ŠøMý8 *© ßZI Š<2ϬºqnÙ$Ë4—“tEN(üú¯$£’üö¢kíçÑónŸ`ÔŸ%åç†n;¯¦I¥ÛêL‰¶Uy;\;oºÌÛ p]Ù0毽7Þ—ƒ4ÜËK:®ÀŽ#_ÞféIhx”ììˆáa]qâ'QYFdÄɘvÆO÷Óf‹2§×ËgÌŇ¢3?Ú9 ¾tßOâ2j b)BdA_ÊAüƒÁý£‡l…¸9þ1JèKœÒç¦^çQ”V3ZcfÿÚWzkb˜Ì¢êøæÄ‘¥¥/æú:Çi$³>™½¿ #G»º¦±»žÆáOe-¡\.÷ÚÕç! †wkª5¯…ri³6ŠW„#OmbL¾p µ¯.EÇ€Bó+ãÈht×Ð$¤u^Ÿ71|9Æi¾„÷)*¦®ˆi…T1¥ˆš2^ð=ò³ +™Ù:[þjï4ãáÊרUá\Mn5x‰é¼ª]‡`ßg(r…²U£tåV‰~¡wy‡Hƒ‹ÿ8Qn" ^»«7Œ««²³†kK1¦(q“×ù0ëÝLÞ̇°°áE…ãíÒ$wV+ ¯F«è¸òÖªJ6i5‘–ÌW_‘°îjqu6¯Ë”~ìFyÕB¶ÖŒ!W@[X!ÀhƒM`ôÝj™kIpÃ\«‚V•&8ÍV&8­ˆšÊ·a¯Y™à¶ìU§‹¾hÆÁ檥Í&Ø‘ùªNÝe¼ÑÙTí.ý®#æT8:w Öý7N#?Ò…'P%Ó{L»Îkklx•0v–àå^êŽNpÛLƒ°Y{ÛEüj[šhüŸûªmn7Âýù˜é´£Ìôh‘²$;ùd[IêÖ‰uV®içî&‘Ì +XT¢üú.ø +RMîꤹ™äD3=h4Ã2{kš—}rÚª¡FŸ• Çì“Yáä"+à$+´60çHó‡-i~ƒ¦'Ukh·áØ”wQ>ꛬ?€ÀïeˆõôX8sýl ‹¿âÓ!ü™ä_/+¢o~¼¨ ÐÈQþqZñ¢þx™œÔÕœý®ª{b¼Q‘ÙÔx£ê ÆAÕ‚7ê¾<®¯ç|3¿®†Êrv¬§IsÌ|ö.ÉÒì¤Iï`|4ÇägƒE(âå\÷lð&›û3,‰^ßèÏæ¶øÇXðU>ÆÏÞLzïÌÛ¡¥?Ìûí’S-æztM ºŽŒ©ƒÊSý/ ­è5¨ ûI¨~AH©!\j·k>Å‘zt±¨>­UþØ÷FÁ…ßgÕ#‰×.|^!ŸØ.tv͹Š%W0Ü‘•º1lŽ.¦ÞùÚÆ{ žŠ\ó”EOÞÿ¡K=µ +[¢pãÐ×êqù•zü1âaº%Ì©p °þ2,3øêÖ÷Qy†bÉ)VÁT“mjeË +¾º½Do‰|DX*"â/Xz p1Õÿk¿a°‡òT¡µài’CGÆðûT%°‹ŸŒÇ£qá=} µäXD(ä” äƒþ¥—¦f2Á€p*Ä’ñâ(r-H¸qp\ÇÆó½ÐSÄ +9*[!ýF æ‚H"v½'ŸzÅ +/c«}µB$`geÐ`¼Mt‡Ù:Åk‚æ<)C7øù-‰âtkÄçWÈÅŸXBú•)Ù%í5d™"Œˆ\%…ÞU¹Ú&òѤLGQ+½¶XnZ†Ê„·sÓXÉ–Ù%±W¸½ø¾U¹\Ý¢«Tñ*ÉA^_¢'`€Œ·)ÅunúGãr‘!ZUŽ€² 1#H*Á7%Å'e ùŽˆü¬Š çÆë…“u~L±3º#;BK´K)’#§uª5"óÍ;*Hº%¦˜…íÙºÒ8gè…ò3$é¼&»ÃÞln BÿŠ$Ñ1ÕÉ"³ó®FXaAg"ng&Ò<~¯w7ýj•yš½¡ly IJeÕ—ú4ä""Ñ¡ºèìWcßÌÆë0¡˜eös+b­"£|éh)ÓÛöF»ù¦póƒéf3–4Oð³äA¼nѤýýFà(ÖeŒYTô5÷[ã£Wó…ÍœFGÈß½Óy%òw3òj¿Û$:ýÂq'b_˜Ågú*‡9;+‡÷{«ij~«ßÖžä+)î¬Ï’Dxa*aÀl²T‘NiQ¿Ú®Ú8c½˜^E™Óú`FV`Y„–{41Äa«h߯'˜a×aK“ã S‘.÷d l<(”µëöo¾lêv( +o ·mŠ‰ÈãbíY•/0»"jã( ¼HJY½¸Eó ”Ðp'^»Õwat§?®8xò€´˜ž¢ô”A´uÝ^À"àŽ”…N„¡1cåèY7¼T ‡[ƒ8?y } Khd°—Eè—ÁâÃýü—çhØ51‰à«˜¶ö F3Kkª[Æj‹õÈÙÞñ:i²È]é9\Ú‘ÕT´¤8ÜØ©„ ÝIqÏó\Aâ=rñ¥9Ü·QÙfÊ:½ !ÅI?G8 ¯fE˜:#§œÊI-Ë–:Œ.w d*V§zÁ¸³Ú5Ú½ž{š=´[õ–¿˜­,5›ÁôP+LO rÛôR´ÄBZœ[³?_­€zWʬxâŒñ ½äª*áÉQø‹ljí®´vÀZ; M­u¿ˆ&bÅ™­p³¾´Õl í!†žCT³¡dí©‹ÊE›\²qº ¹v¡!Ýrª·Øc}ª`«lðïiêíl¡y×(V)¿œÑ3ì=bGÂíÞÆ}5«G=4ç o–ê=mJŒ’lWhl‹¯HEËÑ?#äê»x}­À×ÝÔåNéG™³R‰gê+9eV7ü‹cÛI¼J3ª³Y•æøòÜŽ4 +s:>ºåØVYÛr ‚¡œz÷œ¬³GDÆkÖÃИ÷fž²ÈÂÐ6¬8,óy"‡¾,6ß¹>BW%é·),ƒÍx˜na@ ¢ÒÝa¶NÁ‹hΓ´5ýgøWóº/ÇŒº¯9§"@W·è>Uº ¢EFµ­Ý¨÷”R!Zi#ÀË°  èìW$ä"‚9îø˜ù L‚etm) “›8щYèCÃ,,Bý°°#5]À˜Ù™¿ñå-tÿf\*Ž>ðP:9f*L¦›¡‡©‚AÑÏ…wÎ15܇ |]=“N†6¥G„ÁhYÔLpÑ(,x¨øÈ íèÀE‘ ©H׎ZƒGžÚ'–À[#í +Œ¼X6Z–‹#·Ž´ªáŸ¢Æ¨Pãü¤xÑÌ¡™£ë|Ké¬Zó­ÆbéâR-ÏïãÌ(5R¶Ï»;ƒœ›Ôše9éuþJÿ¤„…û“‚åV<ÁÚš4°™$u¦÷Årèé-v•²Ð¥†réÍö‰÷[â»xÞT¦%¾±N³uÔR¦/¬~;¬v* +NÓ>hiߧMÐÖ¦ï°#+íLz’7ƒ–7Ϥ;;Ço«= y=8…Pƒ&¯Wy³Ç¬—!‡î 9jRMŸ[­±‡îì8rwø¨¥Äqpƒµc OˆŒà1ÿqô½i¾Ñ€ª¸¿…&ûÞÐûÍÐ[6]îM{:£ßTÌΤ~³(\ÜyÞ2ø_„Rþ©×ÞÀ=ÕƒæKåô}Ë6è.–ʼ{†jÜåY-l>{ýñ†S.fü“x›Pò@$§i½ +v9®¼ú–3> +¾%õý¿Ç,*èØiàÕí¥Þ ’T¡,ñ—lóD§¼üFàý ß&\~L1­BÝ™åÅûÕJõt¾ê›k®ßZ_¢,AŠ0"/6in†GîúAÈULÉ- iªUÿ—ójïKUéŸÀï@ÛPï(ä&¦i¥’vµP1CÜçì÷Ó¿3ƒºËìüã™™uyžg­YXsCýO‘ÊÖòëò7%x|Nb@ w¦¡Zšþ™N·A»ü' †7 \{²=ÄË-ð¨Û¿uNµ­‘â´”?àCs3*åŠßꋨ±íƒY¿p‰ôû•;Ý“þ¾×ŸÑÓ5&ÓM=.¸ÝT÷Uÿ#†ËÔ x¶šcŒí¸@þ0_Gè_KLµ`äÿb¶>çWÛÛõPÒP—_#{jëš±^zÀË·8 »’XGÒs¾•i.¼–÷kq÷y×jë«é7=ëg’iécçû`¿(Æ`ÐÔ*2g£À“¿uÕ±ll¤ÌSÕúG]þ/Õ>5æ ¿¡õ±²ž;ßP¨Zó¹Žî­ëÃWâ6~®ëZ Ì-lú¨oèÿ|°¨¯ÃDŒQý>¯%¼« üŠ£ËÓõbd*Æ|S01y +‹)¶Ž9SÓ\Ð0 Ìd…ý3ÕMl¥ü†À)&jSVð5§Y#~0ÍR×  ‘v·B&Áï®±?Ö[Þ1 \6`ºvÍMÃÄŒG 8óŽš c̱  UÇ nV°¹ÖLY.rnU¬Öê†×0²ÆÄô͸ÞLêDg}÷Æ +[›ï&PXê¸L«¶±ô5ø…Ðf€È-œeÜÞ“ï9»á› ¯ɶb®–€Ìƒß;=˜W¼»±x¨aà ®÷Œÿ˜”·©‚@–K];VÂt° ôÖ#ؼ,ÓéB’7ã`Š‹%Ø92üÛø}Ù6mÅ~?b¤ê9Š©)öÏFš.¸«€ªÿ\üÜøbÚsôc ”‡yE"8ŠØÍlô¨Û$ÙûZ92û@‡;nÛÎœ}ø†úd¢ùÓ¦7˘æØÚ~ Â!ïkV¶Êè íø©ÐOçPH@^õ~WÍ?™¾¥ +> +Ám¼æözAY*¨í‡ ø¦GÜÉIð®ã]pò7W ÖÓõ[*¦>ßPìß?®™ÔÌY¨+é-íj¬i°w óU˜æí‡+×ó¢¨CÅÀÅBõ&òDTt´¬Ü'šùÈM’gíúbZš˜'Mé$‹ +†’Z’w—52\bïêL»P-µž£mÖ^«”T˵Ïñl¡ÎdVâLœ$2§ly˜Š³•ÄrÅ®®réΖ['övSÓá'—[)è=Á UL¥¢“¿\µ´à%¼D=Öqö“ naµV\£çL † ¯%±p:àgóè „‹ãLsô©±SêbLõ;OÏœ,¤ú_; î+½°•wé…-­R BLàk)V×Æ!%½šÇ/ŠŸ³óûÒ˜Ÿ:”zÌîÀñÕl냭T£×y% 'C ĆÖ8á“ç³S®— ›n ÷Š¶á³¡ÖÔb'&L ¯å +‡ç#x#X!z' ú:Îô›áiYU•wødµqkêzÎfÒ +e§o%cØÔø9^&mây͵z‘ÿ[nNó!œ,÷_XÎT£ ‚n—ÓÔâ™6(*½ç9[md‰÷RÖ³¨ŠÍUÀFEujÏh%CH+€ßl›Ž%:?§nn-œ•ð –¸(®/'2ÌP‚5$*}í©”…_YÆÄAB CJžÈÙ1!N ÿ~F&7Òìk­Lö%ÜÓJ%"6ôBÂ…!²‚¶„ðÌèW£€ž Fªlž*ƒÚ•»]HÔÞ\c¹‡\H÷>C0L-‘«zcg@WÊÚìzˆ˜ôönøâÆ ØÄ7½^ü²1º 7éô®x‹¯jp1T ä#9S9Yœâ8}õQS”è)OŽî:•Kñ•áä©êp·§j›“syÀ>G½<„Áí±vÿV]{¹ªÝ‘éðÝ7v>OÙ—[Â,qÜ«iO`V9#˜pyà2-‡ðÚk–èó…û¦ÄÚöô®Pjªˆ¡ó¢a“€¼x‚à­‹á>”»‰qßâä M…ð2A¯c’8Ïò™f±~êõÃ#‹-ËÎ/Nn:ë¿¡Üc2€û–ø{û¾“AûHÜ>Nëî….Å"Ë3aJvk£L%.ê¶Èè·t٠ąãUç^›Y¤¶JòM‚•z•3Jqkß%ô¼÷±¸ân†|[ÆÍóLöj¤I‚¶x@Íóêüœ¤¾mÒiÑü¥¦÷bá +ýF¬¿ã:pÕ©À“ÏŒKëïh÷÷©à)áÔ—¿FD¹p½‡H¹±²Æ‚±2HØ-“O·¹³pãÒϪ¬—" PÉÝ3(¯6A7›[§“gÐÇò`oÅëΫ¶àzoM1mÐ1É50Ž^’œÜ¶&l_n(Rí¼sÂ+D¦öìÁ±¬%¬ëœ!K÷€óõEM}ÿˆl ™+ncŽäìΕˆc.:ܹº’ôzv2Ei€žkê ¤Š£½Õ ûpý}ç¦ œŽƒK´nÕ²J;PØ÷§C“y´Êõ¬)FGƒ©À-²xuu™”*µâZ…¹L⦻]n³8?ofR«'¿>©óª/°ýQšº½lw³£á—œœÖrg§C![Z,|b€Ä³tF•ªÙDL¨z{Æ݉ÞÁ¿“ÜAJÜÑ“—•ëÔÿëjìSa! ¸¹åeåKe_³n._‚E»™¢ÊÙÂAû9ÃÍ7ŸºW©!܃ÞKÓä´k¢ºàŸE«{N7¼øuPÌ«Œ¯Z´"Á}R*e/Â[¯y¸"us‘ŽÈì±Pb8¨9`âèÙf>k€(5¯4&CDˆ‹XÅs€bì’çÙ³ôäÂïeJ‘ýÔ4ðY~sï]p&×O®ÀXyµü?÷Õ¶*D¿ a &ŠDE@@ñŽ5^¢‚4JŒÿŠ‹‚z’™‡yš7Ý«ª»÷®ª½¯¯v+Ÿ2—LwH )MàM2O„O››HNž|«ŽÁTê¾–O£,öà“>Upí’üû€É–1pä Œ`ÔgÍ’–Š@<‚ô tS$3Pd>e<¼Ø>àòB¨mîìî#¥XTO‹•‚™ög&ßúlwøö÷óVB•ÄÓåêFæ(d2­°g¾œÏCÕäëE¡aiW]m/¨¥Â.oê$3ï´xòí¥%¾|E¦ ÀgÊ£g¾Òk‚NvOÖAŒÒ¡®}¹ÒO!Ûoñ³wˆÿî~w–º8y+IvFZEÔ›‡vbÝ‚l–Ì¢£·Søýðw3ô£ºCüfÈc¨uµoñcŠ’¯ËâGoR\ Fß"Kз‘$¼ ;%ÉÖbg.æé#Wƶ7#/0Ù örD„ËòÈöSâ2.8 š)v +B­þX\¡ýbbÍæi«Ëwºð"Ð>.Óø,è\ú˜ƒFÙ~~Hš +‚趔°Åy—œ¾ƒüd*B¬Ãs¯å 8 Oÿ_ªßA)C«áB TžNXJ,õI]0ÂŽá¥tõ:l3~»2)ǤSÚiSÔ;h ú´,âá…äT íT­@9kœ ?Õ)G‘ ,}§8Bm³.ú¹èCb˜%çÊ‹Ðlpp—þ¶æÐ?ö¾Ykwe1¼8-xdß…0Ò8ΘR™íÏ„…K±t¡ºQàR\½h”-M&'fâ€`¼"U¤}C÷j–oT žx ™€ N¬UŒ^l7˜5DjÖkˆÑü´3ƒãûKîÿ:³£”jþÒÝÇ›h¡ûnž¨$ÇTú÷`WËØu¾]Z|.ù/œë(üÓ×,Êéöô±Âêp´A¦šÐT!ËøB¨i²OÎ8¦ó—%¼šuÐÀ3›{F?6>xµóÞ‡ô“߸œ¦;wg–Wñ÷0Ê™žù¡6òNÿ‰1ÉóNõT½þ`ÿ²zSÖ|9ÍÀßëà >$ÎÒZû«×w«z—Àd9ÜØ0²˜à(ÛEn#ju±d Ú¹Ðæðý˜f…2éÚeXÅW½‚2YÙï ¸°ÄËQŒ·1Ž®}ç¥÷òg‘WÚ!Ó&_loîÖbâ•áNïù¨%VÚxj½çÁ=\IÌ»´—ÏXc¶u¦Ø…𨥅ú15€§n$4B辉åñ΢ñ4³Ý3׌^ž-9úUö‡r:6 ‰:OÍî{bç¢}¢lÇöÍnw'ö$Í3ßsÀoÀˆÚ¾®Ør‚*é õ+jõÞKOv7Š +ƒ®ð M°Ï‘¨«¢~‡HÑ]õÈg8~(Í›ð•\r*5¹D h¨Ô=yHÔº¼ªpVR±ÛÃ瞸¯uä­Äˆ^³: Þ‹.>y㌣|£D@ ¼uä ³J\'ÅYs /œY˜kÞ‚…Õ‘ÏÎöËx²ÜºÿbÐcꃓŸvTøj)9Sžq/Äy¾ZG®Å„ÍÓ/Æ€`ézŸcöTòöjî> ØIf›»È›ì-ôb13o· b"F å¢F¼*¾'…M6˜Åð¨¹æèzscæë07, +Ò\§”Ÿ¿¡hÄ*ßR9ó–ÁCq¯8ZFÞ‰ñZ*ÖŠÝÂYî|!sëùP|ž·Bº‡ß?`xü`³ÀÆs endstream endobj 38 0 obj <> endobj xref 0 48 0000000003 65535 f +0000000016 00000 n +0000047811 00000 n +0000000004 00001 f +0000000020 00000 f +0000048074 00000 n +0000129852 00000 n +0000129924 00000 n +0000130307 00000 n +0000154490 00000 n +0000160188 00000 n +0000181053 00000 n +0000202901 00000 n +0000225104 00000 n +0000233450 00000 n +0000237159 00000 n +0000257233 00000 n +0000282454 00000 n +0000131230 00000 n +0000144831 00000 n +0000000000 00001 f +0000048049 00000 n +0000047862 00000 n +0000047933 00000 n +0000047964 00000 n +0000049662 00000 n +0000050302 00000 n +0000057587 00000 n +0000049775 00000 n +0000100574 00000 n +0000049969 00000 n +0000059456 00000 n +0000050678 00000 n +0000050806 00000 n +0000051129 00000 n +0000051287 00000 n +0000051532 00000 n +0000048483 00000 n +0000300546 00000 n +0000057847 00000 n +0000100850 00000 n +0000059723 00000 n +0000051601 00000 n +0000050831 00000 n +0000129291 00000 n +0000129339 00000 n +0000049242 00000 n +0000000148 00000 n +trailer <<252CDCE32BEB48F1A4AC1838A61EFD15>]>> startxref 300713 %%EOF \ No newline at end of file diff --git a/doc/pub/BM/html/figures/RBM.png b/doc/pub/BM/html/figures/RBM.png new file mode 100644 index 0000000000000000000000000000000000000000..c8dbcdc3b0b15ee98def29ca6a95da0c7ac70322 GIT binary patch literal 38001 zcmd42g;!lY^FIodQrz9$T}yG-gS)%CySr0d3mn|to#GA$_u}quKc4q{|A@O+cJi4? zW-^nk?6qT&3UU$%u(+^bU|&pEFAj`@q^9#%D*3-xeAJr%3;b}A0*L+i2Z;hhMKos!0S^Yo2jQin;i6>dPVC@h zZ)Ry@O6=n4U`p&^>i`Dkk$v8!>9#voZ}Uo2^%Fph^1b(<_N~LXb9(8`dcGHeTrnv_ z=CHVVNX}A<7&$b&t*uRh`0exa@DuLyR^T(~WW5kM{#uFazSu^FKK%EoimkM&Q`vZR z_HJ4ERNBCfLmBP6pE_#KsO!WrI@hs^y%MMiuo4Oay=oCn)NPhH%ZY8d4q)Hsmq(>} z#0|GczP8i6XmDSb)>m=9xpSvF6Fh74hL$U7E49r;wc39=^sJwItc&;X%oMnMiZ&_M z$ayN;24%XHy;r1{JFlxzHJPCq)4z#-<{D#I``Z3vR9mQ9Ei5>7_fapoBYx*Q5r1i( zRj<;wx$o=xVEAY$<7CoT<+Xn}pFXiGQac{gDN4SZxdJUc#9R4To)p#q)+@ZzRzEAI zni$KcE`)S-?`jG3U!~*{RntWlGJm(r8T`Ylks{2nzh)c&?m3<=<47Tc73zOR! z?VVg7i{Ax|^rP%umBN*IPq$)iPD=Ge{Y30Rq`7Ao_-j<%el};|>!P3Pufm_9_9E>J z!}(dw`aAbz>`$BlA2qhzX_?oz^j2}8P1~F{mT#my<&76{nI9JnLu-$%4_%=<&;3ie zJ!|rZX6kTO2-NHJ7OQ(h42%<8ZFnZk&zGF$!O4Am0Xy_-t;tF4+wlfA^-T!PoDZc} zFZ6sSC0h%Z)A;@-NBr~Ay)8Hm_tu9Gk&5HZHUPnq^-t0)Zg)rds$W(?93xhWt{oaZ z@tWy}ZQgCV7EGS&c5ZY|(&vxRVX!29{b)0V{-JT)$NIja@lziy?{oseg?zS__PBQL zIToX329+I3CmYT(b_g(ZT65m>%qyro#dCfJYg@RuXqYW#+e2s!b2gkDd0H5x0Af2}l6GnSq>&DyE`5&E_Ev4zYs9-r_(H2E&|twT-tY2r)lpTty?bgcp_w$qs!}{J zxlsI4^u76`x+L;7+v-}3A60oX3Y{=1q2gDBkR6Y{>Q#gBp}Loms^}c9@10=B#q?(L z`JS=xOz;{5D>WN$vUG*ODCa#LU+XKpBSn6Z-)a*PgH<b6JOazPSZpB)%{Knme+B}}-Q zg=VcSQ;ULmfDL;#)7H+)glH12E+I$4x(86Ld1ILcfjgSkA|Gbvg6OPqwg=|&QSZ^W zFV=(!z*B?cz3Sz}+LJFYG%}N6Gp)2Q#eba>?_5WJY6ubgx#{-qRndqbZ?>>1bbIFefe%%xqJSIk4uH|=Azbq{fkH!Alxg>o#h0xzH)*-2g4x6g`^m5=6b@{irt$#wF1&A6ovAS{TyrPSLWj~(Zvib5O+Bx9!784L@zqX zv5}~wg-%Drh2y4il{v35fjTJ-Qdhhni<4Q(72XuU)f|uy9Y<+*QUmj$zHvX!qqIOZ zVyW0TfrFrY!QdF&|6Q?2?5-iSx!S97DUKH`bHIch{>!DwAl9JS5tH(Ch=xIN!Y%7QORo~OA6?98K1uFBP7OYznjrx=BJPY4m4tP-7f05j(=dd5 zg%)S(@3b+F!3I2 zbqSC03FfFG&uud-G~+AI%gwPY zSDO=e>wmMzvCuMgZnOC4^!S^J$O;=IdQVR|S*TSe<^tZQ(MDpmU>=~+D3cw*8L`zM z3hmaUqG{g&@46yEj{sYR=W>9Hkf+v}T6dU}NnL?rl@=pNd5LUE^!}K+sML37dW&0C zHIWjE=ZMd0z)cwdfLBVhN;GNs2qC86Vz@Fbs5o@q1~kV0$fcB#aK~!Kk%EoA_a29D8I3x*0*0u;BVN`(KwfjuGKQ z&XbsMfHBfvPWa9l&E3C^r@K)LCh}&zR+_=`^G;zN_)Vgag;qAPiS;*JM;=A2Bx;{` zKF`pA#9#Td+`ZC}1I4g}9&HS276c+2E-1e*o>*JoF;PjD%`-+Kt+dK5%JZ#F`T8A; zMRqIP#t2{%&-wPn<`wxX`c9+WM$&41_BK!{EoK@2y)L9T&`SrBLQ~I!NKpmK1HHLP z7D;1Gx?H&nb)0~B93C8d#DE~`j77V;T*yPd*^7X*qUU|oNIZ+q#jIREqx-`lh5`Jz zS&slwrUBK{d+(Lsgs4K1y%LV%9#)fhb+}by)Wr|gVo6)JyEv?5@$kNrl{5wD@HYiC zqVsFWcE{#AH#l7B2?KkF3zRMqNeXBty;49qVxaMac53kmH5bUL1mLaNv_09{pnQp! zAu0j}9{0_~ib;#2yG6!#y9rNB0#|l}#hA+*8``PI>>8DBrT)!FWnfYu?nR2BPSbU0i<8I-WdiA8;_ody zM@gv^M-74ZScgQY(^z>HF3=KIz8a}eDWF)viG{Arwn{s`P|5N}@I zG7;6k4-yR?id;|lr)C7sO%;aVpM@yu=TC3_KJ42NPR&_cYe50i+!&? z{tOh3-8=l4i|eAlKtyM+m&9u`r@;?T>)5*!ahU=w_Fo zTPI8^5qR-&lCAko>l)hqZN@iSCtesjtgLUlopC7EB{1O5Nfj)M<^t*u@g) z5*t?Ckki>fDBd7PFIuf;v0xSubYcavSb!NPf{P0L@}4cPxq9uV1#BB4(1C*oFsIfX zRG^vewxE(2DhKa*^ap7p)k3{>WvXLXtYhM%q8cV2ScN_d&5;4X67!NMt6RwdNGh^i zr1ezpFi@sGmVJs=aTDF~niJ z5=FG)O-hSZLWM@S=vNC_+latdBnlg1{To*#sn0_a%WC7-cmpAZL$+!VQvSi~1a!H+ z(FI_5ooqp3EmDb$J6{nm|AY^a9B?Ukh(I>T1~Hy-{bFI{9nsQY-gLvTdiidXfUQW8 zL4Z1*`%`eY|9k<_7n$@=hm}~ErI{KInph*)-|rRy2PQvg3{*5@tzb&Svq@-!2cu=d zaLAPnsUl0)xcdkiXd22OCWG0~O3@`Wm&;-jOL5DUNnT2KMTwkFpWG0{uzajYAwy3y zblNNBTQ{JhFfbw2Q78k$k4QuK|D|V8Yo9OurK?3j-@Rq0zOfE<5j2gbu#TOW$X#GO zHNd1hRg?HP3hf1FH`YgQFTZD@nx;n`#On%GPh1X)`Bs7C@6-qbi?Ra&i+x3!Mc1WJ zo6d}A@s7enTbJWeg-t4kME|iD5FuYeg7m_bX=3^8K=Aw9c1^Xd7y~*!9PZfy=MxG% zvD-b{(UhXTY-_5#aC?nLEW=%54?VQ-T7uNMkrl7Eef{r+b$9|9k7QgHF?UMw6>ckD zC2Z1;xMCeWjpfAk+;v#2fGCK{5V?Zqa+2LcNya=GI(CEtO+f-b_1mgvx|mrU+~BOV zSvD*W86byr7S;$#tGtH?gjpdJH#o!5vK3&v?1;s%s8lKREi5|qYp$q{b}-Ux1*(in zrlPho?OmYQa15~yN)QaS!T2UyC0%~63M5t}6pS6mR7#OFH|Tc3l<1z^ZZ-~tzBRuo zopi_sk6(ydYM5Gl`3;7+qbHBiPPDx#mq)`6st?j7#aO8PV65qdoB>%v*?5V(iHFF? z6t4YdS^dURPZd@%@^i9G8n#xv7&|I!&Nw7Ztj=jEORqy%dt%30=`q zKc`1xUy2uMMLtRJj}66W*wmUB(SbWcUHA_%^+vPxIprSgW38!fL|hy_MZ>(Y=ner5 zfyR0-!k8jCHwJ_CYUuB)`(IYG5|}ypTF?aW#kuZ{1*%%-(8bX#E=k0=kbKE^CyEVW zJwF1QtKw4eDa`Pgy`b6Q|6$c}ffTE*Ao`XRpaJg1e#rbSuxHg*0K_2W&5WY$MjEQ0 zW~sDN0LMbvwn0M6_GWp^NRKE3sXNSHO1qA;iOq{PCdO8H8|sbcr6_9O(J!FYW>dd6 zRn~}>rM)Tl%aQzYkoM3rO{uU`6>#>!gvw8!_|CyR2d6uuW;yfIioos2T;6Fgu9r?7 z^I&K3DQn1!!06kQQhF3AQN&812(KETlTJi3)yuk{he2&QAZCD7f_zI{{JXTs9Ezz+|teFoY@-D-P@gTvL-4U~@oX|gA?Er*hWvx$!4EA~^D5V`7&+qH0rFvHsw zH(tgD>g)tHmV%@~Dm<0v4%zouqj9l*9-?R&8E-eK~*qJCrEL;3n%aSFadMlEB|X@JC1o}!G0Lwf-E zQldeoG^SR5Gm2|Mq0x6@BPFWEyoix7g#%D#LdQ`<-?8$T5JP|cABgUO#C-8kmeNFa zZLzxgcMIF3X(!0I01a=>do~A=vv`$MbRh_1@5AbcFwpvBA1d42%D?ZZhKs!w(S;*4 zC<&;A#lZ|~Qq2FNfd>)?`bkq#qKyU(be~fb=5!?r;~5U;^{y7gyXBg|XnI+2wa|xf zpQF~r#aLVstae-?Jz#7W`<*LY#lNNc?=3OOz!g$f(vsoK{TIB?FtA3wGGc%!#qm8Iwv2v}jIX*Pn&Aut8|i0vlJ)>S7UI?N-SIMnPj{vh9Fs zVEj1=iyZn1E~}0tY3VyX6`fz{ID)mHP;9#z2_2f!4qv4!MM-WM)3WjsshPHe^omwq z>f9)IFTt#FENW>H`q++TeRN%I;1bUUZA|k;9DFn*4L{=hxZJX712{Y43d$;yM~GJ{ z?l?x}XG1VAig}yt8{A5@Nj1}hs2N>z6D-EhFeTF5*xteHvS9dw>eG5#DV{sQSkz5K z)dCYb**c9-w07b00U3IIonBL_UFg!fZu$K|l;@>EXO(Tj&Y#VVohgP8Qt%PVn>;Vn z3>VvRdE;%SwmCLR_1E=QR@2$8+6do-(J|->sUs|zssi$_`@4krrp;avSXg7!b3IRM z39D5k)sp6B4Lasd1al>UWK4J@4hDqfz|3p~saS{%!DRgJEap_o!Q`oB#?Cbyl(__` zSU;$WXwP6sbSe^l;vU9k2PBG?T#|+K$MKQATo`Z(euKw(?<<8T;n+Uqp&iRD+9=K~ z{pm7`)jt#_jg8>pNn7o3&{r&PpBO0OUApKihrYucqE0*PK`!Qr2_pw;;DEzGmNH9I51e=YB?M2>!mEM2p$Z^tqU4n1Wr{R` zEbbC1;jCQwP387;J=Oq2Oi4gU#`LPXh*S<%)R54GKV&JT2pi<47B)i~wc5~xDYnri z?5S=)Fir6G^AZ#if*NDnFbC6Q;2^gPyk3l}unuae3PdtjNKcl-D>D_)G-|R^3l!tf zq+h~f^=^J%eXde!p=O*85}~Ch*I>9cFB;t^e1TqfS^K45ufjo7mP0)tb_Q?0I#aB+ zZc`#U)efFfV5&x9lJ=(v=&MzCh#cZcqs;2#JP9(k(I&wjMV#n;Sdg3D8MBENCHqJQ!6_kA64e;#b3VCubxO;(pAi@`lesTw*=*6NG`MgykNHwmZTs zaI*jl*LvL+n%8eQJ$pe@V-B^5Av<8 zNzk2s=@T#^uK+t)xJ!~i6i^<=;{3Y_mi*fu8snmkyF!F@(0y-ffWK%>;YGCab1IIB z;-D$9MQZ!`;K!|Zv@YGKYobU4&YPOpHb`pZ0>vDk zzw05*wtK(@+u;7wwm^z>FG1u$OU=*QtBftLVg^~se|!dBqtLeT#`qw(fg(G>GaPXazrfAN`KHk$m%s0XCaVK(dEXFAB@6EY>vu> zd0Hb16KE=?u_az%1lEIOMhJJ3b^Q)7ZKQ-m%sSD=;j(%4&riOcN*H~pTq@j+MdkA+ zvsP3w3M#EHA8LKnr7P)#G;tII%WQNngXyRPnOQIKgD+={X0ZaH@37HMVD&!90pM=h z{c^!i9?Mxc$q>7xir6UFF7?l%p;MX zg!m^5yIOju8x<2`ws`0*vxE3tQR>6~DYIMP^O`FhL`Y}C;L<{{z%roXuel_7428;g z&qci<`rc^imMbJ{q0gM30N8#qRJbdki6!r}Qb}i}E6(jg*R1Eo1~`VgzAopEZX`h0%4MY|D$B8f zbyPv8TXej?`1HmjOGb1@4d}ZGfRZ8l%yd-1Z&E6L3W3Z#BObRJ&) zA!k1`HtaZk$(c&=txjB&#tdQENn~#Rh)SQ(=+xXKlr%AOQw%Um?<6N2H=^SF8d%}; zfSF?2`4^>{oF~B`|6oo;=1HOrk}%-h+_usR?+PIjAuqi;L{2WiuFvJSVk__fhQ!b< zR;Ge>VrT(-FUATR0{1525$Vnc;$~@Kkp;!4@VzSwaN$KnSX3|J&wyeuU&dc$+hi}% z70G>yRz@9k6*r+sO-D~G(6oI06f%wku!L+a-po(8f zVwfjtq>PzQyE&^Vux7HXJNQOSwoqj^Z!gBY6v|geg&VSslesH6$MAMSIe^S zZMN>A53(r3suiFKH<8IoSH&Z?kH`&~uavZk{t?WkRmBYLka%wpk2zDb_4^0W_-};B zigpT_Mvz+}<-liP(_6JZx;yE7Y@d+42c)V&E#|r=!7%0OoPvyYAb$dhaI{VBeVd$W zoJO_>ho!3XU6v0=6PB|GS&j!4b$aT^JEf(&*T2-biauqK({*(EgG-Z`>i#!90+kBq z6>VG9&HLaeH#MulT57HbPn2W|O_`3WyWwMd1ooRlEBy=e^B@|utmY4HiaSZAo1o5y zmc3dE`JSn`&K*S+3k5&Js}flyL76~CQV+asAxBvaQfRvz1a(QLDBLo}aPn`2t1~o8 z{9?K9XPH!v2{9<)h~rMkyc1LmUWdjvsVEMwDn-+b zhLq)7*A}6L1!7hqgAyZrz#5Mo$CVs1xj;meV|jLc3w>Sia@K%i0&=5(vI7UhyyppVPP7>cLm0+*+sXP01%eo_+4ByDz@kP!yF!88Sd0?#NY z=l~ri57>H9)2dP=v=Cb7`n)n4h<|Nm$cIA8vkzEEOcEG{#9ito1e5s90~PkBWfLvK z!ep)G{Tpp1J8_lFUTqA~8)w^OGW{2CIw(~UGA`0mIg3HW-|$e#OU|>YHS(8CrFN2w zm>J2?L#Zla7_R76Sh?`O;j?z?DGMPmGjttxL7(UZCj&~30?11aU?yD>Vnlo@5yu&i zv)B*Ml_D0ei>c!dw&5;fOSf-O>W)hG>Lb)VdDAYFMbHa_`*@%$tV_4Xpe*>O5cjw? zx{0~W!C7e$yf9npS#@Y3Gp3GAK|z;NW~Oir2kRTp2dvUd6KA-pNYR8GNBU|_ShWc1 z8y@A2(<)hdX;ED(5lIr88gGgSM@6SP?-7!49^w#(X-b6^=+G;Lqt4`ziGFRPsI{HV-p$JK;QfQzz(cK%5Cn zPq3#s`#iX2`wcd?Iot)}a#Tv~LEP9UXOz>LRLyt*Ik59ukesy0ZMQEGNf}96c``5X z+1ASaCbyDdX;U`FS>lyHp;6A<+#1vbd^`gxlh7yv^_s>a<{cb_>BFoXu?$A7QPyKqOe@dMCdCfgn#7#`ak2U(Z&JP1AS93Rm?X;6 z5Q;?I0TC@(4S7agL8t*mTYO2f>ALw^!|F;xZ@u70 zUvcSgzguKakRs$kK_Yrwls)NChCAI9-dCt&S@cgpjdF%QB`o&h`aI!gp!S=z*#h3U z4&EXr$4+d+I4?t3`;8ts1irS5dZHse(bj8ru)k z5~D;o?a;>+Y)vT(Nu8B3#oZ7=HVALL3scPvNx`VxoVm=cXQ-xJM59HVuL?S(vMcDn zh#(GA4)vZ@@?%#PTwe$^8;@yzkskx@#Z1pTC*c*X{T-yd1ML!XtB#icfz@>%MC zO-`wjGTT=H8H*-E(i?8rG;!_!5{pB4tjrt&45mhRK1D25M?l3#(4ERHvE--^CRd-6-7q#|r@aT)}t?G}9;Mky`+4eeG$hNz*ZJTf^v(s$4t zv_z1?ztEC+54DRw&({C)^V@zmgn7N94aPHMBS5OYlIMR8bd+Z-md50!(aXzDh{X>H zrV{pAFImmj8IlEyMiI3)F#1!CkXtg~OP%$%^(+DT=Nc*jC-kci=62nH^ZCeQpD6)V z$u5wIz7| zT|pHR?Nic5I<|6?bcB?T{EKWIL&JF^=+|13guY zl{|8rcX8LN$_n1IIttdynjseH1bqFh)R7X;y{2TNwg|Z9D!5tT9cLX9a|zF3F~5mA z%zc^u=oRs)R4X2&&!yI@qN!MqeVJJ zaexkl+`r%F#^w3j{yoeGF~Db^|4@ym^1UYTqkez|JQpU2yh6Uh42eZAlQ3hrR| z0@=C)#P6Wu4m*se=NYT>{%tJ3v|95B#FwgrC#u3Oj(3KPC_!{Be0X|Lp<5mqmO*l{ zA;MiP`Gs&vc3D1t8t8t0{wim3o;!(I?Q8DAB#( z0Xn)F;Mj+GLPd1?X%X>L5re@;1Xc3JZ7K-g+u*WtBR;ORHE)h^W}|>LmTx|` z_Sp7UyOA|ePpeZiV(o`8zPZyht#I!iE%+DxnFGIT%qawNmT>KUG&2h2&pGeW-54HP zto13!dCNz~W|t;)LM+ou)@A%?d~)*S`fQ2Md+&nVr{q{fmLXxazL{ zck1Zc*>513jntMBxhyT4H3nBoLVSI@Rs{Bu${khpc?d@z<%7wmqi)#m?!%nh#m+slOXy1^TSHQq&sI$Fv z|BrnH14)!;jQyq8uRUJnm=uhv*E_Zjo4Wmn4{z%>Iy@9DX$zr(+52wk+jG&H$UM7Qjt z5{mP89=!P_p1$&j+2OXkzt?LyU(S#w!#c|k|G{m$iei42ex?6uehJpP+&#rPhK2{9 zu80U_rhjg`-j6?%IbYws;4hIYaas;o>$bYJ*tx|IGri(K6DPC)5kxAk#}H{h(7A|kWj`LPEgJFc#8)+ChcL+ak0UOp-AKu&zRxb z=$YAmWcgm`cIXlD?}!2$IL#)F4ba*x<_FV?uIfW#WBO&V`@i7L2WPzMnFVtEV%r&0 zEiiCtTsA|5h%>}6L=%eHkRZ}iBBu6hzuzbU4Gc{W*^HwwrZ&+u;U=k{9`%@iY1+O- z3x?3XffJ7RZ7=^vIE4bs;(je8?NEY|Q*B8rbUn&o;SQecUYXc!26WlC%@6gT$&C8< z&Di{g#P$qYbDr&L z5y#g(X1um89kzBOQ?Gl00j<67U>!VYPmm-q0PXzBCyZ=o`=`5y!8ANIKRoiw!VJfT zDJ^duUQWhKwnaI?qpZNuwrEBO;7mHKU5WNGG}qO&=`Gh!ahVMz4Rm#VI9}3{+@@zB zKs{1Y-u__SMMBW;{Ugz?YUH)5pY-Xd4InWa6+E38z4X|3!4H=r9qijyj1HA{8zID~ zTl~)Ok{&-{K&Jdn-#c|+VTLH;Jn3u6Ik!0Pf-$Gu z>v|1DKadV7vv$-kU*;w;o{2yYS6J4Cb#bv_`g%Pp+_Sy5%dVJ{mAh4!-rv*1!~11? z-?@YTrhL5hacUE&uE)3Ve&uuhl8=-l$G1KW+IFwbjuCsw{dKTmwrPP&)USkKvQc?o z{e-3N*Y!*G+uwdngl~U&eV)MLbFWFh=uWw~ls|k`uQWc;^I$AB4TiYH2Rj3>?x1+2 z|AX1k^+ZHIth+#PLqbXV^pE{=i|JTD&d62<wWd-ents&MbPnrp?|%Rk zUl7~dLIVQBSdzuMR-){$YgtHX zF%>^k-l0y-d0xcXZ|E0&kz?tLIYcb}Gb5%N!Q}=V3ph?u85q+o5zdh?>wd2e!bK+j zqWS-7kQ2u!evLIf?8MIf*ClcC&(K^xek(&%)^^X|D0!NB=JJgeu|6!ui($E30H%K2 z=Yq?8LvDk+-9ZWH6|sO?wSEoe<*s0UYR3m?=+l3Rigk7{UC`Eh(ZeD{8Yqh+SWBq_ zhGI{h6V@A*2u5P9LEkUoCS{ySh#HUdqs3w{cIJIXM57NPu%{8Aw?-m6n=EYQ_CjNQ z4wCzwdXI*qSB_(7oLKB5Cy23j@VFDNHM2;um{QOT8Ej&jkGgbkX@=a7u0>{a5b@aB zIcG*;R@lD$3dCdx|8frc&j|L$#eXh`WB(~U{)23c-+=X(3~e}Mq6GU=Y=Uk#@yq_B zKTY5x5mbLr>UKO5D#UDW^x_27I7 zTbQT#AknkvV$Lpz=diWF64mQ2)Nyn<-|cgoEt*&}hF@9Ot@u9}0E| zk3VRCHVvh1H+U%tjho?tpgUpDrGHKO6+3(pgLKN?EwXH^8xC`2@e9822e0@TU-v$- zdPO1KS*iXL^dC&m`_909=g`8Yr{zMPjpGAi>|+yT5)`TSrB-`&`ui9Mxn^%Grif(X z!G%olBE8W6(?syA-QZd;^}h;@AAW${RQax`t3ZF1`FqC%=!AD18Pr_+%502BYyc}f zUW~gh)N8(sT0H5G=jQcRyV;kyh!_~_fgl~1oAV5ilwT2fLX5ZzL-uVl2C6E5LdD3& z_QMn7P@u>JnB3^}vcGY1V+}eFGft1 z%X!xu%4`hpw=&Z1Q0zDP?_;snzbEXSA^mxr;|wlJusd|6;|h`}n2b4R@O>lo6RmlF z3szoVID=gUlGOod&pm>{VRCv(qF3wdV*W; zC)6;`|FMSq&`01izp`$>P#bR_tg8gedh3vOENk}!&r67euvKQz)1~qP_r3Jg{5+H` zM33ZNaiBXRYxtdy+#-x%GDV@JFkYTBr+bKWtJL80i`hEA@c9}J=~^))$2AMOsw?$- z=jngyal5#VVjIGg^}ht`x#*tUl`wb<#eHazUjOQo&ZeUz*az$4gIm}KRs4DZCz!8l z^;xZrsRC?P%j=dz1j~6MR2(0UjeD3`H%5t|$mHhz10yFaW}y>QrWJ1V{2NiFhYDX9 zlU2}6uF_&BjL2LsJ%zLMId4h~O-SKbQo}Tr5!bwvXZ)wchn!b>Z1xTtP~eDnoWU|? z3$8i&qe(t{@pg(q5u>~p^-jx)4I^Izly$X$0_-Mv{KH3oL)-UZ)s}@0QTb>|DGfF= zI5Pdq?1?{3hG3CM5U9VvP%rx?+-M@+>?U%KOw5h#d<;oyI!->XqAC6um3>Td(=`vv zVj_Hf1hKye4BqznlmRFuE{A0IzGu`}wo{UOAm zU{;v$+Tu6l&B?MPj40Pl7beZMU6w*m`7$;(+Y>L(e(AIgTr`mN|B`@vFR5`ps-hzD zYy-J7zq(pa*i|y`Ke{5`T8({lEOD=+F_)~#+bUW;4WW>*4O#eoA#{Gb_*3qB2VKBx zd*{eE{)@H%5(*6eI;V!DGT5u$T_#KZHJMp$?UmXJ1x zrl&r=3Sbe@j8F&5D5Wm3xS|UW;k}&9#Oc11VA_fx?vTKg%SL(ZU~zZjJSbvrT8D{P z2VD-W3pBB`m_NTQ#I`j=DSuuKt>4Uru_KbOL3uktR!Er$Hp_}@uen2TKm_fq4e z>{H)gXih3h|-LJ7PYd7vcE3AYRVdMAJBW3UJ!TeWD=a+x9_-ighpQhxlnFL3U4CLi{ zqtQ{r{&d^K>&7ED*~M>EmL4R*ZiOa6V? z_}HMhnN=j&`i*3qo|r=OSUIs<#d6KTaO=G~Bh2G+{qb!7?N?0?bNxu1*RN-11CtXFum_xsZb~l~ z`|OyL7krej{MD2!$(gHP+zBaxS2*(pOZ)>rM)soh^@qXUK*E7C=NwMZ$FI-Nvy}N!KkE8(O=cZm#C>i5Fq)>rxbHM%t{wuen)s&xoJhQ5yxtP(-%TE; zP7hz3m~M5#AG7FfW!928JG$Gxy*^hx9Bqd2!ZM((j=XW_R0?ETYk%@6hiA38!+|b2 zK!#_W`ZRZXCmZ~gCdX&!pK|{kHuWMjYurEJGCwOHVc#4lKfi4~v>klP#2-$rVe|Qs z4Fj1xE?}0HnwWipaci1^-7m*i2v+OH67L|7DY)F8d$YbJt0RuT9y#XEq<{XdZ@_7|bm#LVk?n?j+qy<}hY3!j2vF-*>lZS2!M>Km3UPoObLPw9HOo z@nv;(2PeSw?P@pq)!a)c!?{mwRt>V8%XBI6C6;P;JNvY+8y{>|eaZ3GlKDUpBAr41 zA%g85i7Rf=3rPzqnp~rs74O>(B-mP3zV&!K;hLJw;{RrNF!i5EC+XgVQBhs1DS_do zXU?of)mB05A0S>jR3Fg%@#T`ni^=$LUVfG?Uit1u>neSvpbqoH`EXGd1Ll3l*;-F- z#@Cde&OJ81m@Tzan#E^#0=MOe?aISoSa_T;PAfIw(*lIr?PxYB7;P3b40jFv*ig&Q z>tb~5cG6c$@n!5cx4Y5Y>wD;DVczy`|G|89G=@wgLSGs4Ji;3~n61Yp?s&zTG|-MZ zC%n($NTg;epTx}*&2v*OCW9d%6O+f=vX%{6r{K>UdHX48V}!^_fv65Y&aEM zNlhxo6I4}jpR~>1p3yfHEn&Xz-Mp>&{S&i!oF@lO8ZMk&ZH&45RW`)_95Jo>yF=3pO~d+XQX47 z{OLly+Ht+l)aCI?2F%X7dbii>d}ki33Tx3cOGeY2eHQ1&#AjmtlDhS&VaOz@Ce(7g z7bwHkwoMl3v8uFAeg-t-vy4gku`f}V+`J1sCme6)48>ay9Jnx> z7v5I+Cg=hyLsA-IoXlpo8x^N&&pgF(_)|2zwAkmVm3tJ!F3}Hr;4&*>odB~2HUA@# zN_^VIpR4B{u=j&9BOHO*+K4=s$vb$V7<>$wmdV!})ZSdVl`LadoceUJPDQ!zcF8+D zS!Kg)97w>*EjdFEwox%%Jhsv2Xt#dy0e36kVynh5K)n~oXisC0f1uQd{o2+i^C*|u zoz?H2ee-W~&oZ;y`rKmn%~BtiJYqF;KVruYqB}TEevfyKL`a2Qb`IK4O$FB>A1)Dp zCJ5OL8(;KTCpbH81Q*!1pgG?Z5>e#0FNM{qoB(yX_W@M-87Wwl>Pk(Vvv#!YnPkB@OLXQbNG`=n&DwhH)%pwbn2$ zg!ZUa{w*AaCb?r%c2Gikg9$x#)Xp&m4SoyAAF{7jmZxhPJAv;7wexS-S7Po1lrGB= zoKV_??F}tTG>n0cbIze+r&ecb7okr)Tv~Xg zpMR^X?;#77&qa`)<#rOZP?-%CR{1m9Nu4u{mkFyMS?cCA)tLp`=qq(5|;?T|Z z{pB-I_cdnXt+KjGE#BW~X^8T`Uw283?o6Q0x&qDhE8hCj0JN?sdik^5w2(soAAcSv z%a?aZW+Sh+{v1*$f3@~|iAGh~>3{c%5kIjmR^nO;q3gfyU@I$1i3+KBte)qX>!2?! z4KInKp~%ov8v*_X5s+yx(?&&MDef2i$JZgOd= zd#JkKzzQNRY&~{bJI*PaepQtJ!2NaNC^@PvHeWy2z;Alop>KllF0x2$jT~~p0!`Av zO*-T~kP`BC?fviZsDlYo=naoI;F`JG=P&oQ$$d&O-Z^1zrmofKc*i~+yq(M5Ucob4 zL67%O)h#{>>jaJwT*$-@W@wUV>V*?SL~kn$6u)Tb*)XD1t@@Ekj+T(^RJ;n+K>@t665El1oBdFpSe zKaSYq47C&NiIzoQPD1?pL|WopPye!+d0@lEd)D0*tqm6qynnm=XewKzTADL2Ig94y zC9bOLewF+3FM$J^^w1bt7*nyl#cqSz_Z_N(NC^g|pk_cH!Qa>EW=na!c>SIp)YW1> zn?YvYoyon^{n&N$s0X_ST5N;%PETTAa*S(}AMajxEj}m0&}QP4qF!Q-GLDUZCS+K9 zwCHhsciUcfQ@cpxVIO2k>dt9(H{#yMv_D_%u->ub(Q)_6KlUshmcM)1505k_6!1g8 ze!3*Y;nVSYF|;Izt3lV{E>{xL@O%4zWPD{nRNogZQqmx)AT8ZeQqtm&knZjnLb^Kz z=@KdFnxVV9JBA#(yNBit`hV}so9}b)x##S&_gZVObB09vM5GK;Q?YG|nrPT-tgL1W zqHH4S-Yu9FzQ0d<=a}D-;=21^qN(b~5*16~C82t6f~{j#G2PSe&P6jccSe+gPAARH zCutou6z4`yOnYUujgiMS*PV+|r1vdNURZd*J2E`YK%gkNzhyP0hu=0A!W!Z?x9(P- zv??}fhdsfornno){7wZO)UepG0BpPEyHI%=8_j71odapXSWRu+Nt1+mx@Qc{ck@L* z$bOK0LT9mnE+ut<@D{7zWE?zG{Sg&GNQ-a@%F%e!RJ>XoV5E)@&|K(r@L{-oeH6r- zqUr$VyVfpg0;`8_IzOy#Zg2J;%l~kDd(JL#Q!-FD*527V(K_MqMZzMPWF(Hn^!KGZ z@45XXPU2wi;djy_pC~VUoprO=oy@7vz7N2Yia1!{nM5w3WLUUBne<9jQ022B^3c&0 z0sWQ7J6^Yi_5Z?!+>9!PlFQJl7quThPno)#usouVR9Ocd1!qL0GS}kWUoJ207-i}{ zuo&;(Ze?6<>>UW+xig>cG6?1$2Q4fD?8-DM&bt^B6F=37nFOP6HEn>x9%Y%JUUL_Q zpPp7T#0#*HP#2+=TaOF(HK*ZfHXsshs4PQFT%3cnCmZ!MvfWoVyDkzs!I~SMET7Jw zQ!R5F+LDbK&LEEoUJfpM?)xX+Cg17O!b5$QKiy;6v?>!x-buM^hgId#4uljpnkTgo zsv4OVJ4okKTBO@&n$my)HQ8q3%AFsurKA9Csz^sOau-eKR*#Jd3(NDMe&pt6^5^^F z4Lx|aB$(W8P9-suF3=9nz!HjlbThKmZ87cKCcZn~XzL&@*i;;-v9%LKEN(kk&A2!y>lVvpuwF<+l!kEa^-#fJZU4w#0Jp8N} zZxo0l|D9M?a;RQ{$KLf>(&TJs_knY?Nm(6XqRN(P2=piI-(18;P;a@s;zpEkn_FQs zB{Z)T(3Gq1^T^X5?;7Z_WP9(={5Qc= zxP7Fgj}}g(5FOp5m>JAjVp>(1;{1EZHJK-D|5BxN9*iF$!V?X=R76t8Hr9)m-c@?{ zf!8hbP3kiY_kIhE1WDLz0YLV>GD>UWLE6>~CQ4~9Sw+PB6eKB8+JzeI>w z=O$q-C#U8$$BN}Vgk&s(#i4-``!VCgIziQ`@?A7T1Rv-ZmACH#2a;pkZfl?byTfZr zSn+k>1dz#cx7zd$x)N&Msr9ZKi+5gyElFdh$>{80@@63qPbi>dI%&X8To z^T#M1PotwVXSwpo4A=8xLC}!t9)ObS^*TYy>-WjLUMu5?-Z>w+?*F0%>~VycpL2n~ zzU4-D|D?IzHKX~Qk-AIZaAN_+5<}wvMJw*|7Ugwi^ZPqDN3RI{b~eF`#*_Pu%xLl| zAWU<9)2oZ2f3I#7kDq&g^GCP5TokY7^kpl%b8_uyaBf#az2vxzl`HM}~CMHK(3FeVmH zS1VQI8KtX!W27T^&m|4r7u)ao6~V!%pBPx&;w?5)qFPq0WjTpOs0GUn3(u)02t8lPTHO~m)49tlLHm*%S#QQ@58=St;=km zGrT_B#-|a^;p?tu1{wV*HaDWlOvrE1zNb7oK8%`oKZHIVK3<+1OM!;4w2Hc1L% z=Bvcyw2xOFyn&&ypvKjAp_0R$s^mr#8%>km;$;Z{VeO~`g1kNy71ru&lO`=hm56m@IKhKzG>q3+!Q7d$3%nt z43=}re-8uGRO(a4c71g{yek%j`UMjN*wNs<9s;i_Oc z!b3W&ZY@v46JXsFlJ1J?x%vZ0RWp>$TD}s6mdvz+-0yg;kD*&L;ZkJ z(@1FTGz7Q5*8?_~E&YD`(fAROrt%rCli(G7U-{}y0V55?VM5Os%s*>VJGG-WXGatV zluV&O>y$OEHs=qXwmG(=5aKeUMVssl^ur~Bg;%dn=zeMbdwIlRH~q?~?}E(Ii#ZQ#EJ z>up zRH@_1uES(AWQmC{5Mc6oo{m}QySVC?Jkfkt=e~1!oyz+wlI1mT!pr$QCuGeRei-%) zhnJqmO1SJfSEZ)igxe zaK%Qi4Y+*ku`}hmQuCkIhEtzYUlHpXY6l_Me55pYxrBSJw0j=yx$?_$z2uHmdT~nM zgPM`U_GBqRraF6xLHLhy~Vr`|+elHFJ{L`7EhbcR$lRvAeq?YWaU z{o?}0qY8JW$@s?V&fRehPe(dCv{$S|GTBvUiO;xhHNN)L0L+1I@{eK(%fIvv4g=m5G?MRKAFkA*3+a>$d7DrOI3)S-s)kf4&PBTb5chdi6SB|D)#FmsBh^kw{Wh zJVcLUtk)&tFIvVKdADc!u%I)_V$a8j5&GKL$J*Wa?40n?INJ_4Nfrq~rd8=kCcc(U zH4Z?1rmaBNNYTW4$OdtS0aScY1jEouZ-x5aZhUN+4%znYB{ZGI^&SGE6a2D3eAwcz zuV0l+1qr;zgE9tvigrt^?XA68k$>`0TA`-qdd<*S?#+6B80Xkij>$1QQFc)@T5yMS z>Al4XwA!;WKGNP;ENJ^&)++jp3b zB)zVB^+AlQ3i+D(SQrXPOSvOs%0lhOH8{|(4LwSSu(^V79_dq0SE>}C%Nddt?C9{- z4eA%=d&m#CbhD9A?qdOE7;+)<=G z6Zn8ul_^MHKd6u8D}1}CK1^(V({T1Lf77zFXYP zcqb)}%>x)`nWTKD-(w@~@$tLlV-#`HiRt3WkzZ-pqbUt=UZhH+icRpq0Vw(kxp(wv zB|iFtuh>Ko%T#T~5;2|^cfHSePCUs5FWJe+Hevha-C9j30#?J=f4>ht^}X+V8*n(} zZcG^sE2P;XR!$AO&N;K@hp5j?)?yZQ9-=jw(>{|%;)`2H>*c=npt45;W4?LS#~u2ObJx3C7moAQt=bAiE!@ zXj?&O1aA;FSUQknhSwXuPYRDPXjyYJj`s?+a*fH5FBvK zC`GW((A5@IoOfKKfEMw_m=&G~uii@EGo11JtK&bPz~>{`n$R#G4-g*!+|=Y7+EztZ zUPr~04{D0<=_ie|9d+SUvB z(Jomge6@6~8$YTz>`O>Wz0lwI*=1I{Rqp=qJwIBUvU7bltd3}tQM}e_d{cuHHB<)D zYT<91%%(G9>{m|rS4zD#CR2Sy3cZ$f&l_hu_3yqN9E5Fp$M$^N z;gPJMEXWydLcRbgf!>+Fi<04E=IQ z=Az5<0i8*MlA1!30v_HxQvOO9vyI46*5&-6!ZJtaiy0$}6(s|Fshk$BSNfIoXya#Y zW}ts9@YQObvFZEDi)CF$ww0DtSSD%_r24cal@M-9*=`ra@Fb1Sq*M4^)DjQ|A;C#M z_my+|9lTCpEQ->%%8L+niN34`kjh-y*_|a0S_`x;)4&Q|FS$E@6B$eyL4hI`#1AFb zb?^ur8=(8~$VJLn|EF!7)cj#+iRGX8Ve64otNyk^YMpd*!q1H$V*7&*pXJ z=&Tmd!hMZ*W-yvb_*nRRJ!H*A&-A`^ZzWg=ITHlG=zBblAyq8!WzE^1WefPq;8m@h z#*aSY{V0xl6OD+XZ9&l9GVvr78N2*#ZvX#|Pcc}RQhMS|ll&d7u7q3EoZt*AluDiw-#nEZh z3&0GZxcQs0^eJgB*D6Bi`QVvU3Jug+TjydQ4l3O23nMIVYqt5T%Cr-Ex?hfbY}8?G2LQJDx?Rg_Xnd+VkJdZ>gDS z)8jj0Kro{Cmmk5_fC^=@??OIx-F`GbC+=HZbaSyx)Nm%}33?G7GrgaszNKOl*xh?N z1X7TLpRs7YIExk{L4^YnYhET!J!kDJDntM*R#VCH8*+`Q7(iHQ0S;XVxrOnkv z%?S9)w1>$@^I;IB%g;@f_fO~FS7ZMLQt)42@qo*>2Z*y~pwgLn$BovxIRWMEWIAt* z#BMD5&K7ZbB!G6E9yFA;G4JrnJ{=N;YaN7uS%58bJc=o(fXN=JkIu%Ja2-JeXFzpa z+|0zP@-J)!WTcJ0jNj`-Ec{Un0W*iiJqxoqOb8JVaEDfpKG!WREhRF^k*W*CZj>oP zlvbk;PPuf(ZIYl*Ce*!bLtlXBnaDkb1aeQk5orDB1(myO-n#+xfcpfOtc-i8lp@f{OqnQ2Z^Y|m1>PXP zo^jv{^MC0xt)u5EH_7D8R~MQ6cTPzhv_&u%0Phb8m<`1s$gdQ>G#J}}EV>z4ZFEKp zyq5>(ar>3~!GHLzthp>C9m&1auhZpUbAq{Zo9rWEC~)$q7`j@4@E9m!oPR+j(qs`NFP77$SToM+rmlOC z=`b=3^iBW|7YA?3n$a`kmsah9@^a=mZ@+*ltWoDVm+f+){#=FHNmNcMq>|iMsrL_FZ&&oQf5E4=O8PF-s}jHjzXr2=uQm zM8c_gaJSmhMu^{jJx0!YDQuTFr&=2#l%`pkl*eOtKL|@pU^$l!Q zsC^eyzQsK>@DggVG^P%^0hoKv z&@0G6Mxi>Qr<)-79P|n{CRix=MFWJ9ecsfPlwjHoY)&M2&h+RwwD-4%g@`@)`hjy4HaP@n!S$itx09#PYKo1c8HKUg;3=(z`v)}VM|2*4C%oP)MngXys~|@Qa6lP{F`dI^JefCVgtg`ynDx-JrrprQT_6QV?`zA`FLPgY#Xm=swS8 za@+r4NlkVB(Tse=XyhA84*M-gv08URQPTVe>42Fh0eC(R*_fHTbn`M!TN7|f^bw~j z(Ba(Og% z!_@edhdAEv&OZWaykvsVqy7CBaq1aPk_ZJ4j)U9r8p>o30lBFD zefBHsb7S`inI}GEw_ZQzVr^f6v|j-op}S<|YbL&{ed z1zT*oe*{~%{*$7?^fJEH+C0jbCGm^_qAhCth@MkGfnn-7XQ;fEpyp+_{^3_=We-$P zN3Wkp5A^h48LQ8rrEMjB!ru5OcTrTqzkF1!)$lfdJGDp`P_BuhVM{|hD-VVpWrBed z_N15K7NVy&X%U8!P4D!_{OagK2CNDFGT7Uly;a#!9%=a<`dF^K;!WuQ@1O)r!w^el zL&p1Zy7+rU0V_2T|32cbvYTZR!PFyRDt*fS?#=wmM$g(ByFPVyeHs-=f}Rmr#ORqL z!zc6+tLh!mJwae}NTRn3#ITr4cj|@Ev{t@YernVsa91v{t^k?!Ueluk&@C&HsPV4i1))mJF zwnAAKE>y<|4)@(^|G+&rdV-04<=bo>gxCjJJ59GqpAz&0e>}c9Vob0X8uo&}Ux-L# zG&r#4E3+}Ve#CC94zY)jgxtO?T@-UqA)%IIU0{u-?}g)b(Sntu{qF96H=N; z*>nx^9dmQ|sq!!}vsfz1{SW(h`P@IZmSdvd18^jcC9ugfZ-_Ocjk}^pZPFKiW0!xe zc2j#2&C4B3H0O$>Cy1pNWUALXO^H#O3q1N~1$g)Tah&5c!d0z%JA11W8m`0gyKGY} zHGl^+fhAuyj#tfQjR(pyJ&e7^sq#^){&|+|{WID3=cl+M5_Ev-C0^#_S zf89*$0Qtj-N1-pLf7ix>$EMg8met_V9Dp5!_=R)%VzR-9B_`hSroR(gGFp1?h%;uC zyOpHQG{k>B{t4I4TTfi{?vWmeYoeRQxg(laLNekg3d;{3Q~Gr@^Bri&3=Q$3Z}+2^ zmc$mj);a_9l;*t$cnw1>=3AxiBM<5j@V1VCOKh$d;l{My0`@1}4u@hD(jAF8j0C_+ zptM;c<6NN4Kjq~bRkcj5CH(OddPReU`?pD*KECbGEca?BskOc;)-7`%veG2WMt|~6 ztqwQ+a|C2PUUxwP7tQtUXP0%+T^F-P_JZ?wQpkiS4wA`fGu$6;6uURt>d&f1}r1!pv4`Eoy84OCS15fTev zW?BmeJE&EyAXN{42d(g*8;wMa5`FHXyRBV+LcqsG_?t}^hb<>L^o|H7C7T@EUQ|b zdYH_IBu6A8!b-L3YjS3ZY0lq1>0L6?tQ+=s;XHx6d+E)zb&xVUtG+vXMG#hBH5vHs zC_>cS)i&x#vh>**fb(1>W~cJ0M#NfAhzFT~QOr-gFoqUC)QM_K2Cyb;cs#xC`bXD9 zttZpc{mg_b^qasU*m+&ZOO39>`=Ncf!%S1RZ+t%A9UL8uqr2%cyp6^f1fljAo`hOcb(l_=KttEI^&06!4Zv=bzP{90 zM2nY}d)u|o$Z?TC*7fzmB4|O2xAr(iiGqe^3*Q%T`muE>-bc_|H@F{-XwWtEB2?Feb^DL6HN6W^$y2p`+@{shWWjjarpmDX2t(B1Ruq{N5Q|HGZIVG#tV;-o3r-eyRCZ2IjUMIWEYG=$>^(%q%MH)@nP5 zC|G~mOYz&`O~e%|X}uHN;n^@Pm0S$zNbnOn}*KjcSRyxv*LJdPJG@96!&>#)Q9}9az=#%Bmd0v}X`{CDuFR@FkzUB* z??cz7BuZ`ZHfTs1)44co{{9(;=+hYHs2?AK@r|Q}C$735@SbB`Fj*<%DU_9~a3@(o zOK^G^myn3(UXfs+7oi8MZRH%R@%$pcUBnOXZ*^!wq=PicYYjmXjylVy-s0-bbw}6> z>YgWh6}*B;0L=Je8r7*0*!QbukMgdq1YNvmzC&MWuL!)QS*@BQ@?s2d)rDo7xVT^8KO^BxBMTHq^=k((zBJ_~e zvG`W12P+(H|Ek`CyG*m>ca*5p{$D~_HbTB?BbcI)9m=L;tapX0^DF`m6VlyXx($C$ znmTAIHr20Inhgb}tMfH*-J!1-HHYex7I~S}8fv}FkI%DGp#(g*(cs>x4KY4QV{d~4 z;m48N^NroNp9!CR+~>l@CN9nrW}zCWdJYLOHr@@1z9rUW9|Rz568WJmA@Ybnb9XN=n}70?11?CjgXm;yO%~^|lclyJqVAn(8t^47a&hNe7dnPDg z_G(Y@HeZGYzIM3w#~?b$eT?hL$IP#U!nwLG|IzOG@%tgIl)lzotVWx9`PiSb z{gq_hHFRJy{aM7=vM_B)40^Ts6Q72v-V>=eU4zT@zdk9r4r;sUTIXziTTU-Y?3;|7Jo1nv1WFE`57RVbIc~;@wHbo9 z&=~=Kr8GiLL`nvdaBgj}$y#sX6!G2g6#Vnsygt?fB${5yT_P3X^)mE4T?(}z%hWqh z?0KPKHC}HV8FZSYDCs7Vm^A{iMk}IJpfsN$iw)*=)j*A!Z+$)kU2gX=CQR(e(;tq3 zh==Z;FXO=Cv?>HOl~HDQxHF#k8zGPPhtI)9>@ax$UI@;7?dwy*?}Y$Tq+S{q=l28D zF2PCo#YmO~BuvLWY0=X0ub+2EC}nYiMlI8(KQ_)tN(dQAyC(FUe(7?;fAUW}2@{Z5 zcz|A?oV3Vv*1c{z*{fTWtm>DsEI%guaXDWog->+D^*l9}+520EGsWs7wB_+WbLBb6 z@>unoOcUF!nvVo+CttR=nvIUJ&y~p}2*4EFXP`(#@@6QY8S!bPuRgjmLAYhqII_1a zJ(+)2m~$TpnQyS3W?sxZ&e9cj{agq#K-lvX=L3XwJmt~5+MW7M8*5~T3JpHyy+!{e zBK4M1No1MTv1&i^a4Rqn7z%t}`O#g)0klX`&P;N}K>k{2`0IZ+3`Wzdind|(k_zw7 z-?Ws=m0Ul>H~TQ4iy{8c+a!RNWvXEG!(35-Ij}su3xNIly@UA$0figfq6R&i^{YM# z7fmP+SLIun*0u?E8}OR*dRkDEy82EJqC*O3hZ&+;#!Dok`wlszfkR&^)YcmBPavf6 z#6GVlLWCae?mt~~^zi@@iwZPVsJa3F)PSpJRs*foFKV4yyw$dbM?GXKo@Ql}LYY;b z3QGiw8mB=nF$1cW04KNcW3$C@N6(Q)9?o9tKbDeJTGj%c33{C$8y-%VNTfByG1m0M zVGP?j3jKTUg{`( z!IU`#ZEkCO;FD3bQ83ZK}R!+qIxg#?sriGdDXF1wMwl@`rCwo)@2a0}7Aj&gYb-~tS{3fCeXls=YiBV_Jc;tGw7Ej! z@CWuW6b-8pJ!BuoIx9gxgMND0F;%3vd(mMm^SF`Jvfa^!(3Xa<5oW7FWRV(m4toMH zB;Rr;*;b)9l*f?v9NnU`GpHB;#|zVg!gPX;m76YdD`#Qq?Gp@Ifmi>FV{4}BE^#Uf zV80&-3@G{F!Y_Ow_Spw8jb8kidySV7057e9ATnBsgpUKSxNd;Dg1e^L$pMG|vcj`bD$F`+ZOOedvtiyD0}od>m(OlR1P-!A(#Y^ z!9^qk8H4rfKHS=L3SB?CMPJ!VCo_(ha(APlQT)v6W6i94WTZF5gnJhnTR97<`*eNr zB$5$sWYe}+-!?@J(LhkrgWTxy!5F4*`CEJg3dE^(tcZO?@L4a}jZV+tV-O+|1a*)# z7H(|_PX*4GMV>%jH(>Z|?%1<>rQSzi0oxVZR$uatIn02k^H)vATbz+Qva{L00S>QA zDC!67o~FA#&8<{-T}beq)M@1&lvsQ)a>Qb~s|X!oq0lWBfpry-koZd-*iL-j3*oE5 zr%rPA`zVFLdSvWMSwFkn@*>!pBkrDc5DV#D>b4-HuC5kC#4tSi)g5KGqwUdz-w1f= zoma{ogoM9}om6!xM~udfGr1Kyy2Br8lzP>GW*E%Yo#Le6PuUedF^v4q*Nkj+KoRbr z`30vIK;U}nbR+!>Q2=CjUEq#TmDFh*l=s^$oY#`*C0v-op!_NB`#804YKkNlD8k^> z)H*#s=zbx9k+Wg}%hD8X3JG>fEX&T)U2}WA#s0&4@}SS(bhW5!r#ePn%bd$?E5twgg7qUs@d}RHw+hV;iY%S(o&S3#PRtC9VBgbUr%#*RW{y-|x{?8q;W$ACepi+ec?*kB768RJe8K*XSS3 zWT<{C*o1n!0qs5!J=X9t!fw@l;gX%7{}HravOw$B{?b1rfWJwqgJ7p5qU^fYWrl2} z*IQe(>4>_3mZ_jzOdk?-)M+_&A;-4WmY@&&C^7*4T~}rBmLM}mK)qoPlL;YhABTVj3&gw73Ubuo)X+xDp%I)KfjWfqN=X3iSd-9@#@ekmUix+sW>z&6c#H(SE8xb;9%R=)1pj})YZDcl$qaEPMx{*#h8K}Co6^U&+-QdXe`wvg`&(@>%| zb82fYbjsC=-oG}69av98ty#jn@gDJzbmen3AisF7He8yjoktA0Fdp zM2IrG74 zb6~pu^R>u~5|KHkTsRZM1^r2Sfrw|Z31CBv$=z z%!-KqbcrpEFqQcK%GuhmcFm+sWgc27Q8Sq{b<;S#m9nT6OffLao-&k^Tlw8eJ1c7~ z*A`=o6h%<4w;HtuD{r*fe?&A7DoahJDpv;C=>0s6^;70Dujwl8IWNo`&G?I$HiEKi zFM&#_GP<30{XP^0vZb)LN24D+(%W6!EGY$#V4wM!+BzH1|Fu)iRq@|Bf1bI;KZmHX z=-L3i0bci)vjs`=(Q8`QMb~48Hq|q}UsV$>B;zV2PA+=Jh(e>)_C(PCS1vv+umrSc z4rY+UFgJ4@Zu|kABTG{(t7?1xxYc5jTpgHg(@jonHD`emzQ zL$g$N(B}DD>iM{;-vs!uC}{p`mW{R_ z>sfK-eQa;u!_c)iMy8~j2z(@bzO_?n>5T)?C4aapQ3CoWrWJxuo~F{lLciF zUr&Y)@N@63O1mA^#kV`P5bt7pub23td{(Q*1af;t5=3ouPqDL%6NRM~XXiea(K2?8 zddj#6m3wBjzQ3_1n0s;zH4}WCh1J>d!{ybw5ZI0q`}|J_3_tpJ;K%o$WS%b}h~fU0 zO7Q8QaJXl@8JMxkX|53nvk7l%P*BfFZI4jZeY7>T0RL79Gl*_H`h!hMDkSABJXjkd zLL+ZC5H0pYGwX8;WrF>0 zh@=s?y?&aYW}yZdTM$;*0Ty85ESBNB)J#r!OuLUIn?T3AL1l%sE>NYzEVL$Zc2g42X(iiO7orjOAae0XmC}4mTYQw!8TM>^d=xHh zM*N4CM(CFW-JqdiflhcdqnX-nYbG1aj0C;<})<0NSSBN`dt;EM!wO2@k1_u2U zWGc{LZA?6xfAER<91k+|-{Dwk(=0IN74Gzk!1oi#Bb!>0v9?CBNMk}_E70GL&gq@< z_e*^ZtLv<#&;%&&8EC{K^^9lY!IZSRMp{=F5JrH1qi`&~8amgf9r zyR&Ry`aboJ(c$~_k-Oxr>H9YBTuByBeO>3q+Uteyd%;EzTLf;0e?ahP2_5a7UNce?+JPnOj)Is~_yhpy@R z9$Y5~XDdE^sKf=w;9Z1&G5~*`GusGEtm<8eAc?T?qU9lc3O4GbH*^O1pP_B!I=kE9 z1aXML%-;R}{R{yV1;hTZYz!+DD(XfD)*Z%ucP$wSp8SGWY#TKp7mRetU*;d%c%ND~ zywRTyi_hs|YXFg2AX5HmS+kbnQ|V5;u?=n01A_{^f$@eZ^k8UG?GG=23BQTpDE;tp zENeW|b`b$nS-TzPEfPMgXVIVjQjOn-nMvYsF**LxMDd5gYo_Qqer*H?JHqOUmK@&1F>&X$<485C1wS|Pq`b~6C-KF6gN%_4=okCZ5h_%5pM)4=a&XDS+5G~;-EeQ=<15tp1=zw`?-^Qr z;PI^%QJh|sm&&MEn0lQloG9eG@6!(ZudmZStq<>e!08{`IIpgX5H1Hav{NDlg#}_G zmO>^g?y3-2n^N>G+sg)(>d5@}S3>v^X@b)n80~u060e##qZ{Bw1e{VA{DG5yg|r3i z2yDz&(*@9b!xsEZi~|N;b5{Rz8q1u^q(kq>LYC@`4x(04s8T|DC~RWYhJsfIvKfW= zpFZ5))r2(mGtjpyi-@8EPIGK!SsbKJwZ->511Wy~g}^V*VN0LOG>3RMAZ!T^x*K$$ zku?I{RO?jy8&AuIz>(Z;O7>?tuy4`T64y3Xsrv)x`wHnuOPmvyhH9gS9te|65E;5_qEs$u_{rLA3uzR#<-0rAIVr;;cv)cA zb9bheK9=BAK8)9EQzAZ`ei%?+AeN4Pb4r9C!}`asYEwg;??$NR4B^|M@ZOwl(IV`g z>K>rK(D?~w0J{s!0V9DW$~@7c!FZqb8rhaqLF$vGUN6L~S*{v(z(`3oxzt&YyvDH3 zsR&+okd$!GQW=&58(=WmEp8J4W8aeG34v{Y{G47Pg+jPhxwEHhLXM?ihCo;BrA3QO z^`z)Qx}k!>R7Rgf&LFVc+4k0+dIZcbgJ^j8chdy0*EGRi&Eu{Mt+o0f3(*HR0s-@} z*e?09@@IUD>pJs}^Vq^LyW5=@|5=k4EHix^Cz^znaMQnDBY)VdWnyW`e|zB!)=La^ z+Ym{g!*uzLM8cY};fzR6@R3Kyfgyw~)$>b-vw|BmsI{RmU+*f_;JQWOz>)nCJM@$+U|L2nD_#&HS`o(E=7~#O2iT8y=TYRpvM%4C)`PXH8;PM zN4;D9u4lqCGJ5kriEK3F5>8q8#k+Qr{q)GIO)H<0$YV<3RiS^kJPxHt6=U6ur9sIF z)m_2-){y3Ny+e2%i>;5vv~Z3&#LK-hjJ@?ax4CUVTEk7x6>F~9UF(`xl$laLc364*PQ~r8aWMGVqh*uUB{rhV%H_@buiHa~LNq!LRVp=kP z!3Hx=e8?~EeW26L@|vf2^vS_jPb*E-RA_QD=lPd*;+S$V^4inomuzM(0!!HJ9_61< zTMIiQ(92l7c6(Nd^w@mb>uy`lL$@<;?T`A}qm6sC+Jm^)C2C-rrR6ERZAI46xr_ar zvavVj*giwDr%qIZQI|31Pa*amVdo_j@uWi|V<{DS68sJ-8_bQXVal25+<-P8tShsY zM4a!E9vZIIpPvlO2OsgtIRNeY5eo0TqrV8{3e_;XX+HiHCtJ_!qGK?+d~K*vEML6J zU$9db9!Dm4>sHc?+X&*z7g$vPu)zvDIdiM@Ei zCm46vxPi=9#Vyi>E{#Up+)CnH23Hp(Nzxn6dj@`di^GQiv(p4kOnoXcbt;K)3z8xK z@j64^>BP3>BZfPprf!S~n)2|Zlo9{u(}J_I2laQIxU4q*4+vj>vU$^qoKuBS?P$XN zSFiEx)N1Iizccc{^<1^J!=^Whlg&P85ILDgkn?Xu_N*7!S>!T;71YkKXz zO_-~PSi|m7+C3@jj^bK|?f+|afnVS0&AblTkbLcl1oB6+)UdUYLa(qa3Er2UJTW4X zSW~COkzqc+rW(GC9`wL z;0{Z}3?{@0dQ<-IPJ-f4zQZEb-1WrpV}ek*JD2t4A6ws`pS>y8T261|g@pBNp7vA|r)C z>Un1Ggh%(n2{bpuvZIOMdm@C_BqM?_^q7 z?zQKeBs(MSeJ&tV7uZaKONot-XDiWK<$WqZja$RS!IFzot;$MG2x?8;>~iTztmO(Z zAur30)-nB|_J?c>_mXZZ8X((JsdLNZRk!PFPO!PeHV+`d&17n7JI9ivN^$!_xWGs6 zeqc>>Z?VdbmxwH-3KZ+=(b{*bx9}lh(VBPsQl3-0DvwVIPknAqCkYH ztu1mD>0%vCtR{@$3*y?F;f{kcT(a`CV2J(!a|hzBra`8www$O>kwaNMyH4 z6?p^jwUjNIRe#;eO~vP@3v0AMVmUPF4IQJo&zVOT@*P-Db(bsMb7a#mqS@FL4*Sz= zF#D&lYd-dwEsrvuW_7$i7<>I0==n^%@>^TGUh5~TYPmJ=Xne^aa<)gG(EW@H^c+th z&+7FyO_4E&Ew@-B73Ovpr}lWZ8k$M^cgsl5aaq*CI@UxTO*J+Q?y!G)IF2l2|F~^G z-t{v9ajov3J;kHs4wPQEz4EumK=Mzv2^sroYIo4g!gG7VV5@wrRa0YY4T%?m{yWN@ z4}82DPq^zn&m@Lj6OpA0NQ2&T9**H!mmVNk|u=oNP-e8N9g;ZnMrqAU^mdo{Ff3y8& zdq>9c@*a7>L&NrxpCdv)F>|jiX4fm8OW!x+6o33%@6S*4)?mJ{2`m^i>!fJ!Lck`w zt}t|-y_1e@HZP#{_~6lR*`{SN0>J`z_9?l=YtopfPeFJ89xp}Kq}~~KSE*CYn8DDz z+{hSs*On8Kg%oz(ey5PJ@g~#1ik*W}SV;m%T7Om%q7LENN*lrkU~ZL)y(RiuaOEI> zcXsJ~=ri^k@_hM!x;f9NrnYU5M^Ji~rc^;e2!w=?009&T9Ew7sA|OafLV!SEV+uw& zNEHD^MHCCgW96vyDgr@KnhjAY(gZ;f>4G3F+)aY_oO|DR=fnH>Mlwc5#{SQ__L_T* zz1CiP&fj5Jl(onU)ZW;=POlsbmyFfA_8;Bz2ip$E1&lOpHQn%4?C%? zEVJRYs%PL>sj>|0=2cF8gh1Z{>ug1p&$P!KM;1SR{xEatG27~X_uWjo__%KTkr-pv zKsm6Df#N?$B0X1&EvS<&yv++{{sK_iCSj$>*SJ+}#=;}LH?CG;+^%1w|IMEg-BtNDL-E?fN6(9OuO7UgsFDOF zPF!BGiHdJ5q1XR>pZ-}zzD+oHFmk!CHCHou_R>D9U0X#y@aO^O#(4!Qi7J(WU7yqx zRujjgWp)Es@npKqnfU}LZ(lL>9W$g1E0qu(Mk^e+nlPx`5jd~gu)pLKw;cIA>Y^_f ziPG2=&dljZ0Vdl|*jJh1rGzG!J^$fu*%@pGo$FaI2?4cqpx6LNlQREwvMBu z0CwT+Ek`F^l&}YfmnTr9Nh)}T?pcKvdue#vVI<^sTRu3C9`5ZC`3R+z1?~d3MkBW4 z%%Oeo&USo>%~*}y)@uF%FUJ$XwEVN9Hzhlt_7$6_KdI6>E8REo?X2w5bW;2I7x*-CcqAlx1hprNHqFpJpqNXiFlz*ltgle?Hu)iy-jIdOKJ@fA}Cy2|jCrP49a8la1_ujmIc?P7XK z)!c-~!)^(;>)wIKvqz>CTL&A@X{fIDYTi1v2e*59i;GCU^1*xRdU3nG2*E;ajY_IZ z<#`|Mr#*l1ZG^R4d2cyR1~B5`M>WkicGZZiMg{qaKaJszi>Ardp4QaUn)8eDs^5^e zv)?ox4q%SI@4DCN?*$M;tX+v)4&Ni+A?~T~<5(=%f7j-C-&koVcbYL8-K2uzf>sJ- zWh5gE%VtVPvrj&npVNE(uDB2&+SO^NgfN!?%O>GM+fR}kMKy;qSJK5L1Ja`H4VW^) zPKoYjV<(E<%TS}2CvqZ&C$kKCwd#iJ`EI7}a5mY`jL6?Y6|f4g7HUwqyLz6}W~8KJ zrm;Nnz421F6s!wRtc)(w>rxL^j75pjt#sFJe6TSZy@)h5JLIO$x=(5In_3LgGi4s_O-N2^b|j5yE@eeNSbQa@tjMJyNk7hb@ItXC0Fo1<()e!OyjW1ZwN z95$TeTr$mXDOO~)NoaRJ5c2l~Dv`|_5?{X$y%cbQu(Eyn-s*uUswPDF-Ncyo+@Rvf zRG6t{yW2SN;~kHtm#IjfSD8~cr%qopc%G!vlI(&`G<&U)&pLdh`b1$kJTvu+^YN8V zdSFXd-R~)^_$D7*7q4Cx9#cOt`5ik<&%*w&IKF-#e6uUpJS$Z0IA*qVO3_`owW#p6 zyRxpV#Hjt-cE>0*)ptCptg!?21RV>>$#53U4$|$7r9~q7hRm3V`j^hol6O+p3dXC? zFj^}EF2j=h`YArt<6bP-D9f$l&2`0Z3G#*V@)@%V8B$Nm;Dje6 za&ue_{mhQ}JSlya-#}X&$(#qBm`;-NH-&}9lK>v=r!($+07w2=pMB>DGxwlu@Qojz zT{(HhD;LMMvtgC)TvABkI4{R38^Nbn5nJ2CD-ND&Xx73bGgD=zHkOvqYI|2a#p+Cr#qE+lA-Z~pOBB%?d!k$n zkCdyf95c^>KjfRq>}0pY?|JM;F(Xb#TTo z^W|P_%o(-nKV3ynTa+_|Id(L={j;IckvgsZ$06ET+a#@(SFQRb zJ-?O)crBVoV&_b#qTo|DuJeXX7Eo6!+&DdUn}0DHk?bW}VUiouL~8sj{?V8r$M)Tc z*l&%z$*s(nU20LNb*_iX^+F{`?*dCk-b#sRyhJ3KZCHSv*>N^`eS#W-Se?^W$PQFA z;~Vl8V2iM>3fiF~K0eDyk}Pg6J|+gKGQ912L=Y#4oAY#K^og+tj?TP_njhQaj=1(g zeX!*vKXE|H`ZhTEbmLwbWo$byc=phO_`sJi_JHirtWxNrLOG=^J8H)zeonl4I5NE^ zZ>C&v4!eEZu{5}bAm3>x1-*8OIjr8NVLwf?4SjA?{>_|2$z{S}qgV-%2qDQ?9@OYk zxm<+Yh^7KJ%})a8eu!C07chc@3N6R8(e&X0@ZinspJLKBCeI;emx1HRPoi{ckI+%snxjZ+EzmX=YVrpJ7)iEfk#J~T?nR}VUv{D5;w8n>(1mvwbF(()zgBd zck@)g$8jD+cs&!Iw&_-UZ7(cOs5WWQ*2t%MTO0jas&q&YaR^S1uAXLAgJ{uxlMB1W z(+exU1<7TziwcAN&#Y9mU$aYx@9nH8yE2qyZXWxI!h;O_YG_nk=jU%yY9{vy8 z0hJ{77E;fQ+nb6DTHL1KHl5}ZRB!KW!*WQYreJlr#UIO&Pg6S!4h`jR!-D&fTr!|f z%&Vq&3lLXRSqVJ)g;tE;NjnRq{6!}gZTv-1iFUm)(k#NB8|4&FGxXZ_locjL} zC%GyZDOe`H0zG8|fk1XJY~2}t1O^#HqLINLh#?$dq7OIHM;N;xkQk&X28jUwV&HJU z05jB&5*lC) zupmZgI16C{hk~CEm#x!5t>&8FH7)4Kk4YlaAXa_=k*tol`AwZlA<_u6W9lX_INW1C zIs%l-{*t?rX>{-mA)rnl0Yiyf#<}eUkMCSN&LV_iu-stT{1q@meFWMC0mq2cVog zfY;?t1QwnUKt4b@4vt9xbWDdS@IHhU0bc5X8aO86+n7#{LU$ C%>~T> literal 0 HcmV?d00001 diff --git a/doc/pub/BM/html/figures/RMBenergy.dat b/doc/pub/BM/html/figures/RMBenergy.dat new file mode 100644 index 000000000..56aa19939 --- /dev/null +++ b/doc/pub/BM/html/figures/RMBenergy.dat @@ -0,0 +1,100 @@ +1 5.59097 +2 5.38551 +3 5.14253 +4 4.92049 +5 4.80505 +6 4.62185 +7 4.48617 +8 4.40662 +9 4.2445 +10 4.17527 +11 4.09137 +12 4.0624 +13 3.96858 +14 3.92182 +15 3.85881 +16 3.82252 +17 3.76908 +18 3.73743 +19 3.69007 +20 3.62217 +21 3.63773 +22 3.58399 +23 3.57401 +24 3.53176 +25 3.51544 +26 3.48965 +27 3.4685 +28 3.45745 +29 3.44185 +30 3.43143 +31 3.40337 +32 3.38956 +34 3.37394 +35 3.35488 +36 3.29924 +37 3.2945 +38 3.2913 +39 3.29587 +40 3.28508 +41 3.27126 +42 3.25836 +43 3.24323 +44 3.24687 +45 3.21526 +46 3.2367 +47 3.23167 +48 3.21595 +49 3.21003 +50 3.18076 +51 3.18035 +52 3.18534 +53 3.15874 +54 3.17502 +55 3.14463 +56 3.17313 +57 3.16107 +58 3.13238 +59 3.1367 +60 3.15066 +61 3.11037 +62 3.11367 +63 3.10958 +64 3.10312 +65 3.09725 +66 3.09705 +67 3.09602 +68 3.08657 +69 3.09206 +70 3.0703 +71 3.07878 +72 3.10305 +73 3.06263 +74 3.06851 +75 3.09004 +76 3.0673 +77 3.05662 +78 3.047 +79 3.04692 +80 3.06387 +81 3.06418 +82 3.06383 +83 3.04044 +84 3.06274 +85 3.04205 +86 3.04167 +87 3.04238 +88 3.04487 +89 3.02483 +90 3.05472 +91 3.02654 +92 3.03249 +93 .04819 +94 3.03152 +95 3.03445 +96 3.03407 +97 3.02488 +98 3.00511 +99 3.01687 +100 3.03151 +1001 3.03374 \ No newline at end of file diff --git a/doc/pub/BM/html/figures/RMBenergy.dat~ b/doc/pub/BM/html/figures/RMBenergy.dat~ new file mode 100644 index 000000000..f85552e62 --- /dev/null +++ b/doc/pub/BM/html/figures/RMBenergy.dat~ @@ -0,0 +1,100 @@ +5.59097 +5.38551 +5.14253 +4.92049 +4.80505 +4.62185 +4.48617 +4.40662 +4.2445 +4.17527 +4.09137 +4.0624 +3.96858 +3.92182 +3.85881 +3.82252 +3.76908 +3.73743 +3.69007 +3.62217 +3.63773 +3.58399 +3.57401 +3.53176 +3.51544 +3.48965 +3.4685 +3.45745 +3.44185 +3.43143 +3.40337 +3.38956 +3.37394 +3.35488 +3.29924 +3.2945 +3.2913 +3.29587 +3.28508 +3.27126 +3.25836 +3.24323 +3.24687 +3.21526 +3.2367 +3.23167 +3.21595 +3.21003 +3.18076 +3.18035 +3.18534 +3.15874 +3.17502 +3.14463 +3.17313 +3.16107 +3.13238 +3.1367 +3.15066 +3.11037 +3.11367 +3.10958 +3.10312 +3.09725 +3.09705 +3.09602 +3.08657 +3.09206 +3.0703 +3.07878 +3.10305 +3.06263 +3.06851 +3.09004 +3.0673 +3.05662 +3.047 +3.04692 +3.06387 +3.06418 +3.06383 +3.04044 +3.06274 +3.04205 +3.04167 +3.04238 +3.04487 +3.02483 +3.05472 +3.02654 +3.03249 +3.04819 +3.03152 +3.03445 +3.03407 +3.02488 +3.00511 +3.01687 +3.03151 +3.03374 \ No newline at end of file diff --git a/doc/pub/BM/html/figures/dnn.png b/doc/pub/BM/html/figures/dnn.png new file mode 100644 index 0000000000000000000000000000000000000000..bf388bdf4b4f36a1f8c4d61f4b52a4efba715ec2 GIT binary patch literal 274267 zcmeFZby$?!`UX58Dj*0*DP4m z-6c81w`QN;-sdoW=iB?bzJHI`B_cELyz5=-S@(0__wy`5)Kuh&2`>{uAQ0la3NjiH z$oV}88JKKa8zdDzGnmz)BQrS7gEv$SFZ;5cgUXY^AA0Hy@Ja`?*da;s7_%a{?2isZ(45H z*()6OFCZ13qh50(@Y$2<;d1!gXu|$n3nb-oD^Y3$N4}K)IKZC;42Ol6FdLm9p&Tq-D`($Prh0#2CE8-Ap$AOn`Ai^ zn8?hEERIn#dt)z>Sm~3$QV4Sy(NVE9KYUQylITFIdDVzCQ%>%7udla3&Fi7&vM^ra zFvbrjP0@ID@i^xo9Iu-%5(edx49LV$((=FEy-OZg$nl8S1)}aoJPNB>0iH^K>hTv zrs>g@qMLG5f`{$=&)F`9sP`M254@Q8%%@uAODS4U=5biY*Pvr_Ify$-`MlFPRRK;D z5fvU@4xMP~2leFNNE)JEhiu}SJX>}P{g4*4Ty)OxLAH`TZu$KejAG&?w@7$|WvfI; zXJi{WO}+{V_D`i#@{bEM;o|u-)!=FLYuvur1o7`p9%R^4>>$VOmm=!KA?YW!dM1DE zhgaY_?FGVGqv!5Y0s=fzqs?db)lST0-n$R_-Sn`p@|Izn_fI6QyG?b-{W|JWWJ?Sr zSvq1oYe1B*_eTF?1Z*Mh-B+jyAAc~79MM(XDbF&J)u-|`S&TN3Wn!!RbbYtHq^VPr z*5e!ke%>wWPM4JN}a+tbp z1^3ffBNzSmJw5Iq@935991jzg9pR$|Dz@3$D`NSgfaVylsivsaT@e$q&d5Z!&sWsyytu zw$rmSvUBDRf+NZ92nY+f9a|IdmHTmOk%mdD*(7f`0ZOq+|U*b=1OwvzsF+lex zL?vKrT)^_ots0542hdu@pBF|0eDH27;1j2^R0NZmMD{aB28&Gw;IWV?5WNU|%d`+o z8qNGM`l$k^#xon45|;Z-YfX;RH=W3=o3RxInSMmGknK`rU`$6~3!;%5Xlso2~U$E5BHK z;ce&r42ye5kA*&n4!oq3oqb{PLi@Sfi>2?{O-a)}q0ojGlgc6KkJYHudNnVqwW+Ua zj&fbUp1|S7p{hm2(XMS+;H|LlzsWMAszv;cSm-(QDOUzh)@r7(#+uq#r{&EH@r>MW zV&wHy-lrv{O1wpArM?eWf0W~Je>O)eyE|L|zDLHr2SVE0+DVzZ*}D0jGTWg~qAovd zmF8B9$P?DilU^ws=Wcg#_hTz_%a|vteKtMIJ)J+Ddhh+@`uXlBG_|L;hK7oWiY2Vb z-!abd<;KGysUf3;tS?$<-<|s_B+zu{LuYp=EA+gJ-Mn|N+2Xk+^TqtmydFeK(9*!X za(6;%sq9*!VLL~wx?#FV`aL;%1{ZC{p3;t}4pc_C{TxnJ~ey@??*g0qb0QI`bPsqjtkjhV4zPF4=25 zcKSK($Jx+5a+~B^%O{LeZD_+`N z^!Oo_?gPEi(zJ*Bx-+`>bqz|{bSw2NbnQ#*KHj#~x7ib>6xOy;*VESXTQMKn=?Ls# zNPn7sU4HCC`r!W1%+UQIiOXMU$oVrZl~&@kOP{cb_+PPbV%3{D4yoD|4#|IwNgqj@3IFfsEMxAkDi!n>}4uBf-98kaP8 zGKmV)`(H0tX+2oX>^4+kvY?-1?Te~>Hq^Q``+6fe!z1f;hJ@;_ic!*N>+VeKmKUC% ze}sP@?IT(}LCME*W~*&*goet}mnWB7!@U|N zCJiph2;A3*%_1Nf;MTalz}qRmFS$>-&uDS2MMNXTkm-iy8!JKhLW{-It52(2>nt42 zg<6G^zZmbNiyp&E+GS@m+ic@4p!OY~+xAnpc5kHIXwqM|h(@SC?5KPia$$*@_C|d5 z{CD2eYAx#b6k%cZ+&j9TT?ndV3g5cqbafIdR>wFf^1L35jfuT@W9mlbulHXImXP%N z$t9m%o-T!QUk|zd^tyfiropEyr7yXK%ik9X-dp$b^}WjVD0Mv;44Y}2(d$$1z1!!K z8>M4ZT5I$52l1IJ`&aC>@U-ZL-&R{KWRfJ5CluIi3^^IUukG#3LzNPgjEpi=e97!g z``~@vQT{kGKZ!sr0SQ0JdlEaE-)Ugvdcn${x#4#T8y-uMoEA zfSo@2Lb1XlrYYlvs}2gD^9^g9)nnpxK4lBR=;$-@tx_9I53nzaL*6!FXb-9+RKbuR?{N4cJeq&p>}Y4SoKSnx$Tj|%A?)cG4^3z zkCdI1!_?!iKadYkmW+d}8$;x8N9i&&N>=R#ZcWc_vby_Z@Yj#Uo{s9HW4nc zqQ6CuFofeJrJGKbcnOV2@`FXHSmGte!MiHq!+u@miFghZ5wilr(&cKc}?FHoWX z{MuPqQ#H19iY6tZfB8yy)dTQfU2xQCo`#Q1izqt|?v5VZ8Woaf_^75;)}~updFg(Z&s;aB z_R;RB#g7DJ%ygPBE+I$C-+iIv=4%rmg<>vn7Rd#%K4RyzT5esT$vELg5$hv;V9#top_~NW4m11;)4*aHDV>1UT}NCf_rd01|1E)~EP^ZE`;Z@;!{&l$Nt{SND%UeKhd^$c7rZs^!yB_OTGAIo& z;z$-}QVVY+h zDj=)8Rp*o0^ScOAQd6%A2@B80(4J45PZ`J7k5U-~fOrp5IU=x!$dL|V8GUnriJv8$7gyKJ$jHd<;bDFK zv%J?&M-+z|8D@hO(drZYwx|!7{2d2r_S+QPGE3RiWfxiKkg^;jZRm}QHl-r}(jMXK9|lTw3fv)18^zB@A4jUtK4O|D3g^`tXH5w7U2T6RHi0mjGo`~Z!HH?xI2#CIQiuU(*GU|>)_KTn!OW8Rkd@LS%T zA{H8OtRul3omfn;VlUkjIVNd)-MW>ys^pGkrCQj{o3Fy2qZanl@po_8^gr({FhT{` z)*<9Rb@kRYxs1ZD!Wx#(e);lcp+1CsVfy<^k84P!ukFD{tE;QUb|V$TCQ^It97uEC zamS7~kAA%36#6)}Cx;c!TzWuae<+Qz#&$$%s0sL-Br-EC2oFTqCAjBM7V(~;9m~#; z;Ep3D`D&+`AKjy)$?_Y5X|Og;)f9u33p^mRv^U&54hRVF_$A5Ba$M>t| zf=u%Ttg%nV^7E@tZEQ7OFcsS$G7u=NrE+1d!QU3?JDi`lw5pe6*(E5ajZigicivKh zG2WKMaSb0d`t0slg#;-nglo4OO^!ZG#ybk!Pqa^7M=i|d=;H61zIPh*&jwGdov(kFn%7W?nnz#Z=;#PT1_my| zpc2ktOW%M0V1|m9r^{sKH^>nMCQ@_ivh{C!3I~@=qw8C+vs?zqyEmg07-?zmmfDT{ zVCf>rVJV!|-&9ssE{{)0(6qL;&X0+SsV|?S9bjZRWHwE-PRDS%@pgx7WNBj=6-zzM(m*usI4E3 zT3f||XeCiqM>~%g`~-p=(_fOdczyh=FdQUax0#RP%LAttPpP2bp;G%;sN8ta%v594 zRt~Yq7o5nz!xn$3D=|s~?FvY9g&veU~T)e0zjUrXk%2&_?bCrByrwu?@p z#64c*8-CH{l3NyJPV(Gae`Mk;ITw}1`aBaqn_!nViJ@zge3|^7e9ueHq-$|U5Jg;s zioKKtfv!2MCf4e(JmMi~8cpk1PtWTI1%^rT+k$4uj*rPihytuYA-`^111qYMtUNtH zU)(94npcN6_~~JOzXHyaY<6#g_%An`7ESNqluNSjFHF@^YHnA6*C8@v-S*vZpsIXn zjM{^Wu{;?cE!t;I2$4`J8H?bZot^HHk)-z@KMHc|6sBk;E;2zCf5+E#A+RV$>2qVh z;j6h{Tns8EdYYdY*3~pt>BdX#cXI-ERj9#v=(=^ zjHi79cZvgB=h{)KAoQEM8XNH)2SI<%Y{z~$n^Cc)I;S_AFI#t%)jk0VGX!gMN#j8u zT@|>)?&`>}pFby>V1Mc+zB9zw%i$fKK5c6?2E}*`KZ|&MlqhSL#qtdO3ds1zPcBQG zJivZIfKX;$>ABMUa9>|+B!fhJ80R#>%XD2mJw=wCH!qOO9uuaX`|}iP-%8bmc%Qma z;7Fz)SkZn_%?dS{BdqfYy@(^`x~x#?y7I$TQ&ST-)|yvIH!HCskmJ9vg&ayv(_h(ukykaZ`34fq7(P$1`_0TyUe>|F*tsi0%a9@`cK>?D21e)RL3te`B`Zjj^J zy0*v_mCpD=SC*&@n^L9- zyEx1K8-e7E_wV0p1vV#cCB=)v%}uzDvw011PWUeH-G>eNdu~qOU47g!HipUuIW%T< z%BSQzU|4YecVqv2*$Jq|>DKXn6yxn95Qv54fsf@g-fK$bFd(~bB99L{yVt|1P=Xe1 zcm;1%8F2xIwbS#O4Ejj>gbq;R@9i!RyxdtGNf5Rhj*Dk1+Tjt43_w3ADaq>l3)T#i$JL4+c>QvTTz)23ws&@BZR=f4%%6gI$OND> zoDa09TTMzvz_Z`>ItsJbMRg^A-(MGotC6sgayUD+W-a+;awbiBdU}row@hD)K#}yF z^h}i`-C3c$vDhgJLUpjW73Cx8?jn!2we8eXr_Ltv10Az!MX~61Ya{72v7KTDAj=aj z8G?;2Q|#+Nic+{teT#6O`SIF$e{*KoDJcbI(tR9BFV+b{rK`eaY4g=u_;8kZHG0oB zs@)?~K8oQr)c7YsfcjAp_gu>x&G>xe*Js35K+9&>ng@lqtM3$@no+4EP@L&oE7`>w z7NJ%OdARmZh8!6YwY#VG-&BAl`4z)!*cYFVabP z6>NO+p%9ay4zd6hS|GiZfLK(}E3vt&S8CVF&4M=lb~bAXmAEqV{xp{NumMNHRSFw#l* zy{91Q&OwpK-jaE^Sr5v95=eG@G}o3?ZkEuBg27r7Zt?P0{Jy`Ufb>BsHKG6Cuw|F_ z>@i@P!^z?v0!JH-j0?MC$oBL)+N0K2heIW{iq{|HHqE)t9j+?vt_tkq+Q3Q-D2CR? zYXiz@dfInPhdN4%i>Fr_6^@@?pjJ8lQF-pv{QKfk@`8uoNNA$Os?B7x(NE3=Z@By8G9WE{`eDA94625tJnru7}9SJE2{}SkX0_1td;?k00V^DvNcx+H2R2 zCv$DA%+VD16`FQn+iF3B(BZU#ixKab$9RY4d){mBQ(P#v`HI1Qs{`6vnC)P(MxO2` zE_=li3-Cv_ic$L>)~!DP_IKUdt4tW{)-F7l-0VDAAEH6F2bicJ*|?wv^5|f@yf;rT zwtYkr7)uoQ=bS0Gb_n(#S{4|7i3b@o&cc{!JnQSXm7!8N$YX=W_G4AWz)%=|lx{NW zdmS5KiU7b+R$A%vpN2I*PcrxYO*4}xKAEo7Zfk~6pF?*Z|6_pcS>L{W6Q_rDl?i5u zrGK%qXiqlamAQ|WB={8+6a@dN+K&V~^d=B6?*Z4e9V+3r9jOpZW{UdkAkGLn&zw2K8QL+Dm@d)Q)s~I$e>BLcnc4m2kq{0L z?HuhbH96FAsS{wCJoG+DP~Xt`Lpllr2|45e8kkB7hRWkhJ}kXERV`J@1(~h>;9wbR z9tb~cjsq)>Gj!|2?)G*De6B5{wzgJ$5Dm6i`O5sfLc%TkHral_CO87d+&UjDN8lYe z56`cywaYeM%v{ZyUoJ9hj?v)OFXIb3DayOPqg-x+bqg~}zH#2`rM#J^1_TQ5KOU_D zHlP8RYe%|#SCMI>(|V=zLMH`fR|!0S2Wa)TAP{2iH7{N=t~ds;`g#$__65Ll%T3RB z?XC@ey50wzv8^2xARK9Nd9~P70tf855fW^2#qkl6{)xcLsLK1z5Kt>wM#v+fqs?aC zg~^66K--i(>LqP~($cN+_IP4rqof#f9oB<5htTX2?PRFcBShOYTe>88soF%kwg?0CL20)JwmZmHfoSqCd7hJqA zz@NMu#RzL#!Nfn<30gB$YNbUwI@F>^YmPoYP<+_QEM4> z*qr|UtC0~o{T!8p_VV=Hq&ONqNc(iMq0t6W=s5H1JJ#?h_R1hQ zYv573fM@DtQ08!zY!k#bTU#>&4s7LvS@R~xDq#F<@=nO%k{67(YgFehhWb_)6!4SNJ%N$wGqr*o506rNd z>8H*HsZHj5RC01MfW!LPPEIeinv`DwIa7KXF<`#JIpH!HfWsvmN zx%{bpffgA4FZ&rSwexge@2riXticAPL^J%BsWt&~i|rfI*2H=^CbH$@qusHLcz#p4 zCw6u!7^~z#u1<6saOZM#A_)25U(@BIoFc-*-kS=M%6>3ZuzH!h3_sdVB-owB#bARxs2a7%Lx3Z4_{$+5~A!7$A+=1ok z3Gl%aEO7+_?Fw~YH$Iq(f;lPHoQ$}Q%?y4pQ@pqbDCGmf18dECw~W)Dzs?VWCW^r0 zw2211s;L_eYolL&fk9_r(5e*RO$cdeRu(%Z z!ZH|Opr5Tf2bdRDu?ampV}dQsu;0N!OEC^dPpfX+Jv7yw%A~2 zRz@n_y4qnUi2?f3Bfy0!2Bbhg(H;R7Fe4u7vZ!740cl8#yG(v^Gb;t2sCq20@~yAh|2S=0@n zcD`##s+~&5V0Z9Rk9#lu;cf2siuQSG$;-=I6$@)_3v$CWKXgIq(Xk2IT}Goo5z#}P zIq1dR`AWCu+Ep<=a6b(H@XD1d@YXo~^!U+utdz^a7YTURRPci4>5&j}tV{ZV*~?&8 zO`Zm_<^j}u&*7qz6|k$5eD)@SI`?~7x}$aw?F%cPSk{3SaQ-#lp+PpuYg8qE{-CwB zHRsbOp*;O^fl{TeHEgPf0=L{QA3^kCCt)ggOf5`&NvXjHRCS!eKJ6YFN(l2ksqs4# zvXv(&TdjCRP>3vqB?9hWirK#5lUY=}hKUcr*2MT$t;rIV$G+Z;SReclv{2FC#-3CG z{BW>l#wdpqP|?9;1!5wi>6mp_0C?rV=r(%xX@iwZc^p$u1xLJcKmzH*R4Kc2G~pQC zlK1cTVH=JfFq$Ru?`~LW;yEA^ z1gAC(cwmkINa`^t3&Rb{omhiP$;L~74+F{#!1QyCJs>wWDg(%EGL|wAk)9(l_W>LR zV7CT;8nz@$kpzgk1g}KrV{L#RH>50^DV9MFZ^v1E#i^;G1vZow;SMMx2^75H)h56N z*vgSq=QqGGPbxjGD&ett%@&Y_mHr|YOpF!;-gl+P8a3N(K*BHmMBK%i(F1at%a`u~ z=5iNAW-AVCPTk_2%OKe2;>s}1ZBQ5LVT@c8`2sfM$QQ3$QTOQ9mcOs~pS?hVh0xiWU~@x9)~jr>Zn#NE+7AqhOwBBn8DHc=X9j1e8tOT+A(YZ; z;DgzW0l!fmGoB$(QQgvo;2YV4&_(VkFvgR$~ecIlu zlriEUO-)Tf#synTKbdm#KG(f*5b>Dyg7Ub0lrK2^$5M{T#=!H4{dxpw7@PdMc6TP+ z?9@W~eZVBd*F)6U)2;27nwm;CxC6NS#KeR~8RKmj_wAE3*|5de*{fr>()ZHw_f8JC z+JE&KA!m-v+LI+h98kK2#FvyWa!z?~9$sSm~(;O_6$dmdB_n5q~#lsjpxz zXWtYLS*$cFNe=5OtM}bK`vaue^IX%fW*TDL!GW$?g6@$iGyPMh% zPfZm^0Jx0yIpVkSFeqYa7#SJmTleK>fbUD6fV~s~1Vglr3m3*>>lA|lO^Zo52V3p1 zF5nV&j(+j#N|x|)sx+?k@0|L4ItVz_j4%o1(-%Z37Vf{ihKXyM*UAoxHDjP|_DAzV zozI!$yMug`zdm3Rh&qM{23@58AzM?8JUMnAukjn^(Jw3X?;X64h1?uj;DJOxA-J(n zqJol!hDIT_$N}w_rYpA{cO|&TG0V}q%RY5cy46jyMV+A)a1%{I`~I~dI^Zx_EYOw)cub&z=gW>QCy}Y zIuH@CaDWkc!nT82Tq>0{hEm(rRy%n#!TD0yX?BF2~d|OJaxyI z#)-_uERjY}yaS~L)uwh1+d+)g3dO9WRcpMEQJMMiL!k?Wss9paz*x$gEnTcQT@WDjehRQ#dwf3&3thJVCE)C& zGXi|ipvqgcG_Vvry`tCds`8avb}dnl*dPo`8jQHhyqK;HpzF)~cfcrAb_+o}g*VrE zmARTK?4vCIbwQ?>YA6bqXk8s~>{Gc&ls>88y~cH~^;`3ffmzD^KGY1r3W7 zgs~am9eIN-;}e2C+6Gnr70L8{j1QhQl4T}?q-bwwvIebeiSQ$!CdygIPj-gvyKaqn zzr^43AnGb?G&TKVW3A|mVBV<&k&=n>IpFNW=va}wub8?LroQxy_(~_>u1**iv;Ca9 z-leK5SmRR#9_EP_CnHuHjanBj*<#nFUYOTTu9kijDXJON>9|Lzqo!QtHZ{ZhtAVSI zQFuI=OBPR#fpeA#_U;TQgzu))q$%xfRGpyRr*q|wpA&GN?{w)4JXQt6y;$j#T_m2j%ai>s?DfJ7(* zM=c|#|ILdes{?~4noQ7$pLeH$>ADaFJLb6W<>du+7?;Wix@MR&FlV1rbx17%9e`yS z?NI>yQ&r#O$J*mU+)_2F{XZ0*=t_IwYRioZeu(EoJx`v5mcn6O@5p~PI-|qe=7twA z4s3c}*e*X8YAgE|s02mtcX;^5G?tc>%&HFs9s%C&5$c+01sFE4K$!Y1#!}-}CV*~U z_?ddA-0VRR$k&q8S+-2wQSX$A$6jNFc^2S@$j}#2*r<9CLP^HZe4+7~nua9b$I=;iG|3E$Jps|&CRc;%CxcjBHW3HDWYfg_c#4Eb$@ z7MKJk{050l&2n(bR>{Z7H|3dBpWmY~B3djrBwC_Xrr+HX=O^}b?Eck1R3?e3O-1QjS}!?B1qbiw0yU3P(uqb)@X&Wa zDJ};Jg<^sao@Z2DC5+>(jkSmi#~Lel(3OHbX*YKzN=(CE$vtBvxu z75=*98d#|M77(ZH4aI1@HHt@KrEtCB_Laabid3J^=h_jVu%yAyi~(965@O=lq19n+ zXT;kY4k+#kiLRN(;(g$6DyTOMmD809ZYIx#lcDiG0S`wR^oM=t_FMTs zaM{#40`(k^Vqxwjs-z2XWX7OGoslMp6V}++qNaJm_XG*t5$dZjLFGdGPA6y(u(&fY zCc*&Slf?+MITnMg(y8%130pY4^Cvz5^8mkna=0fvlG~8ynIkG)5MYFNq?O#awmAMN z_;ELoHd4eFWU^Hib=oNo)c*h}vbM&f;$aoX+|ZxZ?k!xSZDuj^h4tAU{%D%K_U&xY zH}aV`vk&y4aYp>Gcc-WHUPtj* z*kv!|uu8M#QB_qHUB3Wxg8ET6(7u;ak3%W1J2|$b3qsP5tVDMY011clU#J1vNAVMI zZR^p-UkCue$&X_yOqk3~^S#}MSUvIEqwU3>@_D*REN+_u0kUZI{J%E?id8C(u}eX| zjH-`dqU!P0L){XahslHq%@}hTf&-%uHF+hW**^xFDjeH3J32bVDJ5@KPL>qlWYpNvU(y{|RMOAz=LqWXzv0JQhSk#EmT^3=_7}z#39V+gVSk<#Q6n+9 z+i_`~@{B0-Ea|AGIEkNrHxNMNvP@KS^^psQ<=k?gHcP=ut9bxO-YTDqBl8n)yb{T4 z8g;9|^iI(ovi7e<{;beT4KtQgu93IZ>0lzRGO{%cjjzg*O5Dd{T`y01FNW7=;qQgr zmcT(c8&B?gAk{GqrtB(-O$|3s0xOl8RfYx!<7rxzn{S?g+Fy;^9Hybmiq;ZC%ob=x zWX}jfbM;Dj08mSo+E`burJ3H_uCngaKg|e?#ev({sM$65SK<4ef^Q+WT-hZ)%B;{x zxW0U$D*kPLAz>V(^F^io|0nwuY-mu;=b~zsWY()5Yh$A_EHUR(NRBiyxKi?vZ@R(AasOg~r1X2}vbz7uu!Fp)Q5a(DpGn{2Ykhitt|&RCmhkG4mlrX7Lj z@tKFRme!X4Gx}#m_xkK^9w*-_E}gtdNTg(gDN$*}mo|&|DdL8WXH$xJhj{yIgnQ*E zQ&mT27|Jb^Gir)B37HmJ3Kz9wSL5BCLvbFQ9HZCjwa|g3pfRV}k2{rnZ31y235TYk z{nZLywa6nWaKvDC?&_fdC`IcvR`9L_gHs~wI~&yFf@G)*=Go)UL>Q~Lp{mVCg>3pG z>L+L5!OvrXW9tRM({31{Rv1tSpKLh$H2(~xJzShRcid0-TLj^ul(m=r?gXAR2+NJg z?MW%>$oZ<0s?(_}l9J{0lPUiBy8vn>qxRLZzge#j<|+oForTBuA8IGFLkYu)S6Eu+ zKIPDiG|kX-C|(h~;NG^#+wXWtes+|FG*yzjF=lVLQ_h#MF<}y+AK3^;YSEdBu+^Kb zf|QYVqu4uNv`vN*w${*r9yaxcx#U7#u}WTh32F6k=H`>)%;UNHW zOzz9+ufi@A5{3g=z<7Pq<*#h~3=h9&(z_RL;g;QSQ-gDjQYmBlJZ%nR*F$m+(s=R@ z)B!tXDpHiSm!cw`C&j3tI^N=`FkZfYFGZc`ka^EbTW?vvW{mWej7XDpeeqzfAj>iB z7sVlPHmKdGRWz4C89*7Rv+fO3i9sBPgD>9cP|tGz9D;VI6LhY(crzvAOd4vc>^wi= zxltuj^?9j8gsmg;sB#j)2&F5M$a_?rp1*R~KBo`uiO+KbA8vn)LP zBd~^W){&8s{m;wJR2*$io2hf&W7ZyC9%nevwnlH zLY24+|3Z3bs)V0TpsDFlTqkOX^O(o*B~zdpER-}nCzqCZ034)YoP&X4+6=C@p{cV3 zH%s)s(j}cuk8o3E{_$B!X=}+ox+u#9CN1$y!0Teh4-9r4Tdp=mnkv&-g!YBSdc-;& zHIHq?9CO+ZRW~FYgnDkAw{??+>A zDCCg@ie6XJ%{wM7U3p`Rrc}21#uasbU6`)i{MLR(#E_*Z+s3ZCH`SDTUD4C_$66aY zf;&~ZC%IH3qB{)~tHn0#;vx+2@JP-RE74h5i*vfMKgd(kn?d)(7AH-|TkN#HvvRv@^v>mE4#Rhsg1yu|c&O1zDQ zg(V{qmfKX3jDo5nK254|FQf@A5Cb5$_h@5A?OAkAS3^yS!&{E=dG&))v%|F}42D z$E8gp&=YjN-X;iv-gzAwsU+L( zIT-tX=1Vfb4|*}z8}t&MH=~BTg0l3uswz(2pA{)a<_2cg^I_LZ5i`AB{nh4oh^z9$v(w#TqjD*bX zOgg5yGscE}G$N&;KI0R~LU9`?hZN939z=&%8&gfiON+Po2)eWOBHb*XG;fD`w|r1MdKMhl;u?>JMP zsBWET-7Q?ZcOnw7U2=+zTjBAnYQfADp^!+pnD$g12V#^(mc|3f6%EHWQ+M3hAm@7TI=Pkb>8mz)wys;mwsY9_gK9ycZ(r5=QUuHFDyM*y3bem=IA8BOBr$2 z5KgE~$HNe^vEsmbNAYyir_IwB);&TZ1)X0!+Ifnp@TN_TfMcComI&R@i*CiZ(&A$?1lMixW% zRt@jEk~IP}&gx^%PF;oFhRHTQi>vj0NQB=QA0O|eST>s`C4^7<;p3V-88fv!0`47b zdN#dgN|K=wvEh-Tc>yia=!gI2&89@%>?C3%&TZt|RC-zUUL2t4$>!fv;CM+9BIi1& zct6sonUdKi*7;t%Bm(y~z0&0rPbuwVZk|d`-B=hW;b%31>EsA295dTIyC)giHTCl% z=nwQeK`zokllVTK^{19N3k);4~a`D&97^d z287LnG>1t6ZdV>ctpEtxCpgaJlbHzKdtMl94D&B1bhjoiAq;;NLH3F) zTej&Nn!W6r(p8B@Hm>(^4D?ajU|`v|@}Q%CG`Wwz*Z8bi>3#d0?-apq!_hERgT>dGuedhpB9e`{ z%z?k~s_+gc!a3;+;wG4+kKE0niWJpnOs;%P;)L37|N1q}JRUVbZn>`)yNO?~blE#U zjys@>laCw5-8bb^rMNd5VxoI2(tw*JF{wrE&Et{#TbHCMMyiFn8~jPTv!!t4X1~qD zbp6vf>H_go2%Jsca!=e2C%h)b^fN!JFg}tL&oW#;vo0dWH@qgCS*?MHk??xvQWjtM z&%Du0s)CxR+pYN;foOiT4Bj@`w??gl<$~b{x&2WdMD3b%q9py5`S8!UF-JtgfAD!|LkBZp&y#Tco-^)BGE?7&>!m$LsdkJy-ysT zoXuRw+8dqq3XWM>{cS_&06Z*WHvFr>l1t6EBN=lDFU~xfY$04F%dB2>E;fiBj|yANkh;) zvz{*WldiONikuQJG=0bP7cZKVUdxg;Lpg{tbjgtsEtrIp)*rOr(wOZ~=UFGCD;G?? zHno#InZ1F36lUu<8&#Inj@w8cOLC#mmZfDTu3q@@O*bv)(G*q+w!OJH8DCZTizO1a zfi-xwbX%qsK2=vlW*c9vC7mRiUxT{xUmo_|`#NMHzBU7I+A&M$z*)m|nqb;M<{ijB zF8p*1UqgwQ5$$dqweAu(gcI4w>sGn_W@*r1&M!WwU-}a^YOWEG;!%e21s6ueFky~6 zmy4jpCvGdDmBv4U?iT&PPNnVUdARP^q?3`j#P?8XHC-a~!9geJb%!_Aap>h*5$H1C zu(r47jd;e&)bNqUL8UW)viU(F1T~>ENw-iu%wVg}GpdT;Vv!UY$w9U*U#99nGarl3VtEMI;_8Wq&b1^EHDd_oBlEvT9d{Ny% zfVRY0eQ%hcEs}!1xeR4;WmAG7sbx>8tCkHqQD0dIS61jhEo8SbJ^!eok-xgTC^@9c z=xz&MC1nvcg<8lt%aP&BW=ko8@?SOPJe>o3CQc?Rjl*40{f3)zot?K$Y@SoGJfU5o z>)Ixd5g-%}W#Vpwtv21;&2=LHLe}($oo*c(f3w{1b5!#%byQd(Dw4fp{~8bc|6%OC z!`bfp_Hi^SS|ve^3Nfp#O-sz!qZHk0RIQ?@5t~Gf)=rGpsM%Vbc7xiAP$Mm^T~t)4 zp!Nvz`?#+AdA`qm-S;2AKbsysj(*GgoUO3stw4^WfT|D3v>0FqG zmfRFH^yfv=YoJzxihpg$fT+EtpA-{K1KmZ6Crd=``hK~$p;5K=Wi1m&lp8-+{2RBh zxA62+0AxIc_~6@4BS)Um|AdpjE06p!AmhCCh_7q<*V!?zVt^j5Iw0*73VHz!HmYY6 zOryJZ;y&h2N#fPIsIQ-PK(~}juTvbnRi(5%<$9fT(b@75Ex51s04hz@80vgw zJ2o?wi0%hR$FFqsaEO?z^;em-zrO9F%nGGkhvfkZ3#o#SN0avWg>i>jq*RzezDaE_ zMmMR!?$Z-%SlN7ruJ`5R>8l+13TC0ju9;Vb?vL47=|YI)cvD{Ak1==rp2eG{Pag0g?RIfq&)0Biik8xT+_K?M{V$KL!exair)pZ zM&h|cga^a;qZQWscKJKY(^QSPW(Sq9I8$^7mfqxhGeCB>wwCrEJHHGkZI_0g{YORr z;K(NL$@^KA|5-2_Foe>6aX+DRHu{j~R2L2quti~_jz#s!F)G;I^S)tgb1#vURk#ka zGvTD??NGdz#aW^zUT6n8tM`R??sK`4J1-&e8oimGg_!|ljK_%or}{9-_Y8DIzvqzi z^b2(%$#hF|1 zo9|u>iM52qZYrkVfa2D1N8n?9{U9(1KtZ<%7~PPcZI7>MnlOoBugI-~tFV7^pjD0S z6zOWY6BAg0ye8swANiTRmA(Q0ScYEs3Y*4y)GnZotaJK^#xr&V1g-M;028buv4Az; zib5o=hI?@H9~&tSUEU3`RM}q*JNB7)SmLPzfBo+~um}KRlXBOa`X8TyZ&7qFG!sO{ z3uBYB6PAAwLiOpdff)Da;T^brxf$HhuXPaEGH%d@@ai{D5WgL4JTjSjL_*LLj|cbI z^My`IQ0Fh7j`7OVOlx`8@}8%YyvD$oj_0QYk>jVUFI~{EpGFPH_bTOfwy}EL zfy9O70uH)#lK*z==ZUP5F#E83d9mx&s0L?C|E(;ZSfXO#F{T`q5$-s$>k$cKtT`;~ zX!?7kl5^#v!`g=0Le>=6v#|`470uA&2Iaj0`q=}I(=u%Xb9N42nY4bklxSiadE(ny zkhrkvI@$zNIHw&@{2dp^%Q=D%Ed_+gxL>N`rqEC8?Wn;@eiEv`sWreHnD2WFVi})_ z{Tlm6&tOsra7T21?@jz~&T|IKT82?BHbR?ZeQ=tYGXyl|4=x|)b>H1(hgZgGn@!Bc zuKkh_OsYVhSJ}ASti=aFZjRKR)(HttzZZ{C;z7~&t_S+NRqw4I=;R|FoK&CsJ+As0)PxK~_SGwo0{ulO;qRTFi&{x1s0 zs2Bp{ogJ4+*NLZCsmbUuRUkURQe~}+1k4S-84jv6PZ6*e^N+NmPxQL(n$G!xIVV?h zt{5YT{>9fBCt(gRca~oiFcDa=VrtthEt(IRX?2MXCvoHrwe087iuaG@9}^_-{sRUN z-vjuGWqBS;s{hiZhM(Y=ChYMn8T14_#T|+4%aQ$AySfR}c*HYmW<1~NdT9g+-0p%ZS71!M@{!mtT+#8S$_W|gph~^!{RGlxRqP-> zEiQQj5a7emLtYoDMI=FO$5Mb&b0}jj*=DA zchOJMp}wZ?Z+~A$+?Nioc!37}@twVA1@si924vQ;8IKAx-7f|`^)}=mb+iP$i7S{73|RmQwNwTvLVfR(k<{z`NNemz(C$ zC)2#2f;rqJZ!DHI&wZT+?bBA_v}9-K`7zF87?o4HnkL51P!%=^9l>xVLp@@J9qZJf zk;9VV{gP*k+qNKAn&6z3=FkeLhophJ9_nDJazodE4y#~Mc`!p&F)aYTg1==%Jn-=i!mY1tU96-n9Rw zBz1C%*H^#}8Q*`QI)PGxV$bQys6^h)!sj|Cw;T^^Lt{Bsf0!7L5Itbxf!4lB`9*OG z{<>uTCO(btsSY<+E!rjJzbZRAIu?HW`e^fBn4Ny!D+J@Q;nAUg0_-1-)JgT@Lx3Fr zZ-nTyw5Av050SGOFr%q+Y8OCGU*GR)2Igf)XsmTYZ~Q{)DcV`gCw7)9$RAFkD}O65 zzZ^Jh8z7h;Nv^E$$w&E-pZVv@>-wOw?BVB+5|Y1NozR!Nke*!WJ_V*R&&Rd$)1Mp9 zN=r&i{=x~Qh%p|$9%sD8xzl9?XO)LhHt9Dw>2)jQ2YWRa>)C_QVAyI`7W zHV-ERZGAsdA{{K+(f}^7S*=-DGk_n~+&aJIwquEz$ci_MW?p%Cw%d+Qhs`>uLn+>Cjw<5DeDI zdkBh`0$h^2^TC zw$&>NarKO^wa`3nCXh4dIzsw%3C0qio3;%=+7C!s4oEHU;V zbz05+te-~Zrp~7<@jPko*06!JgLxz(wx=HU7Hy+=fq3uC82sB=oJL&W0<`hI* zlb&+{LcfCT+u9aY+mdGTZ=J@g2fz(~GZ|Xiz;NPQ79DbSRYVjiI=Dt!ZK1&rdJNxz zIbC*>5!z_!bq9xiL8EDiodzy;Jji7KiZA4qlEYLP1BVpkg($r!a)jWb*1z6QmeDt7 zua*^-DK?u!{l12w$JT%(eLM4c9`h9_GcSV!UV^)3tjD6HfI?Sxiz*Yb2fitiA@WY< z_@?z79=sYP5K($=NLG5Vb8~*5YbrhT@a4L`?{tiy%(93J3E1+U_1bU6n!cSgJgtWv zV7EgVE{}v1G6GHoX(Hs5-OEn5IawMuyo_2nNIrJPpqG%* z`lj?(RTAcl!OP4$yEGB0MgEmYprsE`lj4W&XuPzj*)jRj>0DIH`aLvr#Yx3W@2j$~ zejmw@NvNsHmAUWJW3It-8npeqtP2l&$|~&myc-JUeIfh?^X@H89oy{VIomCJ|cYkmOa1h*H{P>rX}^Q9TVXYGB3r3x^A<@KvY318x_T6VyB zckMFaq6ItQnQ));((47Q^~-!^aPdVShuz}SfjByPt*GQE=2hO}Uxc9IJsW^4&FUQ| z#E^G8eeI;zgcSC8^O0X3kj$iu7jUwV(_W{js;Z(&SQylRj*N$TU=U+etGB}?blbO0 z5ESo~+5^R|)=6OCYd^SOpc5hzSF|l3#$WJ%PO9qKb^zVC#;)9uHu!E2QNdVM@8!kb zLmryxx2OP>)bYawUn3|hD%5#H2`G$j4BEw~$s({zzMvVJ9mTe~V>d;_+CO=kK?Nmq?X5 zd#k_Mlz;g15#!IGopSyW(}b83FM`#U#sf__MZN0+fuRLo&EUs# z=AI04Z7I`_p}9wGshynE3;IDBg;NU1&jve&i{v8tEl!o|a^nmNGLYE{8{uzu_{Rvi zLUIMddo9oP)j7g<=J0Xj{lOHwxX{+Fb^SY z+uoSyJctVFSh<3)h-auk4pbub$v|h1Jq5-4Y=@f25`W*pfxPBMR(>Ep5}bE*Q|QZ?jIRe><}NID(;~uQDtF z;%>$MCDRg+ie!h|V5c0;!5T)Mp^(0E+be6Dw{hbHLbB&x(R~H8d_Jb$7Gep0hp3{-0t6=cKe4sGXc=# z&~^}el0aXfc?Yy(C)1V)-r*h9TRjFBiRbyQ#}|&S-Y@4Cf$l=n9+OG(NKV3hMWBec z?JCoc9aVu(Y>}%<0Ivyt!gcrO_VpDB4d({@dxj5^EnWRHyP*zG1#6&%UZw%s8JFCK zW}!I=oqVd5RcYu82aV8u2!(?|F1Qj=q%e;2r~x1@wPQM~i1~_T8s0t1tvad@lmazf z`_bufV}fg&+cSQpjYsQ>6-!NRzk`dz&yd8Gww=3@*lC%`{sV^dO*u}z zj5IPK^*>#PoqYML0#N?;QR_=z9m#=ce0-IXCXoMD?Q6?N@2tJm*b~ z@Ac9SE_nDN$Au&j8toNPTuLgSs~M|@1tb0YQXL_2ST?!SVbvK=QFiE;*BPWrVqZ8g zoW8zAz$f;beBDj=!)~{tD-K2s{^^VL#ta}i@8L7u-aTUQlWbuNfj;;m06-fDJ0u(s zdOy#DS6z4MGvNZf$z6sOJ1;WR;Hsn`FtWBEHK>2C?LboQb3z*2)8Qv!YF<{ehpm4C z{!^o-W3OScX~#+@74qRdH+=-BX_?k)L1-wKEAu1!r_N;L;9{q9+|Z+1`7KCWDtG`) z9mUEqq+QJg+;@l~_cPBZV^P%ti-!T!)5k6}d4fmnKfCJxMCAxKB#gT<{{;W4%}FDI z=x>JB=>0M3@$jA^SPGauQrcPx#Ll4hcHbrtryNTeX$ElzscBvfh?#nx2KG1uaBs{( zMoL~9Hnn1^Vx`gZr)oGWD5}3&4GDl~=%L5vdDrQ_$Tx<98WYrT2bpQ64o6zZ(j($1 zku6W?LavC)g;f>Ul*?}DS}E2|_AL?Fkm3*{^lswCe^;f|huvI_^o1j*6$-pA-uN**f|o-EHwgD>+M5fGNwh z3V7!hiQ4@S4q^V@5uE6mRo4m7;Rp;4$1we)2P7n5!g#n@rq-u~|^o^4lPdPp}t`ou!37U+pB8QC%&c9h5rGV z89v?`+hA4~=A9$q5R{8xU(3UuI>r96qsyo5#gQS0_w%Pm{P}3xBWWt9h*La43+G3x zUz!P#R7ur~W^Oyrprk*buQzWvZTP4o;Cd7-DlU|``lH=By8MKvWhp9Tg1cr=KmW7y z8;f@~tHGgjGx+YQb?eA(F2Qq&=+rdQHXx-(c~AN?Or6920_yBOg$mg!CK_8DB%6ji zyRL#eq;eIt(@KT^dDhM(@O^Fu~YRS zoG&_r#$R~tw4RVLgJZ;UemaF^{A7{+G1&7uCi`9Ko6=GhKhH_kAo(hG76<9M8g&2m zh!DNLd~GZXjbIahnE$fNt*lXzIbteE8$!qxYToQP`g(ffh_sQKhP}1`P|o>HQ^$x! zTcPIb$@jiTWKPcsFt=r0u`|seI;S6)g3GWpK^&2QbhD-7_&fN8!;lk#iI8Ep|2EBan-+yS?6EcqoJIK9L zRtx+hrT<7rW)yLei4qORAeadcD{&CP?DrzxJS7T!a=Zp-$<9qYwek|Q9W8AzoR`BMUj=qu_j4pWI zKuFBB(0npeVqAJ<>*mMZ@u|ZttXEUO9d*!%N!xLeoUpajJPMe-DbBvY)AiNU`JPL6XI#hj~-ZCiB-o zTk(&!o(Cl>K2Puzm5WdZU1&o_DZQVnqK~m(aV1(m#ta-H#{`C72Df~gYLo^6|4~P3 zJs0YASG-IUqTcIgo9NJHI7LCD2=`p)sQn%c`Qb{oCxIsCQ(qfuo`ij25}Avb%cqo$JG*|cp%0TMvZx$gvi(--%^(Gy0s^`;;57Gk zN!&5}Jg!>E*R!3g=Zf|?Uj0-q{yp?&a_WNpq}zr6w?Y5fEtnXC*ba{U;L!hQ_og3# zqfCtOtW3GvRVhc3O03$bVN@jc6YE0wwE4A55wqkdG1z)7 z?sT5is43tLRUxnP+5>f}Kxc$ZFzWxQiVZl(E_FmwVJL{W&+_Y!V6{^WF;5^eV2=KP zS!ONSt~K3{kai;m)W)Fn+h|P+{JmP`^cx_onnm1;UDAG(NKBIwi2d-rzADlNI?fuI zz%0Ux(~Q*X3Z^DSzbp6l%@plK!kZOQrJv|P3u;PEwAcYFr+ni&U37$Y1%R+sBDTq z2Z70q8Ub32LaN{j2y2N9#T7YSZ(*uA)L)ENtW7-s5?%KhezB&uBj+YYc z_#%y-Rnv`UPkW1?C&yISi%hq^yB{w&g_)?*!~Po01w64nKQDqBR9vS_ew5)Av0oqc z(>*tn%-U)w!sH=7q@dAa7J`%F20^c3s!5}sR=TuD;*teAT=%W;R1;Qh;Zo}Q!X})R zVeU>PY1E@o!%Dg~Wyl};QYg|XxM#ok!yT#vQGqG=xpE9974tZ>88AJg0PSe%ecQEQ zZpZg-H-(<~U-{f8L6ng|XWQ20j$A}tLJzEbF|@FADcp`fmBkUlo>etYwk~tqR&UN+wDYt#V+Xro5_bGeyw+Gov0Xo_-+e;H zlbx5%)Ht3ot05gyoDp~%#H%w;PiRMI#{5=Tt~dU_5%6Do_f7-2V7FQlP7H3JQdUn9Du}$~*nryQFT@yPSljm%mryae zyJ);~MZaDDsv#=F=j_=Y3D#A_tpX1=BOO$)Pm9mBkIfL}YNGeFCGp@Fb$QhrYdKye zgGd+)@ae7nu~#dzZQ0lePKb)7X!E88)yN5J!-cw%0CcDi(o*d|?u3TiV}qW9KD&`} zv!@dfZ@Q-3wnh61GdK{{wr^Ti-|iWC$nnQ1Yjge`A6))yODh7B(w6~cFsHIH8;qAM ziR8QB)##m>RUC$6iS3xE7alQ)=jtohK&&UjH|Uo_SZ;l*d8SOPAG^K=3o5k{pz_PT zfzb^Az;D5ceqZ-^UjCS1_Lj+C2b>jB1>LScnfkE5QG(35%cN7^pG4bo!sxDi$8C^D z6h&7N0X;CRzfGp?*FCb~taH=c_Qq5a4I9Y;Eq4(3FwRcZsZuKK^CZkAsBCpAwqrW2 z$XNBv;fRY|tNZ@-c~0)z)@!|>SE68x`tUkU$oI2Pm(8g8>Lnm>NG>TffOq)E)c;9h z|9kxoj*?jzV)NBeo@==*+~V3DWQ0IF9oc?3eV(70o&||z7iR}X>By}7eER+IJlX44 zf*5#B681wGxnvES^2V$LJ3OD{<3zkC9_EF_kaF;+fJT`VGrJ+i#oCQX z;8$8kQFa;^PBv3HqKjANln$hoyP8(BZKg~?pvtKIZ`{h?uG{E@uw0G3VSY#Fe1Pzm zqMkj$Y2}cnWTFPu{lfP$_vZY!9O9-F!wV&X<7%{*T=>2Qfl_n!O?VF_Hai>COb3w!Z^U7}rSV=j9Ni6i>hv+*|6E2;uU zcBu2{am6v%jBJ`U_tKdgZo&+#+D(Jt!*p9r5Iah(9iyA?AV26j%p+Gb>S5%?3S+O1 zyog6<6POjh58H@%u+BBe#*C!-jbTtyC`GGP?(+TjUR|PeEJKfkOnz~mEJ)dA`C^mFA{ijmo{o~k;8a|BY$3pcY(tYS%NpoMI@)ve}1hd?R^pT#EQ%^_w=mBcrM!e7IB3`?Mo3@kZ%Ji zyra+iD`rdFc)5<8n^lQa{wCr;lZbnZqvPYq((RcxyY=2=N-pC;bsfX|TTXVyqwa0@ z-Kq>5!1Y=91j*0-i=@V?$73>lhJ0?qqq*pf+J|AHQ!lEdf+v)YN4`Ii)`;EI)kF?$ zZPG8<_+vn=WiSoZQh0)B`3b{4`HCOe!t0>*)DoLHYBBrbYY^!Yy^xJ7D2$=Q`ZmCv zvemjLf_Ea4FEOm6nsyg2-8P1sc@Ys|Js;7^*#3%YhfZaSCY?{Jt*RS$Gn8*rl2)*& za_>^EdMID2vvw+BG-uaA{ZnJ9lR(MM1D5US_cm^yTzew4K?#snzA=Nb7`!j#s?&)w zlg*pug=a@>Cu%vTyBZX+wdo2&C6f)29HBfb3_3hETyIIh$ z+jm2jJ{UP#KQPm!>R9^TkpbEF%$p$(eBTtleEyl77uK|X0ZL4NpO~YU|L*h5M+axu zR{I#iq8y_!IVmqm>bt796=h>G_0KE!DvEsZ!1cl<>YdS>PT5Q{!ciM8>};wG!Na|! z?XbRpan);?MIso3jsS!HNQXRAc4^mWv)&Xzt1WGmcm3RLa(%e{)c}4)m?MNmP`>nB zvstg+2>i4jWWURMcy_P%Dx?P58ecC};K4V{kHkEIQ}31XswNNoq3GfVH8`lFtizY@ z)ei3{VPDS$HhseO#xGHNN$pK)BHKxuL1+si(G&J-hWCyOYANYpOef+isL)kH)@w?a zZU!J0-UqZGITu(GQFa8s^Z!vPa)y7`%m>jPs72q^`>R%b4V^Ae{aWNJAw}&~wPdb@ zdvmZyM=Gso>iNT|KU5uE$CnI-y!Ja6YR;7ilsFh-`VER~Md~J#c1BHAhy17fvZ-(d zg1=Qd^Y2Ln=?6R-4f~#*ffvh)FG&5A*{?{=5#hb!TWMTCSXuh;5N;jmEkB^&CP7GO zPWRYdzOIy;JN*UvXW5q5r%MP8wG*X_)^7)*)$Li}rAP-z}L zta>9y_EKhI)i*s2n@iZ-?0W16^#|k^D>8jMLC@6_qpI1w^A?>Q$lJdfDCOQFy8M0p z`khPb)Ggsn)|2VECEx^b)E|>D&EF!$$va(S$Cbe`^d6_x`;}sb_?BnY-iiPoG)#V; zAbzoO5}}VPqe$FoXzck_hsIk)JA=|$eCiN-u0V$$Bp zO85V7dsLtU?9o@dDO=!T<%3}t7+^KswNLHNi-_xTi2H#b1dBdR8VFL7V5-=RVL8V- zoHRk&W?h_7?ib0%W-o|>4oO?1#24?yHl%av=wJt1078Vk3xX z6d9JCx6VM`_3%1{hZVt=go1VEsfm8ejr?r)U1lXNAD1p6wX!+Mp| zS-CMr10Ou^zI?i(k@CLf!kJihtv@G!*8&aEQ+gOx3*L!xuCgc2Llk3z+a5l7SHpHk zF=c0TMXl5s$!^t-tuTwxb1HWr$S)VF`CkmlK!32y8i(K7?B8HR3whiXytVgk6VkmG z?H?zYb{;zoBO8H9)ec)!U{jp;V_qFk9r#^}6x&^Bkooc9BiFjtR*c}__3rM*w`bb~ zzV7`Vl$@f9olK!-5kjrDG2qSz)SPO)g)e` z3w6UHpdZldoOt-Hu_yN|PCE-noSnYRS^zxNy*`iT{zu3&2eBRl3i#-7d*aojUfF^k zq|YS5^oFJE$YmvhVlafF_k0>i4Z+7^oZ{Gv0h%NiN=JUVXB`zd5gsHaNC z)M%W%-KQ!|Mw09CgjSYFM4WQG5sWBXkvsL0fQI#@LGEG6YT{VZMnb#Mb7BE&#Yx*m z*lQ@r!mB2?_ac>mj`;wzT6EWLb17Ano+{!O-g)*QY@sNxA-*I!f@U|}m1c|6% zG)x7YS(iubqQSBSh{djkx;?kk)@_``n6@iO>E0Ds*P?6FnDJ$-g$@RxY zy8WRNslYW}|)ifF)HcuVG>G)S4{1o9=9WremphD$=z!MDtx(Rs&H zlnk!wbz0*zSO!#4rQZFVR!==Tk&AePamAUMbg8aw^zcgJRF*By;{)1yJ1C30Rto>8_s;n%Z$K!_Rl%)Qei$1 z%h3ID(oGy{Xc%%a75{duZ}soSE+>m>9prpqoE(_IyZN1AT9 z_iR_9Lpw*6!X`a-VLlP8ywHirVN$vI<=wx^WL%O=vU6_G{U;Q?@~-1cmC9fj@d|PG zHXLT7fL{yJbg(tf(Onj-;~sgaHbqRjkd`DjcDG$|@#|>rls?_ZjS zKUFd1%^dnNd8d>47a>;JN3>pMYSLd3m>v+kH z>`I%{I^Ug_Z$SFseXmSz+UTlpq7Y$oJV9jg*~9w7^ZA{d%+7Eq4AAA|EIwE-^B>uWlBmgi*tohd`PaUT3AfluEiMA46xa1lvm0aJ{fOCXG4_R~V1Z0! zSOAolHniaAx|fT3N1D+{{ott18V1x-d9BFhDh=uWXf+aevk{@XfKYRlUp}ctu8|Vj zcOi`BF}|P+P@z(4q%B$k=9AA=%S?4w;gR-MTO#DjAHf(RhCDjt!~$NO*fw@~lfSx3 zfSaxTxvCEnek^zs(Z%dz+x#2Ym;_PbwLgF&JFjtjG9!3|Bqa@N5f;ba-_whL1{dKDTc$KnQ zbwK5L@I0Ki1WN8&HoR~L?Ef{oHz+B?G4yfrG#C0(WlV)PON-WL@=fDvWsA-v$OiB2 zUi!-Uo`scasy{Izc!X&idS|%KdilfN0`$vV=0$hp=qNOiFLQ5kSA5i(tyhEA2t4%<>6*QxH@ zFo->?))U$MgD0fess0ZM!z&pxxZ?O7(yZypIJ+q=nr`S zl`*tzSCiVYtN8`^E>r;?kY=E=+m2o330?`&)~mfA-B}qb#4gG&ePTmqzj^hAQWD?< zM`A*^C{@JeYI+YlQTJ=OuwRij#TV?&)7MuD~rR)?nI4X>l$Rr z%9l|t^<^v2Xq~cDL4eOU@H{SaQLumgZU{;kE&a@I#8EL9+&9#Yq`U4cuXr$0WT1gV-GDPX5wI7`7>H)GHx7wZRVj%;de0RL)sUatt}uDj%s zrHLO+d*E^o_ajSv@p0pjA%`sw`N)N2p4A^CV3qbl`EgOj$opl!M*3u8~!qB38$&LxbZNrg#X|==oV*q@;8IDs>LX@BJ0_tt45fjoyKdC`cK@p zWBzPJ>P^0EIX1f2zFGG$VuxGMadK;KvCM2;ewcG~>G??NRaiRpTwOEYgYg9?7bQe*RHivei|ply+{yNh$t#+iJ#olmpb~LcpEJ>=2URY> z5LScU&81+7UcC(=6DlR{UG~pEY4nE@LCH9aLTEr&Fu?-8Ep%8l6`cFtVJ*jHYRcn> z=a`x5li3EmUgy#fVdedRL*HBJhGlK!fV1$Q{)5^%HPz=j`+x^=&gH6}CKJt5KKio9 zt-B2=r32_~%Ul8agS)DKSUoJ8&DF(Lwue|jAz!wH)BDkz*u6)_DuXd~6jfgRpgM>c{(} z7&wNXH<=O(QIshu zz9nOW&vH7$yHn!s+%bo^2mb|OgB>AkZ1BO{ioX_~9V{3JX^Rr!>*my|ePTT`A~406 zI7{BYH3J_-N@xFoP3xY0^Nwg}^kPc5<4&H*IbdOltqNKyj^?d6$sQscLg+{9=CJdn?{VoYlz-hqDFJ`g?P!! z;d1ij$vQ`k)Z`1SqgKM)M;lMW+j~cuDQF9)vFEavo6rLm50v#K+9-0+7rtysiqXi* zXGLpMh1_3Z<^tO<8}5TPG{r@y?y7iP%^Ki7;7SooIW(jmxR+F!A^mct_3rUQWe=Sw zgJDUjx{C51cXdRh2j?~!*M+nagf_2#)uvfjC;B%!Bo)Ombh7#DY!)?Kc*L6dfc3t{ zP!U*?fgVjj=vH4@l5Lec%rkr@YYo(A`dbp$O%cufB%7JfPf^o) zTsgn+pf2X(S2F+O1;Dx_9wDW!!GsMRm-LN^nO_areFzQ-#$*v=w4!_rTOl+&fY{zi z!0$>uPoQS^v@tK%RP&006Q{qJ3JZbPmB;+s?RB4|Pebk1$6X6&9#U_0Lqh>8w(}=x zK%UJK6&23;ap;+k?{k+k%y}|@l}mkz+|fT=&i5V-VgU?oXOg1>Z|sJjG97Ae`n6swlS!yn z94!L7t1vkHv69DasvpGi->-4a+f;q&DJazVCPx1c&f)t@IpcDHB}>K|himE0+12+z z!^WWVV3v@@_cdqVm+|Q^3RZ{7?494-hZqXF&SA%X9a)9YaW2lqL!XI3_V=C%-dh)0 zSGBDF;c@qN;a?E`|IRr7+JxzDpcwLr}Zq& zPpyt-bR2RV=pPRQ)#Yh8^;&McEF(YfYXzB~rMs?qce(pRt_C`I3wTt+OeI{k^T*bT z9tc4_QysU%C?Y7#fC2~ zlR&0v@&iuRl&VuN9@_Bf`9tzNh)p1P?}{ooT!ec4wc`Qf9h}LP;u3r*i&U}^|25mQ zXMLDQqpW>!hJe*%zsYlshMqZ*e9y=C3Y}Do-yCa$FF+Y|v-l(J=7x1l2AFTXAOPsn zz=|S#Ve%(lxCQ?P1JyzZ!xK@5rM2VI9^6H(AC;ULbkRSlQ{e3TyeSP<`%^ zt70=M+Hr4E97B_NukcYf7QL;}18(!%aV;2Vms($TKMg9h$=E-Sg03S;0&)Voc{97@ zygc8-8rd-U$^Q!Ol^6iP(V)YK!e0Sqh#~x|R&;qN;3VY?*s$hSc0!(grVM(EqRyfs zdM=8=f&shN9Ka^mkKJ^`vH5FupiIM`Hc>;GbdQeU?0$nkBXu;*p;H1c6(MRQNY{oG zqoulLXogd&)TuKRKNhmgH4!a3-=V4>&Y~YFGORxxmB?dBBxX%-YeU{KGUzx3Pdto7 zq$h9Pb<=`u$7F&uT(;lXLUc_oYYk>kGpBfYI4I9Y?qinggAX8IJgePq%EoLz2m`G1 zuU`Ut=u+)GUE3YH^i@2v^%26G!bx2!jpJlj0~%-eDSb_NPoPfK;2tmUvC0Idz30lG z0duhuKTx6O8I`D@y}_444%|LJ7@#*|Vm)NJ0TtDR=;fR$CSAxK_tuwXWn>u!oQ@!> zZ69f8%^@2KfVxJDsE8-f!w5CNN?SE{TAyZ{F_<~tM_d%xw)%1FKl11~2HT_VwXE}% z!C#B5$VzYe2N3R<2*2f|H$r?=ZcR>>Y1)~2nu$y)$Lg(dU>h8-_go+~sBXP-uWO;p zJczN-JYxp|xnC~agb9a)uf(tMCXp4dSeGU$@ePw}x=VFiye!||m79bMAHSmu z`jo)C@6}zmag1ITTR?ux?jE>4IySBok>GLn$wP*>BLmg^mYok{8XD$A(VD8iuK}Ym zPv|`>1%ICSYuLP7uII|yHS5%@Vr+tK4HY$Thi+ zRWntSFBP#dVE0?1pQz;H8P_yCOxnPYBaXJ6Bpm`n?WdR_=cdj$i2tl^3`VgRpoUJ%jT0mS7^tyk~`Jyo4)@yk^w6gIGB+RL!N&} z0Rj3br)(dXcaY4*zPeCA&}QVg7Cm36oR^%{gzLJIHW=W7^)W@Yd8bou@b-Kr`QTIU z=fAgI=QTpj>c*Tf9jC*UHfabP-f)kW-fHiq4B^k2gjHHehj^0{l z!*jnbtAx}WFWOeD&nb|0%3a(O(7d1Kf%=bfqcc=)gLIJk~H z&(K)|3K`fd*1R_pb`rh*vq00{l8*>86s&k;;E}C3b#(auUrqKeG8EHMNNtr{RQl_$ z+;DUT)?Mbh&D+f>fOBe~Ph5!Y*{cGIcZ#x!X^w@T^N${JJK{4aw4ZJ}9ZvM_PhPHv z^7g_4V_O60talQT@8I=-ypwwu5zO^!lC7l9k6TW?3w&93J%! z>`x;cfA*SwKQY$ixIoR-IISh`Z7<@6_>ii1c~Qg$&qiOS$J{Ct39kqPl#T@U{qbni zm0EBpZ@3)7U6haQgRYU8^D6PO-Z4*`_-;7O|7mX!X&t-GL@;%6#T>Y6HE78*d0)#H`W1!i`jY(Of_cj>h!rBH>ZAa?HtoZH=fUSgGP+7k z_4cc)@4ghdIJi33y$j79u3sYh2dmk3iS)aOs?X$px?}bIzK6(`bKZfxR2U#_T8U{Q zWgtH2{2453So=&VJN#XCwhmB4yuFpc=3qR0Q^lyxenoWaH@l-i;de3jSGSFKGe6YD;Z#()F+5ZUwljx|`}2w)xg}sbTmQ^Bq%{Q390LyWS~n z_u=I&+6;GdhJ22bDhDgYv1B0o4>Z&7{BYP)F<BNZ4B216qN6V^z67lF^#=sJ zW!6oGGPe>fd=fBPD76yZX4F}$KJUF=Hq4*MQQN~`wx%luCXClYOzMmmz*xt0uDaUOyw`gvjP&YvP}I`1?;+qHGFnzMFww37N`j1SUFFM{ULE z1NrYMAMo=bl-=8G;UKNh&oCZEXZ3E4na?Ckr@Uq_R#`XVjSC8V z^HEGt7mr%ZL`|L`BhE*zM80sjez{PQR(=ua*`{lhu4Kng*MyXMVx1VtiJzik=m3 z9a?K0g9QO8MIHjKI3r$7xr)|mIcbkPQ&Cpev=85Nc>!8;0`=H zJS@H%90y|jVuR_}6tI2njA~WTIf-Dj&qW3&r}aa3O)`J#@cF7z^-r z2aXcE-hnIgj@!DnudaLAv~4qaR05*{N(vQf=#v{zatW8o40!cQ{`jqq<5i7! zlZCC2^8Grw%fDFBLbhHm*^ zfh2e1CyDNSMFr2EssnxHQ@hk{`vC_oT0?==(BXVTjlj=B)ZNTL-!4oZBc1Pgh8|ry zWfah1@@svwaC8=4{b1&iI^?f20)BA*f4>_ttqa*_e)Q7#PubKu19_Ln6v9P}}PR9c#A_~m% z>wrBI#|tw)#yXSCM^5^vGwKgP`7RSXsVIY3}u&DEQPEkvhRa%mk`;qiwP;R zjAd-2?6PGmi6mKKO7@a{7sgn#ml?aUjOF*8?&tlz&wD@5Up^!MbX{|u%W)pZaXJoq ze!e6Z2)JVCczw8Zsi=YkY}s)PT0I#gk&sz&`irMdmH7+SIO&{$yHmmIN3?<>EL3S_ zz!FW@pR z*w1cJhg<&nwbMe{bP4Qg){pr(R~Q&w8jZ@wqI~K7Md&2F+0%O{TG{);%qtmoCgoL0 zDrLZ!r|SW*J1rsEz5akAiV%@2QD%_^FmpF<28NK0AZ!`5%@D-4bPX_eO?w1n2YgB& z_3}SP(^D{@@=ykwE8_o@YsbMJxBU6UR2AUYp{W(o03Xo*f!`4saInsgzHasycT`(W zK3WJmbSgMk;w5}nR75El^gOx_2&c0a`3Vwj zJepr47E>{DEYb|ebXUcvqujdX)K@=-;G%D)PEBSmjLGjZ(1Wa+?&Pd*Rk&Hcse<}} zJr6=Jc9m=fVhbVE>XC8rz+cWo_6XMPd}0T%V-<*?$nED!v7I$Ei2|12$V#DvP4-36 zN(mdl{A#EX!B|hl4ToXY+k}#aeB$>t>hH0$Y$n~mAndophoDbA;JbG<+qES3pM}MD zB*2|r{p7b;A3S@P;&>>(F7j74-; zOTN!2_CK>?2(-Hu~sg{qS6%q9&uT?BlGq>p=1)6dKrHNFF}LvsHAQ`O!DU&tOqH`h%9S zXK8wj`#P&}o&z$*iTmO|gW|x{`0@i%Sly2!DBEhVU;E!N|9{!UCTu_)6rno%-TU)g zUaP_cq~&H4*4jJU^Ip}SHA;CwE>8t^P}V0UM{G`_IkWb8rZXS&>{UFh**(Lo7`uuS zJp|Z@fqp*4qDpi9$gbXzGvo(BTp-yOL5CYThPiH65D~2Du&fIIT-)!26P0t0tIB8A zGmget_A`#6{UfclU?b$#2OAOgp9kLEnN=LcRoldl_^yFR_vq72$`bn?)r5_X^VVs<1~<6d&vCaoAqP5bob1| z@FF_y=Xj*ODud`T&A!A1d-4lu*e`@IrIje8>s8V}Jt115qN#dnuZ&JJITnFsVFmdD z3zoB}o`(Ki2`0a+{4Z+)oonY&T!YnE9md}q@HuUJ5PhV)QxB6-hr}G6B`^_>z{QDW zq=MneKa-h%8{GV(=jDP2lWxdaHl52=?)IH?PLfma_?+Yt3DSJJjQ5xjxEnctE@JMSt;{p;LZ>W)fM_89M_(5-#@5EjPhFK;_ z@w~#<>~}aU4?i&I`J}t*PMMi}S}k2y)X&;C-|{jg{%6JM(deJTbc~DYPY0|*aw}>* zl}46}7+%svwALSt){}4hL+R2F^{Z3=(7e<4mj(LE_lwrA`12M;H0&RN_tHcIRht>2 zrLVnC*e%?IU%c3uR&=x2l=aMI~aKY>2bA$4bC3VW0yzu3eX-hL)1hNlS$$`Gi104w~yt4W>fM^Kxi!p4H z%@$TU+ABHw!aBAfYbT2kqYtn?+8Y)sk9BlUp6l5;su+HJXOgc?yU}L3D-#EUZ5+CT z>Qs&oM&cfCCz_tA2-vx&n2_zn3gtM`hLAEmZ`t(jC}U5-_TYV=vi+B<%TDGPw zwj3BR9=5#dMl0q{h0~)xI$P;AUL4 z-(Vu>MNq9|wl7LZ3BGlEZhAVsAYijk`y&mldcn&0Fkn92DgUK*ULDJ{226cfVDI9G z`g$|a5MxN^_jP655wvEAwi_;~m88AHYJUXix1DHha9&JdIY204{qk=w;9oL987L4F zGj?wK8~!<$o2`g#U^@SK-Ed zhPy%T4X;#0@DFVigg?klDCV4kbMS5GWrm_)y7JMT{rV$nz0rJKgTjnVv8^W7oPe^s z`ZY!?BI6pw&W}V=oj=`C7i6ba9(8;%ttfz&3)4!{sPT4j!e`(dU4}SQcG`4h>KDYu z?$(DKin%y;YM5_J?g$x;5kEraO|xgwR{7muJ2i|6avU z93ivtRt%U?sN{2$3x}OAzl8QJVL3nnX5|Y(*#`s42Lq%?<^7);>j6j9$()0!lgepQ z?X*z^-V$7zF7+a4Zy?7Y4aL!B$|P@%YX6qjD(?Q31fd>Tjb!A)ooZm0 zS}be?NS4YpJte7n@?iP)Q&Tysa_g@cep9&)=s{bzYOLoh{0yM*@U}XsQsa{(t$1r; zJdvyE#Np|dQke^?zYBPS$gb9-Td~Jq?$0P6&XD*HfAR^gF%#;k)%6(%oPfXTS_-Gbx04i;MPT^2OB zolK7#LNWA(d}f3G*G}F_$yw}R;06gE(h$Sjnyq6kMPbfOfqS2)BE*3IwgNB1p*F;x zRdQuLNR1(6BokOKZh+VJICB@dllJx<^j_+Jdpq>Bx3hZA|LmWG6MzTrcUC*eZ6))^ zwWi1cJ=G1lhh}c&>j?PR*%!$|&-ls=A-2H&$26<6Dk8o`=foaPRtF@{M#F67ezXOK zGNpj-1v!3)#>1Ok-eU%cV%m~Yzqf;qwhMZQ#^A!SSi1x&@F_^yg}m;5IPdNlbTAyG z2;G-5s?^+jc>t@wSvvu49jgb7d-!oU<^6BUnJ5(v7z36USscAxDkcwV1c>%fK1`XSnUs0K%=K((r^a ztw*zggFQ%I-mEJFwf_R#KLWeu44dBEP|T~7s+lOtVA%t23tg)1@Qrj9Xg#c93HCk% zX876|D~t1}Kms~`!tfg~iJd+@?b5C;vNF-(9WM?0%fa!yu#?r(-8@>o=^OZF?|6m< z(1=rh7nO7yWL>`sRjHfYu0K&m9+((BpfN=e9TFp*6_mt>UrDAwd9MqOE=hvSwoJgQ zqfVRSZ#iU@-H=%y4S#f`vgYh9NgS})_o*QN;*2fe#AKbpQE2r0~%mj zP%!BFCnxbpGI*{J=ng)7qzyT|WZaRSuYuM$Q9*KGJ_wcVeg1PcucL_mR)p7&A@ag) z!%N+=M3+AudOi5b)iBa;oXp#FE$tpIgHQJ6|O<@gY1lEZ?~UB1S+9j#^V zXGT?#GKYfC*viCPUB(%*Zt1ihvwhX#m`QA1Zika%=TI2hU}cX+MOTIb`=wX!#xe^(>_KLRz(BY>vGD+j#- z7!m9^8Owh{FxXD)L~n1xpOOj!YPK$iNB zdy$WfeI^oq7h?f=2OWn((!c>DTpUtHi7<0mSsA7h~W&bpZGg9ee+Qa?s(wT2x% z$bI3ZXfqP{(0+rvNc{<5bG63zt5{TKU3C2o-<_XltrX*<4}dj?0q^bvZ~9EmbClE{ z?(sK=gF<=v<2}ROYg0`?CS3HCW;G@?wN2_|7)7|GhNb06$8uf}ux&fQ8*oM(0dzt6 zN9Xsgr-RM`)$|)aU-1wBOkHvSEv$#WKko12D^&)`QXSg!!jxKkGN9hQVBO& zSn>LT{$Xc23DsgDBFjjAcZjpbS8BH%8J>fb;V0X2_SQVSZK>^YTNhQw~O^%e512^HQ{4?u^bH zY=al`nHGq&Lxa~+#-FCOI2>vfeTGUZ+a*PpT^cD7NA8c5nGDFia3cI&0RLNEI5kBp zf@B>Jv}mCk$c$>R5h<;hoMElr^22I3B?h7j>ee2RU;Z%2bM%W446$0SB zAVV{OAV*9SV@)ZIYTd)(Gu-aX_zU0cjWzoeeBF!7>nYB^4hU-J8gQ_HxnxssVRwEB zp}Z-noSQ*KzylITP925e-> zza0>#PS3 z%x(Ijl6te^W)8G+#=o$#79i#U=c0;2-QsMZ;d7vSer|7nzhu$CCdP*oPs?k9p7P(} zm+|_sK_y)2e*@+L(R;_$_CH9@H+`D(HahfNQkwxm1#hTw3Vp*g5wLn{G-$SJQmPdr z#~1q?UAYxMZW#U&5=yFX_pp{O5Uzo&u_*XKMI`KWWX4lL2hVPxn*SXxbv> zbPOG^P?UG^3II4Ms1?$^aRa$b(K2&!Yh z=`#`l0>NUu<70G_q5Lko=^Y#zvq*@VV4i}qBC+H7`R>DA45~S^fa10>O1L)xXAzyr zj~*se)nxMvELAQ=ziGZnzBA%yZp@07C>KM=*T#1F((m$|R`R8jl#6cfMC-uxDd%W7mUE99N1BaAz=mX}@nln5>iK7>!mt|KH(EvHQ z=Ir=Z*lI2S*mu7ZUM0H+`L`6dRGFSUNK%OE+KX*~(j%e#raX;A<93+=FdsP{=G*_F z5*?P{1?!40V;8p?gSI>dj*0Mw07qY3x+0%E1u?JgaD`W7#Ya>GL(ma#Q7i~-H-$7X z!~*)uAHRzxNv!-OkO`VA^KbKbbGh*1t76W9sI_xz$&>nEijozc3P z-sEKG`k!;124R{TKytQL$)7T+v>AY|pyoVtJhLk`8{PagvTK2x@=CFQZnB8BzCE?g zJCG<~ubddI5U}|{f39=|z}bPi+b+&TU*$yhQwH5r$YI${XM6KhvQ{wiPU)%gXY-r9 zI)JT;R6UQjD^==Fg36UdnKvVqo(@^bj|7S40CQ8N(A|uxdu)J|OHqrDvip7#EkJJf zW6OI*VexNs+~ZkGCzROxp$I2Lk2n54Km;$IDqK0wVmEqKOHhKj>L6<+z)AadLa%}L z$cp=ic28mU(rwT$n)bL$2^c5o#P88)!%h1FC4%vT{u~>r#*HtBZ-Qhv>5sViVFQ_6 zPuv~|?JU!{Of=$}xk}a%D?v7-!S{PA%VG2Ie+ye`w6LWZWA(=*!zKo*63x;ixWKuv zV;H!eq+?&(8XOhIqjIh@=~yIvNO=Y-5ENei8zVlTdY->1z&d$qt^r+vLZ{}h6;%$>(j zOTPAl-5hGC?5A)Agdhdn)lPJ%`I*Q-S7|aX?`k9xN%Nnu$8}%yeXz8akEt zY5|4xGcn*a=-(Pm1TrIY^>#4*wvlz|@_QT74b zYM~en95UL@f{jPcbGm6(bcTni@ezl|!!d!S>nk_3?_6w)!d1d#c=7`$aIqTArUcN9eXSt#ny>#Gm7$M-=8w z#CBE2?!=Y4^HcjTSY~)nI7mZ#4S*`YuFU8)2NI)2Dk$!C8``nD-<0Z^Xr-Mfk6~nM za+Ox1l0J^mxC}2F%ES>nBi;{5eM=Y?i`OO338S5X{nbP_5TS0<^Sl=e0SOy8I*W(S zc)XkKb+ggkrg<5!*;Ci+^;){R9!S?#|6e@5^c}6=859~=`s0c8o|A!Rp6$g9iAEHQ zZN;gVvNC_9?L=Nd0=q@{i#AJfSV#p|MZ%bEDL!w~^{sWCE1>qWUZas0h0kbGiDJ3Ne>bTh`y9@&ClQXo1IiyG+h&j4*a zdV^boTeG3OBusH25UJ$EmncI{5Ob+YC-+&pSQjE*StZs`S}lU!vvf%liEFR%uLPSNyej(~j#Nn`mJpTL&Pr?|~-8NLt@r5C98 zOo?OqrDr8S*)6dAKsE?9Dlvp^)ax;*T;r>nxcAR*Xc2AnXY7xt&Hgj{I{^!RnaG&t zrV=YP(%EickH)9)-LvjuYt4#|xo%s3%`7a`#&#)!rkVKm%vO#d7_WKFnTu zqft+u7!Y+1H<{R_wE<0g0BuRocWcnJU?bdlaJkxzje&^O%%CAiq?tJ7>$ai1hU~|xc@(5W0eu1yIGCsy!V7#$}J0RSfbpUsQ zn4? z3c<*j4|Ji=)PXEW=TVvcpd=rt=?YXEp19O49Al; zlv=+s{9}dfKP2T8T0A&F49@&%Cp2v)!5$GYd6(9MkMXvN>#fo_-+)a-K!0noq_oGR za(T}&OC|{GVV1d^kDa!;7o&;seT z`a~~LwOiWJ-7IcHrsc^HSd5F4EAcj+E|G=o<7L2X6+2lIsAfW<@eis4cGqvWAx3{y z?AEYMJO4#45o4gYS0#eJ-adbl9Ly%l(tlg4bitilk;uM)s?M*RwExYd!r68NDkS81 zXUoJmdraVDc-v(BDnf{|<<`x%73e0<5w@(rX$9P&SW$RGz+D!|^w!m8rAD3CiW~b1 zfX-^16_*%%P8H^6kxjinKG#^%{3KzR0@l#@-7$8w+Y_YJW!ZTX!BDe#bK{&UGfh58 z-~ka@Jq=nflpX_6Hy$@0r)w@@fo>r$h4765v1|(L)YjPl;;kS`lkaz^RQ=5^CP;xG z7I5=~4?GL47{s$|J78SQoS*_BMb>^!fz;w8^O_e4e8bSH)&qM&1gRk5f@Mlv!(xDD zgMSKE+g>E{M@0-jcTix?LTqNhn0yP`aCdMq-eR}YMNc-w=u5KJgNXg%YzP1R_g7Dr zkyZpVj&(eb4Zgq)dEWclAbIVuEKhc!6AJq<NJTIfI9QeDdIBmed2Q$C zYdZe90`hRPGhTF1_jYO95n+B6l9X>jOU`=#WOiq!e9}&Ny9ioItf3eblmIg7P$k4p zI7Mz`RO#-+GkY2grJHARwsJO;Nf|350y_hzOtf8Do3Hnz!77ro>u6}mD=xLCLrO8e zKvWk)`r=)XMP8AH-U4pEy6d!k8w-^nlA^mhbC4fi>Sku+4>p?bvA&ih@Hj_y=!TCMuzm)qY06-!&L;XyKRTn;;C43hzo-*F@jCuQ+74C)zwyzJ65*uW4vk-Z<>@nX- z(1f-i0@;i4Q{94LgT|;3CpCtuarZRavZXZ!km?EbHCK~jg4ZM|q1^<<*d?=1;&#&2 ztnw-=4ouA(z&gQ#+$MraanEQ5X*&xT7u17GIep)-U-pm&DN_rD4fzKRo$*aOoWG73 zshTw{|1?VS_=_gFJXdzX`fq;uNiDjB*{pihi6%8&(16rC!)yh)+G-vhuIXf}FzJ+} z^Z{l@ES-^e(m?F|rlE@g)pGQXf#MPvI;^Gp7w>1e8`df+2y@6^8`ql@V@eI%FhrJu zL);N6@$R;TRzV=~{H;kplwKZxu9U4m3dNNGSvQiuz!(@iX72$fc`HTJL}^!EpeYN>l;3nK5UJ4no-L`90$qtF@$#mlbI+=Zj2TMfu}r+#=oFTCscuI zdy#7ybq>{!TSg6${V0_Db*TqFmUh;C%L6^{ky=Q6996C1S5Tzq3iZ1)cdVX`wy^Az z%LYv!yEnE@6C`FVP=)aa*ysH+zx1DItxBAQ?=8^mnUL7ZPUZML(#LjT;WCKq;}LFp z3Y-#p=AP+=Yk2M@vh>Mk<00AM*Zofy(zSl>c>j(ZI`Dn4^2aC~5 z6>?TsV`sb(TM#%%hW{3KxpkAj#Xk$2lbS#uqY7PG`x|NSumQ zX<5{Ha8_x@rk~jOKg{fNdDGQ$sg&oPW@sH9((Pv|h_+e5c9n$}i}*KN4}NA}osPc@ z9fMYDcSIYExO0BZemq;ce4P?ASh2`RI)t#4Aw&mEgw zYnHUdB79%-HW1hJLrld194*WUH-qmf3EXqjC7nd?`;QSGj5^I+`m1xpZ3G^)` z^Y6P^6#%Ir;Ll~BIH)3ZTnvWoyFJU}I3;*UDVosJ<=qynxj81cEOpS0Cnt;*SW0-j zeU;0WW-rEurD=jhf1B5X7EfV<;X=+8TfXU&1&U!)+GDP)IE&J3{-wYttp`03&Lv21x1(M;| zx?-4ndB!iH+HUcKC(5E{LMqVc7cl1964Q7HHs?&zRb2%idNlyM-`99AMn=?6M{b(h zE2_Ty<%Emg?5eOc=WA8U84m)fMz;!~0x*Lzey3ATHS$plyOJdD!0^wrC)uZWMIN}$ z_>C6IgW<@QU>r#=qNHcW zmrYXeeY^DTyCtx1@qw>n>ur#5)^Lb(kegw%Y!9lM=~aoMAnerI*-?G-dFs3&Fs^U& zF3PhFaEgo(sIZCbR0K^w+P%y`OgYS16Se!)uDqjaJB#{7R)cA@x$NG| z?7~hM9Fa+pqb|CAI-8v$yD$?_0A0^q=^Js%*m7R7Gr7&P)kkrWubA{rJbn|3ojmZE zZSNQ4!MF9Qjm18rZYWOIxlh*BCjD}-CNg2Ws04?Y{^csbLGvYse;|vE=N$0*)+~&D zfQ3v==U`1HrLMR5y>S}M2e_b|A!;vTU!;_Yj%(~9D9-P`5ac5d8lvXpaeTso64 zf1YC*@+>>Px0Bl(X6!5GZyD*?v9a12UC3Ev|GQ%f=t`p^7Sk3;l%x!hGVo~TDX!)Q=6XK=8NYsI=WW1~G8&v0PC2L#>f+aXHTq2b+#&k1S7%o9FcWCLNcme)4W~ym6gJ-X1R3}TmpDOM z7H1ye=LkX=K@`Hk@`bvOTXpuf73T=k{{P|r5B)=*dMiTyH28%G0OHYd+@BI zmw&z9gSz2@*>jGo_9{79GZd=nx^uxX@eAKCe3xk?EICxj0}LJt&dDX3#Xhm@6rNFP z?KI-#?!jYO?>hIdT63^Q6XGkm6TVdPCn4Kw{9aMx;@KbSuE+lJ=x}OGCs_$=?wGhm z)@68F@G%!YkYBfOlLWkBib_lo9BM01lcv%rKQ>FK#Q;z}SW$R9Tj*Rlr=z+z5 zXb{tA3Pc6O|NIM+W^huQ9#zDFCif{0tgjotVXB45u<6}=3_@ak40`NCQfvSExb^-tyF6la zh+66}dcA|qMfgP7D7>Ju76AT8t@&GB zp$fx&*^NCM*`5}owgVu%nPE@?YhJ`{pPE3eP3etFPaeaeI`2r-+1-`2fXIm=_t+k` zLaw{&)ogfzG#@nr(@cmj&GvVD>S~L$5HG`I4p-qf&r^rSBY5)kHwh(LHWPwJG+&BB z7PZ??+AUJqek4voRbQoF>u>(z`F`)W^)mFI11(^5NHd%`&Sv>9R;d82p^@vP)a4mt zEnd|CKX-?wYV6E8YV)XXa%I(ER z{_qjc_GzjmXjw=<xOni=GMrO88KV8c*!O znM0r@;4$1Bu0=H&YmOeY9QLD3>D5T_l2MAW`Vq$;COU^AT?)^9h2{0Xn^;M*N|iaE z8DA}2Fa}6D`uE(3I(zwtIx@xX7eLBr5z39As|Ji=FDpNSDjKCju2lp+uWQdAVH-WU zfJ5(Q;|r`$(FIr8*`EDXyO+NdT`E^5^oVq0^RuCUY1WXEmb{&RYtbL7D&9tSJhy1s z{v6u`&P7#`!7N557E+Y*EJ|(sF`mux4cXRTvX|x7Z%dJf~1ZBor)13?=e1^ zGWWpqCL_v3X-xMH8&m~Rny~n~nSX{SOHvH&N(yS;^1LxT5s)AAi+Hcu8zYzK>X&(; zW!g7m`a(InLUqW_$=1l$)j8xe+|HP@%_3Yz&{d>q@5^X>%RM3*;4I2;O12C%S6pk9 zI2&Xt7sl4=k|FnT=D?5@ib^cg<@+znZ&@&qH_u{C;{K%#QP-w36ML2e+nE8qhT7|< zinw>RarLJrJsAOUYL~hST>8$54fpDk0&-z1hA1@!5x^z z8S(6PRvAJb*y*6dbaW@px7`rkdF3O#eyf zql$%>)-kz3F*z+;gij92^4n=I!Dwc=a|(bWoax?}#$OSz+yv^xt|%e;OjOpJYkjfK ztZc;3$x{2Rzi8%a=VhEjjA#?b_I2d2luTa-S4Qf1#D#5tkA6G?taA|G;C|Xx{YL(J z^qP31ZpB)6+IPH*Ptj8+oawex;7nkD9LSlgqw@Urb+P5X8YAoeg3fzm$Sjp~(B+*k2k9_>QFKM8M zPDL+RobkL~SCkn#=-^RBPv!=9AFu*4 zv(I`Jtcn0ucCPaewQT$tsqVZ`Pj&ZoL=E2F znrcYZ>3YSN@fcWR+vZzRTQKg<8ueM;uU^1bTRv}ve~D|i@i5F|Q%?0tEH|~!*?*ID zN2@u|PH|9ear0(a{njp~?MEd>p{Jb6clzN9&Xjz)B?_1(NA#b5n^7N$eC4G?QfB~X47MHPD&Xg zcF>iHU1UOG&~B$~?PjEgN)Xw5+N2o&3+Nbsq2|XkRhVCHP6{>wT}?9Mznq|3onfOaL_R|7rm9cd{(?02X9*S3PZ@ zu?k~b@5|9x9_=iRIXBAet3>ywBlxB`n140V^I*lpnDk}88`<*q@u}{6&R+KWS+mfq z)$Aqp#bg%J<+-w_#?!U(aXe65qBlb@~w6yY-{xSf8vi|<9XJ?VkrI+vb~Y4NYcP=KMyKGfvcwXfqbno> z6}*ktt~56fHRU&@@TaxboMeOFbJ=)9U?KCDSiS0+>r`hH1ppc-J^6AY;*s0uBPeKq z|F7pm^(>IgsVV0f!~epYumVP|lT2C5|0-O)c2DB8BND404~Qk%^H;; zDjYlKAchOWxoXvcbzjo>617I-{)2Mu2`8Z;jF(!XcC0t2iQrq$Hn{Q=;^Q-^Lm#u| z!%pw^z_V{$HJ7=G67SJ4xtP$EszPVa>ST{wf(Z64%TvqioJH9@4=d= zo{0wkN;|`7jXgKjtJwv}hM+RTl5FLdT6uZ^IlTE3cxX?Q0~PVOHa=xPxEw*EZwu(kH@T z6lI#!aq7wmLL;j1v8_c%qIW`1#)qG8-0kEF!;leSIgd$lH5cTGkn9wr63 zER}EHq$T@(`i>P|edKCf)8#~Qv(Q!!5K{|JOw%0C!e@0ESQE1RE5++qPko?c?vk$H|oEnawHJo_b^u_$6RV&z?hK*^ zE@h;^vwQZ0*3!>Zt`wmt5S1&V9fB=lOk9aN(KZ#?R8M!(LlhhxJC1Wp$_nO`=kDW! zR^$NMkmr{&>QG4g&;+^f`QpgNG)e0@&I2v~r(pgv16O~>yc?Q0XS+ede5&G#`Q`@p zCf3DS1L)AVOV{-GWJ&m)WnS=k;2#52HO%NIe>L zY~z6|_R$5M=MGr;(Y*4c9kNvhE|Aidbk+llOW2SnFG(Zkd#NeS&C2-m1 zTEaQCa|I=RmH-;akj2uOcLW!W+9KHDzqMaMYWgLk{;k(si0e$ z1mm5CwvWVF0rK5Z1ThYFCBzBb3>lsmp*WF-y(GK**2TO)2DeHfdW-oe zx4J-GnX;qH%m>+3{}()SlL6|+Yt&uJ^55)Wz%ZyHT%(N5giu(8lr#PGPZ1 z>Pt)Tz)hbX)8G^2U2;La_!0B)0{g%9ddiY3dPW}H z#l9kZPWWNhgK?Dq75_53Rz@Omt*&}>dT$;LmTv&8Fh)G5FFe;XWsK1J9SK?=WxV+E zQya6thrfLtA{JOOH^1pyZN^oI5Nn1! zf@49BL=*dgRXh3S27UCUgE+FGxAJMbvzwTMRh{|A84=biPClUKH54~@3n#ffQ_V*D ztMYj}XG->glv73auKM2eay>!upKZAt6Adey)7e(~gO#}k(+S;?GIKVd))=PFEL9F$ zjY2^u7o>lJ<=X}LS+uWz#RPQGCE}!+IlDPC-PE{#Hh3=Y@qWAHu0yBVD$3jl+3pLB z^iV^}K{}fjNE@}!pTCh?SN}~V#~<+ zX+dU9_$?KWWVqMKRRXf5-3ti7&wBJ$_4S~`ebU!(rX<&QkY;3X!N{?Ztft2b0U9La4&&YIrJ>1MJhEgZ{whyKEt=vNSi?K% zgtpF(@;u|%M16max1f{V*)9>EG<80Guk0S`;C94IhozGRG>LP;xmvknck#tcBwKER z@7#~$-hF#TfoJ%a zV`3X+vt43crnf(yU3>C_o>hBSbbF6H#dKH7aR*KZ*#NT=rQ2;Wq6Z!KicW>!KZh#_ zh}7b8y#}g+`7iw_G)gkKw5-sPS--am?JZf)+_KCwpxu23{G7|EU416fY{>1o65pC! zdf6k?`>_hhcj#4f(7Z3)O!cBkMwR&x~udJIjyi)qKn=ExH$<%+F_2()@48xr3s$=TAiC8Eh!&WTQuNg6yu%pXx#?oXVY4 zYz}KqHrhrnpe4PwLmOZc;L{BM|HmBobYqa5(#G1F_ZOBuUu*j{E8ZUGD<&7)yD`xfESwGfG7Bh!L_Hw<4CQ!rPbMYQ|XI<`M zJE6|QfWf%~ySGR0(bCG^x%Ez&YLjV~4iujs4loev5qnjRFP;J2ET~N))*sgWR+_2i z-L4z1@cG&UtaIPJBcl;NNP1Shy1jHKH}0ck7&C95`UZKI^!!3p6XR#|LBpjDqZ1*4;=Qh469cRF@QIy-SAiV`0t+?II_~uT z->U&W1w=ML_9yR2YX9rLPGW;$71+~c09N{eea|om`t@sMwrS{X5lIpQ9O2birkl&^+0c`2ywMGxgmqTyiou5TUpO(WanL}c*Z;PS;it^k|I%(7+jGN_$b zk2~+N3#4zxs z4a?)>#~0!ceviL1T+rH)FK~rXir>B#IHtiszQ6GM>O$sv0QmWBmcmJy7N{5Q9Cp75 zQvqeU^Otq$#m$UfA1jCZ8E1_kZ>%j(}_EF)Hjq;KfLSBXjV76(@=mkh?+sE<*bdc)a|69 z?r;@OF7xi~UvGX~_sFj0QsRim{qQSjo-NTwSURsB#+pvXO~bao3M>c-awp_p5AU)} zeslMYnZIlP#@#JZfr0PK+Uu&8OLpFOxAC^<*i-HsbT7o+eL#T|7M0qX93?(UmmU-u zcxBIMH^>g`hs2IKnKrXgD!Sps@$U-Q@n0QRf@B>a9H0{+sy6@oqx)D0e5dDi)`Hkc z+V5p`LGgk(NyJMoRbAxw_pu0uK@_?O*;e;S_57W<0PO~(W4yxuW9&@>p?=%9aWraB zn32eC#!^XAvSi=4&}N7%p%9ZKB#bpn)={D?L&*|jH}*AID*G~H>`QjWHip5>GrI5R z`Q7*T|9kO&JM;4VT<3LO$8ns;0h^5XI-kMlk7S2!-(0_5F2vU_Hgi%zuo=pJ`F7Gy zwF>Xjfee4&YT3PXHhT3TJ&QT?&}TIFP=n}8p&?77$0Pu*OtVGD;C?3PFit=j-^x5* zQOwaw^Xu^U+$73reQ*Ji`(5h-3)S43t?6yx-Rbc%9Aj3XdVGmjw3GHA9-+{3QvDP=cWeWsoH12~!^2PIUyt@*&&bbA zn)kXadHPMQU>)dCn>_VzT{cO6H_-vmeZxiP%NfuYPAHm9byUuWsgR~C)xpme&Z^uM z911UZ@^w0KBQfWph|1hMr|oDM6>>6$F-Cn}+eHI88%mgYmvFCb6|)XcOOaK*VJria zk?NHi`67+dD|L#^5i~gufB|q_Hv=|m+aHD|2Rt#8v3yFgEr`G0e;7Y*bp9VjIT^IA zW(%ZjJ1tcI`dh46Fn#DLFyjNr;$u^fV1Ntbr=~nLuFvmQeHmAu+43{hC4}gy(A)By zuRFOoW*DsYE!TsX13HI1$KX9-!aP#V8E3^APDva(N&R^nmSwH;NF+BZl>Rs{o zdzUr_9l;`AqdmS|ABK+2BR^balz7G7$woUTyn^purS|2Z5U)3BZUm1k86SQ=(N);2S=U_5H_V6=apW|cU-ioMUYx@z((^a zEi@8Bg{7 z?zq!0uY4|quNXin@r@Y1a~WL4E)|qKkbF`_k5yex35;i7&*kxVa&rc~F;V1d4r60gGC5C=LD+g1cXx=Y7mJPKzo81V#89XVyGj@P<* zF!4S>h|d0v^#KaO$7EQ!auQ8tP8G)&VCH?(O##almNO3-7b-lMloH-C4-Q1gck*;t zTO5e?)C|`lsK7bRy;t@ckUKfLO7hQ?GW;EldS^x>8=^Ai?ZH&>dL8KNoufA3!itjf z?!acDiq>eb;cS)zlYgm*go^1*OoPKMf0|i-17|0U$j_brLRtk~`%}+vZzf9W$}xZ? zDf}B6{ra3zV?VQv{g3yfnzogVhNo9R;I>8$E6 z3Yiq6hQqC2T|G+!h*Krfjhy5z+gleN6es|c+PaOKFU<@5cb_{MM2(`24OFhUeYbyg zzbUn2KcIvpU^-+F^s@DmOsF%YN;FryL8?PAQqA&4$SpmV4Cqn^xyGKT59^=sIczi_ z?xfRM0t^?{#u{eJ$^-c|Hs83x{s?iPmn*hn&sC+o@CAk{u2v{-ncl8QB%%4VJJ5`G zAbK`-VQMb)W|Qg~pw?##85Ph3^U1#9nY~?ruP4(R7^;<=;$hS2%te{f52N5E1*U+MA;96t#1*@Cy<6Q6#eeB#)^!b>C#a5T#y}0bu((_> z(QVpKXX$QDga$X>qXZG~lNH))1wGC#k^C~hmj^bA2z?ZSvdD6oL2I1XoK$M<+?hxL zbi>#EGtIAB+6s1V)p0rNnG$Nc_HtztUzpcFFon zJqCY@8ocQBY?Q1RFhiW!u5J^X?Gr{ux9OOJ8j2H_%4@<BY>WsHi&nQBmN(E--n);~t5^ zq$~3O?UA`4b|vjPQ1fNW`v{sFPzo@sfd-9FJ{yP<;+WET`cNl~D_K&0WsJcw*vVns zOHjegwo7_OIKgH^sinCBB!RpDgk`01*65je-jxe(UF2bCSzn;(?#Ie|nEFBLkAhy) zPYWI`>W+HH)kw1yZ0VKa@70q09PvLddVO}fKDsQvV&F3uN#_q?`thOj*l9y9!6r}@ zq2Yru^(g1JBOZL3B|n+pnc4*kH%p?@yU9v zYVo3Ad4uHMQ^+-jKcOahX^nC3^Q@c~=4&IS>c;hI8n!o}trZoIT9Wim_fs5xP7B-m zy`6d|l0^{jIKo?@=2dr06}!2m%B!VA*rc;%3c2GTFQXL>gFOafbYt$Zcb-ejcL1sU zq()MG-T338<-)Z#RwUNnaZkoX3G`|1)#%+duZf#WC7*JoITF=J9eWcEv?YSuZOfmH zz*9R_`opnNv9&sWb#|`9lSCWLs8Zsv*zsDa*s+t+e9ykuwY3*xYmG&C+I$$BQ0r9t z={PWkD}c6@aOL-7oXJd|(J@CThEUvyvUfaqnyEne$Dw@Qcs$`Z?)41)of)|Sg%N>6 zNIe`-esm{*X0XV03|(VN1^35?bI{Q`spem~^SDaFBivP}y17E5kyJ%%GyUD_2?C!C z1FO3R`ee8aVRAC^2I{IN$D z6dFOh0DnM^Fm2IGK>J{kl@g|xYk4sM6KwFHOAeC+iM<9rGh|#LdH3WU4tn#?K{flS zYVa9FPK-0}6pXp-6b%$XXtALu7M@Cxs}MT2HFlDD*__X58RHBo{!KMQ&&foC55k%i zDZACis9CdRLWiKbk6jd;puMqi27mV&J-TsQSSt;%Vue~)9Qm5h&1PQ_o z7w!mjt*@M@d>pdO3Wa=RRbWo@x#;!aa+;SWc0B%>L>e!y!b_ZA^GR zNe^L{lYGb^8JTa@xED0k9j(5do0q=py4IeyDD|s1|4ge5hfq^=W`#cUil*FXBz2 z1N*HwdC;w>`|fz zXZ1ZmvkhIM&dm2@|2~)*OaDNH<8G;4TY9QQWK&KeLjeOOxC}&emHGUUzv8vv`Hl}( z@8C;p$_rfkUkgXFdIkj-T^IP)7}JS_1i78-S(y|YOWo+FTvaPcpl{?vPAuqp#4et2 zlm3O5S%CL5eZhp&5eR`t`PP00EvO1#y3v7#$Qq++UjQyjS_E15Jv+kdvA2Gw`s&HP zy8(lRZ82>um{y(A&bHjPg0_YP)7PloPXsXK#QCgPk$ykYja>9B^0n-Tz7g&;$!|qH zkEzQDUVCHDU!ol122^4R@^e6>!9OZ;yBP>|$?aYC zwQ~ap;Lk+YZ=O1Qr9ydtO~L;AeE!cUG$qf@9ZaQp?L1ZZ`#2wg-hR$PFVysA(Dg^{ zY_Ro&ZtBiM>#Tx|SHY{A*&O+TP+RBLH$$c9+#L<*dB_;l z%DGgGxxn+>+1Q9IH0*F2MfH1>!;po{;0sQ|yq)~^%+ie+hsHLXDg84TbLP$yBF9Wb zZ1NXmfgZ?96AF8Di`}E65GY!OO|PwTG0Ea}S8ly8!fE*g3Sh!*e+AL0yjv?Sf*tfH zFqf-to|oVk`wJ~9dogX$taQvkrL(Mgzyez-NCi@wf|glF<2OH@zFitbmiZt*Z@&k4 z?cd|FFNKAn{7wqKi=kcg_28K06AcG-KaRYlWL`8TF12}MyGg?Ndzzo!su0}kGY#zAp*Z6pgrCho`pC> zO{pzvKldXq`MH&yza$h+w{8^KIe=LMN((_i*Z-nkemgG8i$b=yq zNtW@OHe!jM$f{~s5kpqWq-Z+~4JXiB@J9h$|8RDRL9_g`$+cXh-a>#L<8Hj3v8{yH ziRS{l3B?yN31f2<)@jA7##5&)dBlp!KmTzfQ=AA*Wg!g(5a1w0}6aTC{^2XuJC)Iv0*q_qxu7Mw1Klg~6Rqy)EKLHA@ zf{$Jru?{@J8*2c4UFtF_K4PG&MyP>+r4`)ovZruYt>JB}^38&p5}KRh&DyO7s(&To z`-d_Ykp}9!1e>~|$qh*$;x#sW&M-ZDya?+}uvCuj^egj?zudR|m3W5R0&Vg=-c8JJ zCNIfz{d+h24=|O69xOPiw0vRNyF@|!Jc4GJY%%ShY<2N4uJEM=C}Vc}pOH-w4+TG7 z13u&ovQm`r86t>mH-iDr9{i0_hr`k;H_r9dBKE&wQ9Wn=(XvCDj)boQs7H@f&Hg=( z?Fj(bZ5XhTLNGhg6rpn^c3@-@@I7Q22tFND0zOYMtDENmxQFa$If#I`^`awK_34gF z4du=^M(QrVi9XvT4rEoYfgOC)q`&!E*^3R0*Q`r*337xR8cwM_eGG*uVI)-{RgjLz z+me0uKe(zr-+jJD=&{|#s;LVpEmrVukY9!Y<$Y$DF!6)*@Ad^CBlk2Z${Z`SGD$|?JK4P zfXwE09aNj&V}-?_KtfLu8J;oULuR_L#lu5<{(5+wfu7#%M=f`&b~nk2bR-X{a4) zkJaufr$~8I;5eWCMd`5N z^or|ePEdlI=!CPaoJrPOq1a&x8s(!Eme`6nOj%qgH*HP#Jen1xzFP6XXJM*irFz3$ zLxy$8N(tt^=Q14gt$R*AMG2f&cUeiey>mpXt6J^*vkqg^aU5n{gk?j~5TQHeEUGh} zM?Mu_M{^~4Ee_~xz7&a5H$JLx?)CYD{E z=N1R#r|xTuU_w*dd{KQ^us>I9%w*c^*&CA?grxXp@nfUv(O4;0HZebByLWN*CS&+x zMzSc%dW&uJgNzP4b)3Ge4z4UPwjX* zLwBIdPDE^!R6`Y0LJdD~?4bHX1gI4h4DdM{W)o0mAy63E3gia&c!pU5I2l;lI*m47 zrghxF@BS1->`jXSQ+~ZATq+lOCh;syyx$1&oA%I`n($dji?TUdHsc+c6;O^!e9x_M z;5sTl0QZPld?MFw4!gA%g+Ct&$>3Lry*=2b90#)jt%~9O)(EhMlJg9|82*$4BC$3B z{u~P$>)ZTU9uK+NU6I2&3Csze;io9EF_=HKx>tnA&ZO&)1Aux$17c)Qn*X9bSZfc- zMfvFTRT(^cn=+fd>z-Kq(`?VRJ!#2GkcLWd&E`wr2u`D+>-B_yufEfA2JbIE%9YNQ z#NxpncnX5ur=N*KVEd?=u~Yy{3F%K zIfu|U=z45?_QRJNN&l-Q+bskHsWyt1t?g#$;Oxju88g2utXGF3$3FaYHAFd?{rqyUsASf~{&l5Bbx{a!;$XwOM&T=*QR6pV?;?l?zyurGSyR#EW% zi&KrYc3bR(ZJ(7^fvL?9X*s9yK+yaRG``OT_xGHptq-qCtAFC{j3VfJXyTB>)_YhIICYbz?x|$2ti)%Pn>r_imesergm}(UK;rpcO$GH+{UBJFSpxIar4t|D{2TY zil|!n1*3k9TRWoVbY44{D_e82Yd`&I4WbY;@Z8lKeAK1T^K}D$=+(b2(#9^kPZM{$ zy-^LZY}kKz^=9@#d9V$H1?a?f(3ao5hk=^&5Fj%>aOr+B5qHw7NhiX~+Ux^l0gsnY5V5qkm%AuI z-_~vUM)Hm(Kj;X2W-#A@15dRksX!!obB+ka2@gRe7D$14SbVj%3-Nd?MLywrGv{FI zNKaVH3Zt>Y5kW{`i+>54Q88>;663bo<#%Y<)w-N8WFkK=K4tJdS_?MZ-JRL++~}zz z{;29y)|UfwmV)o`XC9!(cc&X1zly3C7BO*jlj^kV$2~{{^4jf|?C-(DMaWLjPyPWs zyaGv^aTWXm)+$|@cOwylC78T`-qlEgXtB*fUnbdVR%cdh4KZ?mcL$3}(Gm!u`Qmsmw4`zm4cL`|N|G=!O^-ugjPM$mJ`ZZZv%F9m4-+=^Le2pV!sG2B2OUVZ_b^q(z!0nlWC(sA)s)4Eig#|^B&QRE zPqG6UjF2y7W(rM%fad^Kb`@xNFO$Z0-xuAyo0ho z1rn8G4D8a8VaY~N_7_Ia<;nDCH|4o7e*=e>Aa47XTA}ixs}&z03+AiH0atZQD; z*2#`NO>JC9%~uow=jW6_?_6~=Shh!Ui1gj=G% z)bG|Y>Mv6`;OI3M-A)KVJxu0XdtG3X<|T*r)Q~cy^R(jQs0mhSJ+TN(UtK3 z@TA?3r*y#llWv+zxX8%QxX{W;r$cG5I^3ooX0Lp(5fG0 z-#+i7rK=PkcJj(J9?{&pt!eSiyrbOdgzOyUo}ahU51#D1B0lxV!v3V}!?sT<>X(PS zSGeP$kHD^LwQl03Jmh|){s&+BcTB@6GH7lP_-CsB#eL_zp@Rs4v|fMGI=5rp+7PI! zZZ6yfS!2J*V3K88Ho=h!#)&Z8X23_Bwa8GNL0WerLVwd(s~UL7+skm%v*X_Wg7jX| zKbSWnTr9mS#}(Ha0P}z-%YE*kXus%JZ|^H$4tjQ-c12&EN5&jJW3jIinuA162d^4I z7M$q2N3XaFWd4|_FdNxKtUNERLES>-F`OM%S>94SWt?;{hVr#245};W;l1k>N8x;jSVf-=e&ijMN z(p$DraV2=KBE+kY0NtxFMcm;{>`*h=XAN?Vf1*jWTstJHG9m2Wsw^n`Tk=BFIgY9} zf%7>Jr_Qp+c4>;pbS!tsM2~ zM_bkC_|)}89Xy5^eBLwqaRtchWC$nAEB34bi!pD92rnD6Z-fx;JK}(Pp@yL?QGwp# zA|u=r2i2Hp-IjwZ{LMxyszgbDEk%~Yf#4E7L@=Mk1;Yd5R`~=Ye>_tYv98Ewxv^M? zSgUtXIIyB}5A_ARvQ%ky@lZ`m!Ag1X>p>Q6Da7A;T51clV=?IV<6g?>5F{1bV@)*Y z7l{@r&i5bWr2nWFtJVTa9@AmsLfoG;i}U)Hd*pXg1q}&remYCB8C_8U_KY^oV0ubi zLdcTTOd2QB)7bQTPh;m%D4#~?2V9@|+|j)&waJ@?tMPHdtM3 zEWLi}8s578tYmh`%fCC>gim#?zB{q;GXXz8ZZ^xIv}H)nB`f7Owe6W|U_bSq-~yy_ zVh)v5@AVHoUho^Pm?ze1+Y)yygs@E2wiOyM_(&u((DP@g`@?06eDYy&Qv~2t9nav^l zOm`hB)V*rs&wep3BIa}03TOA=dusZRvAYQcm;Rv}u` ze1{gkceqo8mHJl}{H$-^+@u=46$y^8Ca<``JgLN@fN`&GHNf|Jeqr`&_#~mNXJ4GW z*W%ZTJTIQtnsmIr^7M&pv%nzRouA!XwPr>wRWW42ErrtNHmM_6aI)+N;OtQ-d|N!! zt#jUax81l?<*-xuQ59HcRXxe?N@G*fSVb3aI4>&RX&-1&pKiK$Bp|I>rYYY?E*TncO%>|v}^w^iER&?%d#GX5^tvzS0>Pqi zg_?iXA$)sYE>#ZIV&inZA9?~dS=?}vg=A7*1WsQm!9Ke?byi=Wx?~c7$Na)nluADH z1)Pkzh0SiIR+>IlrGLFn$?3BN1v=c_=k6qd9YTU5f zJ_kbEuG@gyW%N)y;X@jSmfqa55qy<<6q0hEl*3PGXsc`~nvc$QUgYOd071tc-xtd$n7y_!lve z*_+9xbxE)4dzgfX{tIA-xT-J6z2I?u8pkOTn&m`@%nqZ&Ymz+LcU!+p@&K1iD5aN{M}@| zvD`FJKXZ!k>LN|RAO+QOqTed6ij}k?Qp!VQ*;8{E4M64qpRX#BoV!CT(hcX}=k=q` zUeXWrVq1SBwCOMFZZclm5E7`hftY-HhbTtQr9?TH)Fin|TuYzlUB;4SuAw=*%tYh9 zUq1Ni;<8}v-~70JpY>Bw)mn>g81l97I(rk~L8lwLd{bRwQYv(6sYcR zWp|H>TgLa1O~im#X*>X;CkP|AnsQuw#*00;DLDFgrR^f-j6|mlAe}4`(1LBc)SpZp z^sbOl8d2!mt?a{)-n6R}SvF?wbF_Na+B7NJ6dDbWgLcZmrsYWYCIzS6Peorwo|iT> z98BQ3i0w4Yu<>ofLsWzFsKK?v!|fiqNB2ZWlLNRtu-iqX1z(sa$E+U7A-x4<*b8-a z|G5ciI<+~zPk`Oop7pyocM0x#1C#cpQDx$s%Hb!~u9Ib&4b@rs^X&Uhvv!RsP7SU< z|Mr6ayf((zX!GMF+q%cl-@d?ll0N89VlG#xRPp-W1A>oLV4a$eNVsj)l>8FT7bG0K zS{s^*^*XT={Lz}ggJlxR^KK1V3q;>4(RizK7NQqv>m%{mne@U{`#{^g5BLo>RnAw3N%G8~N zGG`0X`4+&@P6)!Dj~U)_aMltq&Hu@*BGbEgB}672q*HE(I?>_)6%#s+GN@ zT0SM!*#Mbb8j{)*@Us$Ed_7DY#v+GSZXBWry|A?F!TH-oRK?hx3{qeCkjoGKTA{}l zYr`LA%Vi5=Y*9tEXraAlRw5zAT%A|QI)I9sPq|c*o7!R-mUW?M(Z}o#vkIKMR+B+i zk+w3vq0)~+_J5Kue?H@t;8Z%)R^!pK@s0G;drrc(;hBD+$&%oz)MS?wc#l-?cU$p; z*3JRq=#iPPN$0|V-D z4ezZBN^UPzSLt+gniXiYr>F@YxaE9zOUs-7NBN)Mpn*h-a-4NfLRUCR7{vK2SmZ zlKIhPTjScorn8`|*VDY88;1sikJ@1nS%`9cr>K)tJ0`H=%;nRdpMdIUIt(~|7~zt_ z$IRN1=6UL5{wR3&!Xp2J$U`+hj@Hy&Emttfi^HSKit&Dh%IlX`(g#{2b zwhG?Vf8d@{8EwP+AeG_-RVOg(+5$zbr7J`0-CQ3-MS8_irntBS{v}0yS#1Kd;x9*e zSmA)Ke}jt`e?w4@ZIs{1v2x3XsRz^rMaaXn158wn*z;~S=CDcw>|0BfKeQ(+Udwko z26TP>kE{VPgK=VC;YjPQPN%Yr;!mCUuWJ+PEd5-$*nsXlZY(?Z8s!mJSZI%vpW3(M z^r1Z%E(^gc{cGZ#M+@O5>F23=ah!c?g>LL3*P7R((}k#B@(Y!4)wioTDeq)xQF{-? zQSSwJEL0}<_c#}Z#rJKg;xuv2oG;DXvR3Y*U!`t&viHUx0j9IgD;8EEHs*vl!Tlpb zK)@#DfZ!IXb*w|2-?pKdb9KpXBM<+)pVd|a8#JWI>aHM_654l|W|iv2yi!yC-4qD; zjFJ+Ckm?xz<_5}>cF3nm=}K8K{EvBDfFeI)<<^;h(zN-<*u&8CPbPmkp66$QLVm-h zaK?`M-OK9i6;I8k_}n==3Ff_H2jxSbBeWnwk#x?Sy7V6ELi8`xUk2&nsSHaKwb8C) zr#ja4^GJxz?r~Qy_S`P#Dm-eqwc&eOV2N9@hnU&(9-+9JMTVF zwUTDe?`w|=tPW{qwIhYMox#^-1JJd}qkK|t?`l!KkxsMW&fKEx}r-Xfqrd%^#0e|f(HO7M!H1W%7bHpZ=nI;d}T}N z`B$QR6m(=Q_)^0aD62vT5S$^i29u&+@^-NOyRF@~xfb z7KItY$RzllxjX(?*=@xF7z0);R}PL7+6Sy#80GP^4EKR=^aF>VXOQ@_Mk~fK{JI8U z_!CnQdgAR(VW{C%=OH+$Rg7c>(9Hirescq@3jECys#5?dkAHL3>enZ~FKHq`b^zP~ zB-!M~c;VV_AA_n)q@#+Pp=0Pl0~5tBg7QZ`<%^reUfkiy0QcNCXXXu&F)Q|G>yjF0 zO#qFNR@e44o83y;UUXNC%<^Q!ZRnPk3S+;5-nhP_?Fe!ayH1f$k9bc`n(=O5Buc>F z?1|!8nx!6vHA$fKq1MnH;w0D@0wbAD-n;2mHJ}GUMCv`9`OcBAzJJlaNTsw}5NEAz zIXvk!YoQNsh3Do7#OYO4Eg-HP@p3y#h~1;t|7+jblE{FL6gT2_nSr% z2+{|B(%dlWR9RM7_$ws-FNle7EYeWCJn!`PgZQGO zHl@A6JE8*!gT9XEK_1E2UE2%@;tsOUTHqiDV?Qihq_B-vqG}(Fn)Xyb#@nih=|gTM6HeaHeydcH`7#W9X^T*pW-(BC$0C zE`q(rqTRg$dA+jruBMjnxqD|%IRy>sdEtrBH|3-TCWNP?G%zUpOj&&j{skgLEZlYpT=iI3oV_m`i(2AbWnR(cQZhR>_>AR4>6@~OAITI623J}O z*#9LKJL3PjXeZ)SJ8hfEU#Mwvjw2>Uw0P@r7tO8ZD0_~k5USNRNG_Y7UD~{N;FPDZ z@Q=x0gP%4Q2893M`&Zi^1PXZr8~{jX5o3ohoKYRFkgFj(F`=b*?M0>8v-Je$+P-1! z2~4lJsz!za>sBjj0XM;giT;>V2-+^v6&j;82@m~VbCdPrN<%~jr6ei%yK_&URyn&q ztEf+Ja2w60zDL%Mdx@P@rvEYOG1dzXcXhcPEOa;E&4tAy3c{Nv->3_aL^yg_K%Ua$oN`GSv1wk!I<0Mj0kO$_e<>bsHq}$27sr3~w z|B&enOY&fQ>WcHx@Ct3EE{F^mtDt{Dr8&~zFe!2rJrVVjeQkUa@0|115`U2qUTYpE ziWB}_xJkc=`IIhS2pY$znBm)Xuxf`6N7zvY_;L}Nawb~>lLcr%MbNFsOm~-bl_($}`Zq+|BAX}*DX@ecxQ?AG-Qf)7T^=OE|DofB|u8$8Q zVmBqT-G2oB*&Vr?5+pS{AaW%!w<4vWqUdIi&p)q!4;@<9_jqY25hyAl)7L9^&Y0%eI32Cp{j6>fzNuS9?JK+0c~0t z{58VPBBLJefu2{YfJ~nNMeC`_XnFO*W^l5}<#2DR%ZisLdvR?{0>4FHhcOf~QleIA zssXrICYE8&ztbYu(>`n+YbmWy@O>lMlDmd-Zr<|>DJI{&U6zP){W(0qCa6*;BNPXS zde|R_ZqaQr;8Qz2@}TAVH)q*A$%uxU4;Ao7=X7?~9kGP2&28a~Bi!Xj-|;43q!!Dr z1a92GszDp20fC-ii6{Jewqog;Xrv_s0o7xPM7XEnstkK4T*>K%z`<^s=H<5Udj3a) zxMY*QqAl@q_fqNmlf7oVC(3`7k#WKkq|70G-r^g{jvk_BdIjd~mQ04^U=@1O=#^Ug zfKcbG288f7QsA)KRa9C*@> z0wPEfb3<{B=bkYK{Wjg~t5Wji$e4LLP;O3S)7X%(ux<3*30||KCyVZmex{9*&P>J# zRU+?ORk^r~m=c6syuA^(QkK`pn*Z1;{d?MLX=<#-sE$L`zhDtPRA}n%)PpI;JnrTT zdo%S8qL2BYPw>Nu>*h_5&ls?@6{xR_Q|lLQOe%zdHwWHo&|d=Zsc-1c_Xg*}Iv@Oa z`eQJUv-M;@cV?Wcde*4dJE3|UYaKgJu@KDM;%merw6Y?Y9a$>Iott#CucJNq?dnS_ z(Tnjo7(O_BwX$itp;MQe(=TUq*K#rh-U$MWSY-Z;t)mpoSdUJmSywuS1p(+SWjILr z@SXZ`t*8n(OI6swx+BUo$>)d+T)d8awS%1$>7_YoC!nm}U)5$o=!#Uvz_lh+T$6g< z>3x3&`bhMy$-3(GWrFOPA2*Uu2DKmhe^wQ+>tiL~oe==#wU+os$6pOgEC>tDo$PN5 zawL{$))6%GIeazmitk^CPbqg^X*2sI)wt&V_=mmssv|Gaw4MCsCXz<8`kGLMK}Ruh zCIi;@hBuwkyf^=}RfeJ;5DoaC9x)foRm7%aB}?zk7Hdh`mmuV6C-kbg##Rc~lHB|^ zaGTQ__4w{akOxrdVssv6r{1VF%I(YAbvMO{ABAQ357!OiI^8~!#Qc)0-(cGV`0sO= zJpON$jxm6S=a^&(FUZnE2ECQstgEUzLxXj~v z)fkw&fwvWPOAly52CrPLo`K7mbexCmZ0HMGxfW<$cH7`Jm?4KqYaXWqBjiNTbUM-o zIY>K!VZariqUXh>To1-}hbn`jGBPk~!3iDm78&fDIuE&|FPZC=SrWKzU3UL-Wfi0# zlBv_125d>XnrVWc8U6MfLx!2vS}l6ec18*%$3@ecChFP!2dV)(kqArbMEzt19CY`Q z76GyQ!1E)vD2@)rKnOKTdigjpWmgo>seiBZq|Ns@u>)ENw(oMA$`I5hrzl@g{W3&` zqx0gLQs6^}c2}u`S`~u>YMAfuBumMzO03DgyuGE}51G>U8*uVR-!Hs`<&cA-eP^qn zw9%keztG`xpg(>f)ek1KyoTMu2Py{sww0)1oBBz23!BJ|93t;rzV9RZ^^#B!)pz{B z+FmY3{}8c#851sKD1v?*63eUWGeg5v znbk-W=2!_exrxo6IuxEJG*A|v+0vpV_$`5|Kid};M6XtSnw0-vfVyW~w8FY^^(c(~ zJ+!{sY8hW`6^>|3qs?J^z+t;s`Z)gsEf!YjwFu51zyK}Pi_b4dT;71}@ijNHeReL& zc(^^skOQ*pke_1<{&-G{Sl|&H5eR+Qkv9!7=gNy+ErWZHx>=});X+Q`1()>hj=Ro( zt2)b_1-e-H@d(1{atg=74v_;|#`LN!>xTAv8$AcYKHP@po_E)_JX%c`xs{u==HXjd z>Kjp~@w4c$a!%l)ILhzECM|vmel+6v=+=&Eh5s@O$_j=alv|zLnMUr*}gS$ z7RyYwpsH2j;AS<$b1woyQJtv1ocRU9XQQn+*C-R$a6Sv(M5X31T>YGM3x%zOJ3pR? zJY2tMmkt+x1hreaJ|U-4V4sc{hsNxrs2D;i``$Cx3#UEthXtXsDw+YeqAwm7&?9oS zTPb(W>QfP!Hd6U)b}JHGXLa>OW{Zpm&aoa9?zJJV_qJJjFgJb`QP#JeTUolHj@ui3 z<@>fVpS6BG8}Xkg+i3<`uVz2Y87cHIqy7H;GY}}C8{Z$%d_1nCnEwe8(Ws%#LBSUz zGfL?8-5K^3K2-zUDI6a+fMe3LG|3HM|M_ulQW(;2B%sPGT@(crg z@`^2Jo5Kb1^)osW4!loH7Jws(n@q2UvS` zVMuUnTP)BWkd1IU^;XNqAxg_CzBE+%Nl4onDrv%>qr9USwzh(bI9zY1);ktF^w%e8 zu2jaPkX;aq#pu=Ir>7P=Xp4E>OyyP}gW=Ce<)eUH$}JmpjMyQu=0Lku^Mtr$8ehMCs_ky1<%&-{S#`hD1S10Mv{G>t&_&^u=87-xBV4}glqyj6$Gs=V z=V)8+qir$AW~^MT%m}ae=$@KMFU9u!vVk^>n(;=b#)}5lS%$D16Q4965=q|`W^R&{ z{=t)s9d}W;V7#B>-{+36hHcW6?lN^=uCaQrh?cSrPjL%s(CV!cV$smC6;05L=YJ&8 zyI!GUNJv1EUgrDO7EJlk`>NlcdpbjT6w(^od(PGSGX2bZgZ#5dFu*5<)-Y7rn-%F5 z zIoHv_FBEMnraq;$zIt-XyXRu(+;NC;r2_{>ERaguw1B7VblfL5nBAHbU2lJl9%=-> zB+0WohKaX!{xZB+IYL=M=h?## zLm}YzPX4itLJ8o#mknvT-bX*Voqy*SO)MPotyF-9VQyj-A(%jK%Nx-AiZ@vo$hVKq zr%0U-la1A`>yI!mLWjB%G}(ZR z7#d6{+>fV07EUH(b1cQaBD0*us5)NQ!R*@K_?JE5;iWjg#jT3(x^Xys= z&=h+em2-vex_*#JCOcKgd!2QX-YJPIjHB;Sf(&`G;{9ORSLQD0qQqv{vvzv?&sDE%TY^`j(_#F3 z4-_MkJ+o+Cv{Nq`sn={Ky8LiFCcNVNPX2^2fO`NW3xFltw;QeYmaf*+jzO}#$Kt%$ z)#zTQAyK&Y8_t*LY)#n*rn(0OYS-*;l$OrkV-aC)`7T-V^c!7$;tOCTKm{K8`fdJV zydOGOx2-b2`501%S3r1ed7l^0$d|meqKjMGi{2IiUntdGh+fP9xiAcdD(kfN6;4}p zw<8Fp=Gr+FbaweA;)+{@@tkl%c0-OJvxUW@$g%MDr-Bmhxl)*$aUU+w+2C$k_e9?E zKbFfctXgc|{PPNAtGlv(I(X90gHqidqzBQ^!wTY#tj^^l(G8k<{mO=1?)lv@9Oayc z=uq#65Y{d!9w?;8ZB_unzO33YaQMDv`CY5@!+pNj{;}1jA#1D}lNBb_XZTo``0y=n zJyjCs1^2JDJfUTeHlg(GguzMtJ)6SK`ycwO?ZdM=*qvQ2=aZdl+@2 zF!kDp1%#VlGJ3!BSX&DBb1KXS{()j;PocFY8ufluqk?Nx3SD`)t!pw+^oGJ;d8H!qptrxr_ zRRv+!i_Z%d55A(M_QR!Kyce8VCg8g^7y+Nt%DSH5^rQFfE{&nwZ(+U}ns54t;Xa@+ z2Zj&hc7~c`y?)6g^cYyr91+hC$Bt4y#q6D?#hoZOtF&fPM=TvMJuPQJN$*1$W)tqM zSKJb z#^wp41w@5=^P9>5OvEX}Eu&a$$!ekhPx#bxy@g(ZjqWERCAc*0{si~;lRu>(?)c51 zwA2^VvuJChVrtIxr)VE;e^QK_>z2bCcc=-(-?LTk*akY9Cz^*^@t>epdL)Rg-U7UY%@=MN9xsld{GD7kV<$Q zLD{KnJc5swqT+ORp(>0Rs5(?*f1>tsTS9_~(`+PXM{5=2V)G?<=Y_UFa>;9h+pnfm ztmK?#f7RgkWY_7#L>ZfrQg^v5#^XulZQT1IC0p{T zQTp-f?tAnCb%NK%QDyo6F!o+yO}*Rpuwc+c5t1MxJtP5C1O(|N^xkX~iGTtMND~oI zs$i6aUKQygib{vjO8}9g^xlJj^xg?2-@<+N{-593H{aEIo^=K5%{%9qbB-})r;wOz zXLyUko&_e1^s(?0)$0wQ?=O`FwYY?KH`q_AkiOb(oOb1&*ijnT?4+{>AZ6Am!%jiZ7 z)1`DZJ|P)rPYMaqx;g|SgpJr0Zy?d&TaJK0l;x?4D8$R*6GP|-OCn#>;{H|2O&_hg z;FoC8Xm`f>@WoA5Ss6r*t$V6-FJrPBqk|4r7p(&bZTS;w>?I@96_36$S+>_L+%YFF zX)KM@XrMOwR9O#GebfDBt+Gd|?72TO#*Ba8rXI{FI72W*3<47Pg_P=zCgIK#gckcu zZXN=jj!~hu0w;R|sd<^|Qy4+mhjiBGFM9l~S>tn)eV=jm2ve_ObIk6K6;AbY>JtgY zo*S)_!hgGV2?qCmXqfwMefMGE3+2|l>yvo#dMGO;vj4Y@&4qBY4-YKIV61#1h|ZL; z=aZp&zT1TC#b84h&3tmkYD2pM+UmTCNzE;{e@3 z6~HQ+AgVCOEi)X^k=HHvKLv0>N6 z09#zg1m7>0VaoK!YH1ZR7VAOd?e8ZY7P?JfQVb(vo~XR(%$!XCT4y80mfM%WQu9{e zp1bdNuBwJlZjJh;AHw{itQYBPmyx-9@0#rVmo=YJM2Ck?!z+fDa;$SOb&q@+4)cu# z{>)JdtCy&hb3VOjx*1}ydYZ!9OX#~zRQ&BBX@HQ(8ra9&)pR(gQDVcX`F;pTwP}PT zT+-6VYT^sm6Z;s9*CI=dyDQ{xLoQMB%ifBJ-fd*C`ca-ObJdHU9&27QtFcO=%x5MX z%xWl<53|Q-a)M~R7iAbbX9noKt^?=kkhtq4#}YLq4~cY0xm(JnuJuA1AAGMp?NDBWor_Z&H#Fi z_)9fUn>uBeXI9H(qK4zI=H`**q0@1n`W6j#oYFK{W$f2;%^51o0H) zbexY1!zp{3K8#1m$_CaHF-HgNf?&LH!cf$>UkKfcr6a=lZ4gIj4=^X_7v@s%t%MZ> z+-$xKKd|4=POrM3WXAA!9DurxY>ea#Xe9_*XKV9^p$=@*AC`XVw;%f>(p8`+gX}CW zEk^NE@PZkR??ax=jg<#x_Am)iCiYiIBs=ol-gr}UvD1YCRoyX3F5|L{y_PMQVwAc$ zs_ED%+L=yiKA%w)9y#t_*FK{0mCSCyhDC4Ba;sWgmJSJJKkb>NrFEu(F2I4Loqwy+ zwJXZOv$_IfT0@qGeK3+@vSLd_w?KVtDV>H;(vZ!7;}Q}w4-6W1jnJ-}aWJhXye^%_ zjysvX6RVilc>#?wHjm{tznnXy;hSd6)QflR4Ie0~{@g*iIc!w80G!kbKYOm9#a#|9 z)G;hKGx~uwz(@T877I`1C5V$gxAGtKO)ws|IWVZQ8^@nyT>(VE|7>);s_8yvAkUSy zWd5u2;-n4WH!??Rg4cgMexOP^@l>FVZX|L8ucKrGvhw#DI>V!6KT+zKPWK3UDUyh9 zYhq773$so4In_)KIx-})+;xjRrFP40CEzRFxSbTFhe_38N1wcPW;k=?m!ogPY0qNxJ3c1$~w;ufG+=p;;_wS`}Mk_A?4-5o@TsPJ52oYCN#O{xX_pL z7N!aen*|;@eCy329{JbX>9`zB>>4J$+^}X?dKr+P?Se49cgEjF7Z%Hw`kCOeHrYAQxB((xOkA}wDSwoof9u!iD z(yErd)oZUbL++im%rrCG{4}jYIVH#I1DmXi*-)u5sm*~)q>*5lTgrNt=Hd^bQl~&I zegjWk!Bf?1xM`ixNpIe&AQ5tD*Gl{T!Qm5nDFe=b7RYU9fj&ojX4Q-n@O~TzgW^~s z46?L^o>-Gm&z<4i#DPRq+*AUInD*Mm3C zxYAqq%V=~tYad0Sul+fIcB~8-7C&^lr3|Lp^M6eNP!0>Q*>ccf<^UQmi%|!)_=K#r z^0crI(WphmNaNZ&02jVIvT1|wQG^6b+<#MdsmBIv`MAFjLLcmphYkfT7)3!6iSZVS zthHV9H~goJOuJ~;X|DpVY&%H?_Zt91fsX8Zt`4!6>(p+1-hea^1ERO7PO1A5<1BZU z#^{WO-z2MFER&qTMD*h=DAg%EBxm~Jx@tp7qsSMFCqqX(sHy=?Y%!0lC8BZW_li6U~iqu2yBD zo|YC7eQ6TWSCDpZ$m zxc_sH=cSko;56%n^zdz$-RCG}`O zA>14AZWD?EB#-58)1D=-U!tG6(jk@d$XGFjpNtgFTH$y}FFCf+m2<8qFp?Gt)!CHR z7yl`W|EEN@I=)R-wx7OE{>2cIii&x6SKotscI008Q;J_*kUGziFPhdn3ud%4pEr3cs(B0 z#F$P5S~E`U?+YypVBHzPy{gn3#uL8BE1G>PUSk4&*iVTncY4Swgx1*UnBh!{^|5O~ z11wppVS3UEp!HuywS@`<1Xy9%U1OI=59XI``H?q~Lt8WwQxtYGyT;g$fSLbF?QURp z^BekRJp&z?vEr0;?*tqv>HJ@Bta6HxL?0@ES04<0)VN0U3Kr=+@L$m;W^NT z?9s^cccT^(ej>RY{eY9N8ldA62T`T(c9O;fu$H$6SpvAr!lz~|u(bw^fKA$Nt3adw zOKa=~x89JC+k!G>!`wLkb5`3EH1)4y>GR?Dj9e4|+cj>W*M;=`DcJ1^<^X&^nnM<@|wBnI8 zn`~d!NP>=xzxS$|GTp>wX!Rpv5#<)1bD#9&tjzCo<Qye> ze75PQ4YQA4M1lQt^CCmd`p4D+W){0MPN#Bir$)6J0H?BBFU|w{Z47^FllV?i1zR?D zy>0NO*7ePY=!xo^4>v;T03$`M8*@>w(26|M%|Z}|#Y-PF70eoO4M!d^}<1*-c>}S9*fLS%i`Xj!9io#evf}JAr(j7vAlwqN&Y4 zJr^j-FGJgu5%f<`$BzdB9RI)IrqiVzpx8Xpe_Slz& zKHJW|8wq-Hr~*Nvo%Iu%Ju5T4*8>{*EY?;i$V9n&zmU6do-0jgUS<_?m-j{9^-QM1 z>jFbX*9)SGDbZCh>PkiF=`G}9mv$-W_wwDDdilFW&Nehd@@B79dmCjG1QBGE$mZ)d zg9YQINB$QRo>qfev}ElmTc7l2+B*V5E*fWdHmqj2g+#Q(Q;6XAL>1@r7PhgnoH-f*~LfaVo1}c8E6i z$e`l0W2ae)dAM-qFGsPsRHhMV3Fra#WpCjyc1aKQ1(loKA_Og3k-@b4Eaz0KX``lX z9p8X7Tq>V-P0klADC}p>k>_GIk_%?sVec@Ng4ueX8%-Nj+{Q0Pn@22 zZT*{GmM_v+3(#XdMnIgQ#lR{xkUTQk#mmxjn-6Q@O7DywbkSKMy$=i` zz}a1?C~sSwP}jHKZrXQEexCB%LxzO@1ZBYiPzlJ$C><%sk(sIdeJ(@caE=eW0M68$ zPw0D)LVhI=Kb@~Rq>YVwV|{K3ro`Z^iUk>1gOIdk--S%23y0oYhif{c-w0L?S6g@? zg@(V4EuW)xBf8cxgl*~txm(#2JT;~Av`x+WB%Y1}eiy`-izF>1WLW)?`cP?m&N=SP zsO;yTezQK>FP==eFG&V0$ab|PP|fNVF}PUGtWM#1ltT>oNnx+b<-M&f5$>q-_XEAe9}Mud)Y-34QV@PIuPw=QTkc9Qr1v*KFjy{~rh9|73JO1kffm zsr#G9ZC=)}Nhyt7XwpF66=%Dwg>ueLMF;FPwd4DLHKjM*M;kQSsoH=@szDdct(eEaJ=)9q1kWm0_E+Z74;wb_aFanYz7nVKG@Jq*6~CBqerTrLK%I z+%Oe3?ocbA&**qxoIX_dycCnFN)68l98}~Lf*0lqNV0QX68x>QUmTyZ&8aHcnKKA9 z7tWFr;Nv;uqB*zzGdqNiekZXBv0*sJGrb*^oFU)kPL!}G+J8aE@`b-W*>IDBcuK0myn?L(Sk^WWnVPo>75_)gAx zZ7~NEf&bDz_@J}>RUwC*FgdLD(jufsvX6DRQhg&DFt;v$4tP_Q6#t}l?;Q07*O}X?lW;0v^< zd4*ZgGiRdfHu$t@6|Wllw+8A5P#Ux=r;Z{bWHem_#)f^=EN|C3#E*IkQg^ak$9^nO zynK=wn^RWb2W)tM+K;?Lyx^l+M;k>pxh=|Kk<0Do?0xJf{F0Q_YHLN4(5-*K%bm2^ zo=ZBSNHo}GTy8(!B+wp6F7xpM2$%M2Wn^ZW4M>Hv*7@{3gAMRRNeG$Y@J&SCFh`xU z=EldVVqeA_w}i?60k!{rj8`A`UaGP8bN)r@lSe>*3q22)Q!a;2W%MH(V7c=9?~CKJ z9AS|{#)7b4IRZCc(HF_fRRtNcF;AUT=3oEm7MbfgeyWz6013V*8~bE?>v@p&i0s1FGACl&!r$@UE`q z(|8o}fpS6Mc%r?>xl%XBT&=~X>Dgbj&)U8>;5W*-l5v9bb?*p6_ z0Sb=Tql5QTx&Q0aV@ck?FscNt?6U!^Q(#e`?zKw_HroRou%F<#?%{KnL^5oBA2{6< zz2>NIosp0C-PoU!Az=+z+sjJ+eFXnv&vzeB;JP2TmnHmtpMyKmRg}F+x09{OB9`87 zxP||0B-7p04|Wp}$Tc-e>vxO1lTClGs2uSjQc(=qv~Ox852#SDci4zIq?kaPQ-OX`6VC>ASs=IRxV7r+R)L zHRX^gf$BG{gLJgNwRd*U(>NZ@V9t`h_`*}sXrcN-a~x0^E7E%t*}a0g{o5L_`>cIK=dxD?bNtlf)n8 z>{XU~A|qi31WJte2^O5@6y9^h*zWab%Gbgr!J-vlm|f=hKHI$LWZLkCz~9nOGN8$p zjc4g7W^=n;{ODPdSl2E2p40i(;-qv2Q`v8)^B;>Pd=$p|qi)tpk!!3RDp>!fh8ke_eQ4NTq&B(q=b8Qdkmd>-Imb}p1xcXcw#?5UZjn! zy&AARZxz;w@SI8Vkk~tx4rK>9@W=l0UR~;4Cbo-#sCFeQj^t73K}qOr;>rMYw&a!P zJKb5257vL$G5Zc)JseAtIzUSJ=o{R*v*)!qu$oNyCp_XfR+n-cVDDV>tW7p&l8W}>3s)7ujsRhO-eG8M3~{T`aT+!w3AkfJWS~9c;ETT9xLiemiUrXWCto~l; zs~BC>qfywsCBJbhU9i()QO;Uljj33_-3cP_c*Zr0`VdRWno58>4+byoF;zbWLYF@H zVukj!_{ux<)y}tl7K_(3HqUK@!E5%qeFZ3jog@CafQ9QnYYYy7<72b`rC0U$*h+ap zAlPt2(gitHB!q&p2k9W`BCg_fKwp;_&Zc#gzz{A;V4fkPkVp+$nQYaa_;R=wO}b|n ztt6PB8liC`qeh>)KcW}30J*}8x3(5;0VSDU4jWf<;Q0tn+%~&sQKu9@ylKF;XMuez zjH|m6Ngr^(DKQG}gnD7MQ_XuZ%S{zrjD6*)9*4LBn*#Ym+J_jE-F-BPud-X{_8byc z%w-v)@@C%38aURS%F#BFHttWOJD+`#@}2uAkC&-b{g&d*ConDPzvhRLvV%FkHw(^G z^u60+sbidR-u;s>uw}UL$94V=sq1uKxs?9yN>re2^(Qo^&vQY|KNO6`k9K~FXEWv* zL?MjbBUwE)XC`HG1$yn#^!LTL2p1e8zDUqdQ;zb6gm#x(GsSH=hHR00XG_D;zvCEd zde=-tJ~ZSubZ*JJUajb_F;8(?{A4WK&v3hqk;Y1K1It!R1-P3>(mdmm0WK|awJWM5Lml(CP_Dh)mDLZ^97J;7^z5|WK~|Z{Mxf6@n{U@ON)>=GN>W&y`S~|C0vtGj_qbBnrbZXS)bt_e3qh%_zB%gA3iZg(=%+<2 zx@z%rFK=CSZ>C*tNkGbldV+3{ZAN4Ro|`5~6_EiLb>xFyCY+ZtWFedm$UQdJ=7Xc_ z@am6wTIk+o>5@JCA)hA)aPV! zbL7oCksP}NVb4138T8OOrBsRpSs=L6ZwQAnC9g88OFdwwV?Wi`loENd} z^C$PLtJ?=Uc58O-7#vm{sc+QoC7UU0UmM$j9=6>KU=8Ky4ad9AlA>CG5KLKh3{fyH zqgqjQao`llLW3YCx|PU|vyP0q9P%I(vP}JUv_6nuu6+p81*4{nZi!X%N8FTXCH@x zxdM1vtGa##TrDZ$Y#S))aXe^ zb;;u$C3oYi$gwk5JMrf2jlReH8_+Uc*tiN-Fd&9+9x5@HAZA+69r}c6{BhB{bb1g; z@FnbY6xV&V?{z<8L32q6)Np4;Pn-?JF;uR8{IMk4z11ByFfEQk!=2N#5uD zJ{pvCv0V|R&YQ-GVQ@HWOueAl#3L_b>2h^6>gFvocmZOrQ2g zw_souDi)7<9HKeEm+&wn>gIg#;%$=MVnm3s=*7?hNqyL46EJH_wXU09*^%;p8uW_&tJN+-x$A79- zrSWUW;QZR~!^h%(<82iW^gY_7s^WF7qBf16|E!T3$~=-LI|_QKY-wU>{w;C7kolv$ z>;RQkCTIoxEz@9l3i+xpHEdEfwb4gadC$B6ZrvtuV)#lFt#*L6Q8?~osEN?!I+xmy zGdv&q42eeP%6F`JQzNA#&k%8^GVe#jIc;}GjA761>4ZaPopqk>d_1RJd81#`w_ayl%W-QWbP75Sh5?q7bRF{z_l6_mG_yIs(IgI%IKO=lytfp)&0+IyR1@=IpgYmmmtQDt=N|;)thdOfwBO=vg<6hMyY$w52A`XvZs&7@cxv+->)(Ec)+zCpv#4+}UfU^61n!uvjst4DFBW(_Tk~ z_q9~Z5Y3M70;J|OA~4yHKQW;lXgokA6i)+0GC~^Z*SC~H=aVHGdTrg$I`)3MQ0;HI z`SW%G*9c-tzPv^W#aG~++&A@%!xt=HSB7nNsheS;1M*nCfvV!xg8ioknx`!vK3GyC z$DF8i8co@Lak0-=Om=~FQtXy?YM9ZAkt9;AA9w0`f&5b6V=~K88HH?v%Bu8>rtNq0 zdbX5jo@*VD{F>o*fxWCG;6FX%4eL-?_oBERCS7PzZ-b}YQKBe3^ z805M8c#_ZIch)HS4Rlh?QCX)xZU-1*km9IPvxV*O%Tn(#kiFuE)2h-t?Q^ z43sJlNq?=TTDY*q@!WKLHpydbQ1!-z+Srj8u}-0;;-)78o{r=oBCO`Ag{rJ(O&&5j zOY36Oo1By}j|4OKqoGb4B21Scu@5lHNz6DI(Qw2P?+IHA)0a6*6)?XcSibVRXX|o1 zSMKjFSFAX#1A8_cahCms!E^~qO&zO%s`J2ads$5@{-f0Xt`{%EYfc>9DTo#r{eXeN zqWJrRn(6eDaPX^#ry0zCW{_(vEIbFymX>9`;zqOZBa3s%bIYDru4l@md>gxsLP*OS zcVOwRwn1Im?Gs4U!xe5~v;2>|;USdebN#q^@_oEC<&7n#b;c75A*omgxrveXuaE_t zw5s3X`d4~p5E&IL@fIw@^>^1&_aiIG>oZ5Nw-B#2VJ>`0M9)m;b@KwwKbb>|WmYW= zUMB;pe-1Q#^ET!F)P79AVa&y~v4rw__Xjyxw$!u}VcCM7p+t$n^#iEwqF@7D^`RBg zr7J-qYX73M0_nm4TZ#Z}C~AZ>lRl9jH-dI(4@^#}pQ#@hO)!$vxTxW97j(g2?LOcf z^goX_I)C2X_D`R-F)YXLpPvAs6C>4|ZnN>c9OsD|ilmL|Fc(PRAcbvLYXunjmOJlU zme$T9s&Hw9Ws>i0+408IhZ!fsQ@XQz)Qxz4eSUi>==sxD^5L?FMUbI))%L*S>auTJ z<>)&CBNfihyq5+u^_v2WO25k>dAiQ6srP*HM2vCS&r#VjbKUKK%yCn98^2R`-(WLe z{$^lAD)mzji)jmb>zl4Dxg|`Pj|38!jBMMO5BPLdZIKio{hZ-qARHaGZm88FOL^@bsi!{l{rx!!z5shG$mE$x=q|hQUXtrnKVRc>ra-?POJ?d@fao>w z>&*a@VBb*SF3asI`QS=`w|83B%NOhz%Hp4^{*oj%m>}^!=%Nn3B$TsjLQEtXM$$|E z;5D{8FR(26&ojH=zprs0a1v73(x< zA^xJqz`7E{Y%kTvJfJQ-O^R9jA)h9w`RnCu?am8P-N{vUk|%&X7yKxpLxk^ckc%2drN^ELgfK$|CVu z($u$w)^gcg89uy$uRipH-ZkjydP1APTjXs$(bUniB>wh2=Ds1Fkxf1vtHO`#jq$1J z^S9WnwQL-gP3bMr4f3|Afi*P zfRt#h59`cnRQ$dJv8UIwCqdVi;-@rQaUsEjo0lGIH^9XP&`

k(e!^2x}^Bst?2r(4cGcYG9NA4a2*m%F%@DYJaFtIUURzo!h;BRqGvcQ0uW z!I$Ou!WX=XjV@j{!~C)sJgeY1alw)N7KmuO_8`wUwxY55dhMPKk_}r`ydSk98$^>+ zLQ*^{S}6k-Rux|Vyk1<4v^uK_$;AV*e}mT0rvq!=w~r0R5VGaW@KG`UV3;EUTyh}U*;Tk~dHc5u+r z$RPubAgee*a7MYF7&N*Erma;41q=jPt32KU^Z?h9%lCfOf9PlVezR_067XwK0tnG? zCSKcmurh9-m}I4y$!B^`O{zS{9f4=B|4d*5ZEFLiKtJ`wJNl6V@Q5TWzJ4Mg?y9d^ zKrkKb!1C#l;>PR6xC=J$nMDqw>T?`vps^1&XQ`Vx{Uf=;m$~}+7vCSYThl3)+4Vj+ zjG1ow4<(-)w=4k1vCkF{V~%3q>?~4c~mza`tTk&ov*s&-_XP zx=Cp3rfNzyN#R1~-X79Yp_5Btx0ccTU^P%dPPeVu+{gLXhGI=zE+K^P`Z^Z#%VRL4 zw2!1>cQCj+jdV=3qYia#Tv(~z1`gU$CCaO-}%_G#yx0l!5 z=2W>CGPZ2?@67f0(4J?3uhvpcu#Z;k|K2kVe2s3H3EfTQ__rJcP)Ruxq-XwgeXXtx z#=19x3%&-DHZQQrX(`Um(S~?KZn=e^m7}yu^}AY%Cw&E!bWp#LpLDOTQg$_Mt?!ab zWPVlg;#t!T;U5h4^$Dh0{LlGaEnlmyO1}lWo3it;?o;32M%^AatN7xDv&tQyXOWO% zdiL&jm$=zGlTj2zb0pASieIpntMf8X=|fnlGpL?m`3fg~tMi-K(83(HGiDmh8q7bp&*+Y+(-I52+%4j}b!bz7>F zKc$#2RcrICV?TtZx5ykPas+yD^ z>AucsS)eActJm?{xEZc(&>3`8uHO)LychEwQoZn-Dad-aWhiPA$$O%Y3SdGiBhStCj z7)NQn4*#N!HZVnDl|zbe#_*9HBj_o4Py^y1A% zqi~%wd&!6MUtu-+zZj40x z&3Fd_KZL)Y+3~e;^eMDkdwvgwsOljwYq#`a6kW-gqShs|`1^j87H!JH_(Wok(!Jr~ z%UQlf@on1Ed)oHVU?R8Z%>)1)Ewj_CY3m}IoVVlgMX+Ib_a3dF`izaQfIe@Nx+*Vh zV>bGMe9aEr z$|OJ7Bsr+1q(|-<*}K1VjaH`ZmV5wnNE1#`Z(p+PxtL?t*wAW#q zl%3<;^ilyy${8dlD^@#fk_PE{#&k;FnTNV}cFo~YfVBWc?@hO{=UA#2gPgPV_VQoc z_y6!9>oFcI+u9!cHy+HIzJpH7)>TpNh8-G(Yd_PufqkuxQsgbb%5R=1@%hc0sWbF$ z$VdT)FYk=Pz!f#kK|j&)(1oV4%lPvt+0Z>;rL)Dc2qtjDB#GbO0F^{Jhh%W7FRNny z5IdQ-RBa1N(xSz)p<8ul)w>d1yR_g!0P$}JpwMs`6?C5I3e(f7-jY{6I{~^Jo@h)p zPccY<`s4WF?(djq2qWo-+#^Gb4;)a7FSDSHL5bwduyuCR_ ze6n{2;cAM{_TYo3CuE(UM!W9h8h_>%fB%ItW?cC-CnvLdG$XeChYq#j`{AnF4q<#n zd{eA$L1bIw!CrxBN-vtx8DY16GMn7V&*U6^vmo+a)w1h>{8JJpLB38Zid(B5L-1`G z;87Aa9B=1hif*djA6=f=?G~2Nw zpAMH}Nm=?Oqu;JlM65hxRZNlueyfjn{A8<70vxn;x}O-@AYJ5{MxU(9Iz9|IWtl6< z*-@1LSxH_9)~8J?yic6x&Itk8W4> zGoUjmA<2K#%O%F${qFfk+H84K&Y$5(dovBPlV~VG{2(@)Pd0LK&UBGTGkH;9*Y8*p zRbSe6@l77x_PbCvKM9Prj;q}4##(;$q=zL+?-2)aRAl%EOhAFDIr1ol=axiTLNZP)?GH!P>&C6MAyI#heuR}$;ck;=cC+QA-$$2KBn7RK}0rk>Ep*Z8< z)!}RfYIR0Rk;J{1%Q=7cb02A!%TITcQ%~x*ec|Dzi>B`(v)|VH&4xI~!VJO*_U1NV z_3Pmd*>4RX+fxt?0(xzhl9pU+hYdommECdk&n+fZt<={d#8`_iYLbG5kDIG(RM8%1zKJB@c$l`GwAI&$+>>V^KS~2gMrW* z=z5(Hl+pCJ8Sq_Du%t68$Q^l0HS>^BRjGWGV zFFpy!yJ32BgB(g7Id_hov zh!IzqP{IBa{sYyu`f+uw?wwomx0qv6tNH1ws*Sty^v4+oQ%knmm7*Gatky-XARUP) z*UNCZs876idHs5-Cr$VR9r(Mq%Ua;|s&H(|@YdwJz;K1e>B zTWm_N$N8I}(Hxc8FN;07MU-_t;~oyB#zs{Qad%#R5ZZ$BWwA3Ttkqie1309opK^ER za_aJ36H&C}Tnl*iAus*unN5_ zyQnU!(C$XJe#X1a`npc|tWSGRTTEwnOUt|Y^a-cubrt&3s`Z|9Fm4r}GHqFOlo{Fq z977Io0efJ)Ted0$m!D{jch666)@S=X%lxpla7_gte?kSNwUC0mEjwM}O*;oM-sRh4 ze|pDm3*)6aAi+0bwar_-P|~195j@0nda&v^yod)Qm{G87#03)5Wb%s7ir2rO&2#H8 zg+W`=^*wZ2ij!#2B*uVz-=fJ&;>}{pjKRm) z%nbZXOwpQ(6@H=wQ!^|EZ-??(u^RV~=FftM$p!x?!t=l*;>U4Ryw2&SR1x}0+Bsku zXe2++lG|qUZS0TJ+yWPw1Xd&OfEiWKR4Cyi9Pn?3uXdl(i@{%uj8n0F8~66haw1Wc zV?;|JMNjnp_o3P=oM?pri#l1CY5<&ADJkBw{0>X7bY^J)~G6t!L)ir-k4oXUd(FiZGG0UP#JA_h`r3 z!j00t0~R<%l)9`|u`2Il=##wP(jJoJISe|Cb6meX(5})~8+WtClD{FaL9juYn4ijg zfaBDaee~VCS~AYMf1BOay=CSg;8DfIiD5x5-O(dp2N@w$@epPw7kx@DkALsJFJD8Ueq%W&KyWjZR^T z(qWa*2)#4;aEn{z%r^L~^kyjjN?fa`W5>$xq?q(*sdwPO+!B6hG2y)9#2d&J0^|ft z#Hcg2$I0?%<&l#xP|+y5>tcTh9>J?)^|}{JpIdIR4(l(5r^nQ&gLrJX_CZ;9Jw$jd zf1y9TPnG1!WFGXcF8R^{-^^}M*#<6}^Uc#p^P{635_O#{g4Y8F@^#NCRL1)EEf@W{ zjUEXR-#ocRH)N`y9K8SF(QtJb>e+mIk#ypttbPrI)B_zM87IFWn=nXK!dmp<#S$v?f(+qWv&zhTYBu zx4Zb9-0sIwn7bz3dMcC@@`pW;H5is6`-mffv>Mnr<|n6@Ib=F|ipIBh*ME60dj{f@ z<3n@_c&;yJtowlaU$L^{I967uKW``fy?WP`Puxw@bU<<7sdQIe?napftjd!2+JcWA zHcpQUoWuTc4SQm^5-eu<)mR>xpP@Ov!w%`tl0oh8KCRLoI)Z<8YZbH`b_U z?Q+Fq&ZFTt_~g{RXn*)Rj6Uis{2)>*YR71k--nS;J2&WaHd61tm0`>jCQ?k4Czbit zOsV!2+EV}>7_W&=*FwNcFwdli_>%TYPPLp-LmOxb)6Bl z6Ueq?`^YIA<2{ISmf_1{S@T<)5>h-WBBU zo802LF5JThx!ZkVfV_ocvkq;gg9<3G%KiQr;)I%eOo(rmr2n zy?3bAc>mV^JT_Dfz72b`SaSsS^Km&8PKj3|uaNnmjy>!&w$3IUC2=>>a;*|VBL^N@ z@mGh8_yKY(hFlwuQ9e?V4pv{+eV>ZF=3|;b{mtkn1!lkR3Y!A^GDU=_)n^Q!@Bg`N zXC*x4)AgLtD$dg{lCrRq^vbVP*xg$W5dApLuoKB^k!V@W@{Z4=;l0W2!xFot!`92N z@soQ5mpd*=)9l{<7d2K5^OGc+Kjo`MN4M|oZO6o!y9+(A$-X{qx~64~+A@ipC<3={@Hm{>7q%^r($?2;K6BezOGIwR^$! zY#chJv8$>e4yF?&dUg35O+C-*hsp-Fe1=cUzJq?!J0DaZcFlSLuX1Em(J|DyGl*IL zL3LM0!d>2T#~f6y$RtrbS07{$FqiJt`}3K0T82XyAyH(BRt)#!iB}St*}WPhOVRU6 zr~2Lw-A;M@Nf`EvXayQB(;a{Y2TRV2HEZCiDDdG4g9*vVlTYH89ww;fwOH{U*qPWZ zG-;=N4|h^n+K0t7iu_FQ!T{tP77|eDGc0`){%T-#TW_3cA%rG`4ll8EPSYXQz?1!C zHsdpA)e0wR3s9I3<^SXAJ^ZQe|Nrs0R9p=l6{4(jsANS(II9v-j9^FA@hbd3E0 zBR4JNBnR+XXRnc5K|(b3u7h&@&GIgVt1Fu&qL@t(+DRbK=cnr~T@E8mdT44(TWz@0 zQ%&1?5olPSBqc{FV<%NGR&lh}E0YCr(M}$i z07@65CdPN!!uzE_7Zi%)?#OW*qI^-%4#fO>VgKuWnRG$|3K$ryLjMKhZANk=FKj1E zKGsX7{|hoXeK!@uqT;1(TB2LFjAZ-nZ@BtJs_l)S!w&`ZD#)N1Ks!cJSV=DJB7hfU zF*jj-@psH?J?17a$dLbyN)+vMHj_ZWn`;uhl-dgD+U8dSaelGy!VmrP{Boa5K4kGP zViYnZ*!0+eUJ5I_`6D|(EQRu1BtBYRoDRnGtgv)8iG63kFrYJNEVMB-+o=*%VP9QIz~3W!Ib&P*Tn{|D7oIm3A9 z!q1mg?){tP;?ljJN>~Pl2WymDDnUe^hu)PlwPLZH&xJl;vP%&eU ztFyB4yEG^Kn2`x*b(*AH{Q}K;4uXquK|EklkFKWOOqZaOf>?IM3xaD`3`^46RMyK@rmDg*wr$c%9Z*%Q zQq3EUOsvVILfQdHey$S+BDxh>fjz^uj_!+3SkNlBf18ZQTF%lv!}J5CWW=tw9+kXU z(^Ts)r4JIAQ5rRWVDy2yIdXqQo3mR#axqPblvB^8sS&Xy(=Dh1`DAM8g^|U?C13%* zvd2~H7&{QB?P+2Pehn$|#G2Iwj%RIh?j=~hug!Toht*1$xu@a5XajXlh`t@B;!v^w zvWe$;*zM-aK0h)gs(q|5yjK|K!{XCof_C4ag=otD5@p1(S^n~|$=xqMPdcMsPo1=O zL9qsQ9#%k;6WSL}z1xt_#wpe1N-+S7L*(&3@@(k-Yy|s7 zE)6V+H?7)&5{i7|Gqy2uV1$$ro7!wwn&^~47H=qgiJJZzfkWUZ3S)nByMz6v7Pom} zux~{4!7*5d;GrtOg5QZBcYJYd)gFig)F^ldXYA7%>q}FheuM#2Mgv+C@M<6b5yeC2%e8wvH29Iu=78c$aTQU)&U&0 zQ2q^au`sbOn#=DThjRk3V0QeW6v6za(4(F&fVJYNMwTQXySj=Xf~({mSC;|vXOogO zcee8-0ILs!8z5aMw>(Ht{|rP9FUrbn+IC~>rtPo>DO}MN6Od28{rm=`FkcU60KL$Yvxwi8)mjT zBw72~Qij3SWv@lMrnz8QD3e8kTuG((M7m+}`DF@BU&hsy*&8>qe0k4EH46lCnulQ& zcw-=d-c`=dRaw*iSSha0u|x{7IIq#wE2F`ctH)3{jpeu*LvM}B z;Q5Ld-lx^dY^hwFBBLMfk$Ywa-or~J+?_ts#EzO6p}$Wi_htK$oRL~KTQ1_H1a?Pu zWsv6{I1b#^BAWI}HmcueO@S=xH?VjBCYGudub2%Vl+TCNqgq!Ni;=VeGAay-ezxk1 zq^{NGUtvCQVB~y>9afSfT&YX*E+qt(I8?W`8~ldaOOUd#GAG zl)x0LdO(xU3Sb{tL}RlZ6Z~WEx-W7&H4i*(B2x4Gz5ZLoGAnd)%nA(=J^p>n8ZuN- zni?DnU$ON-c?mYL35f|IrttTItZ{T>XS;eOAPK~ud>@NFjR0{D zC;xs|LN@%%!MY4E;0o;m6HS1%`gI1cP|CqWAhdpg?HFwR``ujR<`;$v&q99VYT0_B z)Z~E6vtobG!rsj)fuoKqSBL2P_U3EbH`fe=iw$mU-}Fio`v$O&(uDN83`B)>Y=4X$3yc9vR?k*$-u=W=|IkGHeAm15uL z11JAE7aSYDxzw*MTEB#A`55COOX#n43+VURe=#qD<(v{$!y^^03&C~`tOvMX$A!HU z*z}NhP)Yy2QT4CwWIMJ|Z3gaFTd9cqNZW2C2BMMWYQ=O_)jef030n7Z?cMf=R5(87)?9W96m-4oQ~V^Ie%{ZyE_G<$uDJz>fnq@wZX%{7}E#A)e3 zC;~yhO;C-m!9403wFT(f{#$59{|jhs{0MJ>a9glk0DEB~ec4C=Ke_N~N?NYJ4=8J0 zc?kCBZTv@?sAuf86hfrQGd{(q{*wla9q?;vogAKwU!JJ;9kA2zmkUCWvtNw=} z@za4vv@4-B3vS2zUcwH~?zEn;jn}b}d@^q? z43g5}59i=;k$w6#s6O)B*ZjQzPk{OA5U@yZ0*A>`Wk(YSpKTiNqxVJ$518|7#N zmMW-~z3ags=0K)SgUZ#VGLB~S4xgRo`#augf@&siv3uy)y1GF(+@3tZTZ$eVE39MK zTpatpb$+XJ9qGK$8DT3ny_=}U=kWq*vAbweW?K*b&lqhgp6n&xy55=5e@&Lj6{aG2 zQluVCk7BR|IfAeXKMb9&T*!<5eZ5EkaH%I2@HUlcT`8qI30PYZ4YLBA9U42<$zat3 zzA%t%o%@#-1bjrDZ(pW&D`D^V*b?+v*dCHzt{1bMZYyRHo7RsszxSpXi=ev&V5f9G z=N7aY0CF0}8<`(Xi>bdUt-R$6_^Y@fdSLGQla2w7i6gbD;~a<0c9n=iguVD#0?P@@dgG9;}OjMr@vA z5<)d{dO&_I$?=Nt(4R0M@J?!_*8+h2>(8~==LNM(X17weXxHlW+N&Z}Y-ay*w7mW5 z!tD-b1Tj*g%AoDYtrp{^6RwO0&RW94x9K+Ke0FLmjMq=i9b;GZFQ; z>20$+3W&n(N#6T)v~3nVpx9oqeN1|QicT8GfL|DvSIiY2Wb{`G(7uM2!0WH3tIE$s z%7;`~)bojdf%|gUSzlLR-e+lc6+pshOX`L%ofi|NYN{K&R&}eF{KXx$(MBu8-KCz@ zm5DI=eypY0x7X(|SH|6td=&$SBYVmNxYW*^7xdf`Uw!f~o6ggVF6_8D*hC{af_%v< zJj0izAKh^rhizEncWNYax!XKHP#LmBFOJY)1gZJh3tf4v8y4tAE)1T(4v-PecH-E(gH-tO!8JZXY~1VP;mm&)*|+nIMTR;5l0Nwl z0>mdW%k**}Kk3OWraNh~{-Aqp%R+Hl_qA}qaZiJb>B;H9cxUA?y{gnWIa6SGq4UNG zZl~F1;dei;Oy4gb5$6H1Oc#_0(J+wyt{C1CWE-849P>E`k*dVj3L`3OtLxVSQY7cJ z4Px|M#ybHMLD_v{#vpAH~NEG;ZC1a=6z)o1*gPFBKSVa^iF z5teYTdzccFN_EL8ic~P-k6^N8E_b%*ktX;|_zdqS-L)}T9;zG=@XS0_RP9=~mhU=T z$|Ay7j|nZ?(BmnRo8H`zSMngx58p3=&C zc#v?Jn*+DJ3A=`o+jXbOFJ#p3_oSgGEBW%=+M2j6O8+%-9+ z>+P}EI^1ZA-pb{zHxaK4;y5Anc+C!WROIns2i3nlbYpLh^T|fSh}--bXuWTw<-i!% zk@vx%?4VMoA3DotYeh6+xHDrHo@o1HZ=8s8+FjY*u7+12ajq4UBxxk7m`5v4-3*(r zm77g|GB&Qtv$B+?s!7RJSC(ISF-@P^koYh0&vWY&HdMlo&ggOe7s_NUG|M#69?1Tq z1+Wi4T7+LL;a6!5tt@W&yDqZWnWA|G@viaCA3Yfj#8i}CM)xx-=J#4}SPG(#ST9%t z^-lH7n7^g--()^_95=cgNZ7w;PSPM{34Tm&NbPc;#D)Oh>P2!e|4=oNc-c@j-s0P$f|%J-k2q zL0hGhoM{pE3-$w}oALd=eh~t9PzV$f(8}VN1f3g=kjmGb8zFoR9EoDMV|jGC>z^UV zT%HT2z-!Nan?O^UluOz~a`?}QMc!R~nG`&fDC8b%Bd}Qb%WuRv#iP<1?<-* zq(oyXif<4?X=Ozr;itkFCOCFs(a|oqmF7y1V68TCvuN&MDiyS~6w|AMcSY0X@hc-O1*gW$=Hh6XcuknwzY*VH<0cLx75&3p&i zS$dH7yIJ+@{?M$5}V_|B*YDYqge~2eBOBG>RT!)g<=GjHt}(EC|!KG;~~K@q%HG9vc(eRnm8R zzfa?qp_8Xlt(YS+ZfkAo{YWpBQJO0`8eeD85w1Q)7>T6mvl!@5F$t-dnxX|DL$BibwrOWPa%D3GkX)6^N7zZL{ zSrb{{HQj4SYJG83;m4Mx<=216u^at}M;{9@e>_pzA|Ud_ zuH@z+%>{M`&v+^}e+1Qu#N*W`L7ntIQQ(YzBoWvs@3l}jsn-tp7*>vjMLHH0;Qmy( z1baLgD@Mp=9|XLZn0C@@Yg{Cx2!I%HS9E2so}E=r%U0&)7Se@qrK)UO+hRStSj-{O z68M)d5BQ{&gPkkR&AHpDZS~%6^IFhWp5tqTI_(C`d01O~<2UcIQ3J=hRMz4W$MDb} z6j_(X40)OTw!5ym+_Jl88AMRfMaa*UF<*k^xcL6HiHtz0YfkL8$d*dZ9`818k!jC* zV$gwI3tMQe?;DAqNp*2|(Gi;A+p;9(W`uL)x`CG!(mk(%CNO4hh9s#n! z)eM+a#x-Oyu@ntwp=N8u{UqsCY*HE#nzWlo2b4)40a3uraUM8Pd@ITEf<&U=WDE>O zIo!@d3v|Lv6x{VZhZgC%T73bfz8A{iIWXr?{dq^axG#PO_4!r`r6iZzPelB zSaoJkcFbYtSB59fHU1g@_kr}CN4t*ng@V);yBDkX=($0O4~5Y=I|@jLebHl3n5pfe zZ0G96R))whTS7qN#bRL#ijkEC~T}3j1@0qxmKto64kauuR-s zwVHG}JFs-%S+`OH^=u}Xh3ehNU>P3+wYei11>5HVNyQ8lWwX%@Q`c7qoT3)L98LrQ zeEqGfDrrLwMjjsYn3rH@hYCbpU$lE*7d(ctqY5*OlS!d@K4&mfRo@?5EO^Oe$@{*y zWU+Vl6RuOe(&ut*-O*Yl2X#s^<&gH8q}HDgTtYNRJ27F%c?(scqxqf|--JnUuA-WA zp+<&sUyT^Oxfv5(sOlx%4)%K*L|>A>Oq65~d*SLBm;w3PPhr^~d8ng+GZn#MSI!p$a;g%@djgASHdF&sR zb>a*OsOpX^<9Q&~1dD5H+{Fi-NXMUfW3SG4j{jm{9rtl>3r)pjr4z60G_1R z#C9-Qn&^RYM+8I~M+)6Z7nCWdOG|wh^d&W6zxr1EVlXZy>_w)|iq}5yf3WatgKBr{ z=_F_8-XNQ9wUHkeZ>B%V*80Gr-6Ki|$%{O3IRq;IT6Rp62Uf*+SWP@otr4$EPH(-% zGz$F|pRw@N0`NeNA*oI_(CUYWyuTnafS$8A8KFlNP`Nr$ z#Ywzic}&OVP@*RS)~HQ9D%icP6~~&;2bg6wa`x_1u2o3~E+g6#IFUA1Z}*Kt@Hp$h zZiII5x4wRz30<&OE8=y2i$P?yP#H!&`>>-9Yo)8o{2pHWHU?SW4f;5%u}Io}=N)MV zk*NSPiSvbEiYIlI?HRbRH`3YGDdZkA&ey59_khw^H$8U3l6w8HEN}&ozqzudG1Lj~ z4m!)R5$2J620ttYEfQMx`_e-cMmoKSpFW!eyjPt^IAI%_(MbDXb2-sF;M(bibnx|a z3@T&g_~n&vU(4$okKGcmFqdA!PrPDbnvmtsJ)RQC}c7kRQwj^@#_Aj>%hFzP-|<$kU^8=%T2>hnO5Vlem@5u&5# zxf81c6iNkft{)H@#&IuCmqnTsT}++-p$!-h;PL_dzPrG>xyQgs1W)P>gxG%C31>_h z;N63WVgnsuHR^0cM3iQO9?6h-0M#4oY#I2bLHUiXZg-=}nu=Fu(x}UGWS#iZ3!A+- z&%FX@=IDmx-ngXgJXMy@C3M1pVk~mYL#nAqlzac-B1Wn#A@)reYW(L&YRtrYx{T7B z{y~2~eHrp6uhZ?_?6BDT&hso8vR7;_hpwM1p`(rETX70zCG0eJ@eLqLeZmelcHTpo6g z5_S91Y^i~B>i!e@hAQ#}?wQktoZir;C$9Kp5!XDpTf7VWtEp+uu zhpVnz8lQIUGR1_9^1MR4qG7kPh({jrc{0uCXc;NycZ+UHKK8nU@ek>%0m z-C(NnjrBg|MmGV+YO=EJd9JSb#eD%{A+!!jwc!k<8q4(IVV&|w-3aYCfA!-3 z$Z-@;#BRu|L#qEb#|a#6wXdC7+CDrQlPf5Rd;<)w0LqKo63Kt9nu~_%`e~k|6u`k&}u(a7g=iZeq)Kt|um4!8w4gm7I z?ZbbhEEpzRR0>KXBmaoEZcM)zy8AXuC{-<$A9OFbgmRZWM?F4mz=*=moZdE~WJV5^ z6U>HhX9K^4LK(JJjA5%Argy!g=uPG8(HawmhpM1eyBTOc8kNHGoKbh-o%dk9puqV+WqQ$|1(M87d z#Ruj&{2Dn-Ui=P4=l7wv*dxLmgKB*<2 zNgw<9zZPdBNE!0)q29R)dpP_UnWovqV^U9S4D+7KL8@J3*^^ucyq@>HBHXaZg)^KG zRlXZiRR5Bv@ap1(E*_-(cjqrvW!=Cvm60*5GM~SPp9pPBc;U4Ns}fme1#*d@B(}Ji_1n1`4TD~z)k}~fo0XcLp};o9|t>PQCWa3Ey7PG9 z^C2LlNSN>&;rO<;{S*a+ht2`j*yr?nkaU}&U*Q7#h3c4iGfs}xZq>n5o1+eO;UPOe zgQ~u-JdVRL&V5h!02cwBmG3BDjPz%eB~wJcXqtqxpOO9G(NUow%Yjh!;oQ_=x~deI zM?&jYeRS~VQ&)*AdQaDX#ziSPlq{jEbTxWhX?bUAi=%GL?L z{!;)AU2fCqXx3`JyjSA{)7uu2U$LtN(eSA0;;y{RrAm)jiTVv?j&p-xV7-v^PA@HbP*}`s}_Ra%f8a}t`+;MN@mT0#Ez+P9MqD@^Hs| zJTl_+GPG9f=caR?lM}!tQF|F~!S@2<*M}>)k7J{a>a)PSU<2h%){v=blYVutegn06CzJm{{r?+y7*79=8*eb8NN<_e8VfR{^5lO zzB=}@UW3)z^LoJTryuEk>^qN)6fZa16do~5hB6NGo%cG2u$Hyt+LgPX-s6UTwSR1w zeazgcWjzq~CJO+g&33MAF73xf;V2`Sk4fwH{~663LB?nr4H7ma{|#;B@-wmb|0!{k zn@P8x&=V^LtUpWA+qFPjLir(an;4^;`}cZovq+K$6L$lv{r%36lcXu={auQO1<#Kw!`aB96ePO^^HFVU45NFh;+=thP z1X3QVd;raC-3-oasvKR93$#BZhSvSXH$*t^P<&y+v&Z(GYCm?#YlF*M>-FWTM*Ikn z*NQZDd_XyKzJ%Iws?B@4*Gy9ON@X4-cUu*h&-R!V4-rK~R61Ia!g;!9+=vkhd>kGv z;nS0uO?d00<`!Q=F@jjoU>s%DpHNO>9mvny)=|gujqY8`U z)>M}2l-H{pQel5nd>>8YL=eEnNn1E;H|M``x-Oh>u&S~)Yy^?(E^nqx_^0k(WyTZJFKN5 z+A$<*GSu-`7c?Jb^z8@x38@WMn5o4$k=6pbR+D&FTIzE|k~ip2xs0&Bb)TeuC*1F3 z?zE3wlmC)RImnhkRXpAS4X7L5f1#qh)914Kq^g!^Z$9OOdsyRo&dH2)Jwc3N+MQ_g z*D3koPA)qtK?%W?7SV2X0-nopMpxe9`Z_m_mWQJ{`;B1I*K->q*NUa8vD5#h>rzl> zh;wt^*Z$ult-*{%|F&L3NU+YPh$bTP4NDN)A6gK^!%0f$gHK6ZtbmuQS;|4|KT`u# zBS!$j=Awc%Ai7FS8f4wTMXPcO#IugNth2sUm;|Po8v0efL|TqM-g;n)()!zI!^_TO ztLjmBOi7)4Pb`F`(Jr`MN#k;z-WNOdxX-bI1tO8g*_%Tyu_7`cstWCH#ovQb?uorRrsrI zrc3i`MwuQ@lPKoKzUYS+_I06Vlg+&wCe0b?I~$yUK*@pDM^l@GP)7Yu9H5Y}_Lb^4 zP0#!spjwciKRZk^;bO;ov)NcR?^?l7pA{9L8%=x9^EQn7@RyDuPf~5GrPA1ASVT#g z?773m75*4@a^Ae9pz*Ibf?w)#>NFwAt(TCIfOt$G?inpC6RUzxD;su*bKNrJxB(mV z?DpB^1d^y(>9NUG#2{=cK^J~+<23n_(357KC1pogQ99A9k(W8l{+kei#ohG4McMHI zfcHxo{_*^E8CLhyYnylFN4`Gy?Zv9Gs0num(a{^;jf}K&zT+Q>2gY!sV#Wqp%#pvi zA8=Ir^N4fMuEW$Sx76oK=Dr77IccIxZQD%7bP7o8>FlY&ZS!Dm=P!5hDGzrIXAK`Y z)~`vJoMj4hJxKm7V92a_sk*w;rpEWR%p2;5q%o-=$*% zwMDH2?PwZ3Bdh$(ozA2$HU3Zy9dHK}kd2(q(3)_Mw!s4d*59OD25EsfUV?^lt0vzd zp;a+ISh7s5xE4fQ->2G;@$~=t^8f$xD}4^rRqXdWNx{!95ovb4SZS zxZl~9F)`#*#=zfJ4%f4Ze+G=-el6<2NDqmPp5ec5kTn(HTM;8q%I8!Abku~J9h56cTQ0Uf(}!ItmH_}iD=|9DN)kfRP; zVF0e)H>2--Sqc2QeXWc6=<_uX+)@tVOK>}S+00G;@$vlLgjiXbQGsAsW8Ei_h`3tM zC^rFLC(+flm4*dhuOfjZ^l#R=m|{|)lVj`#7l!C3%Tl9@WtE6>mGUg66z%8Yo#8;? zP&29xib;s0(Ik#jG@QfP;E+8*5$~tQ%5#qFQE(v+X-?k0IWO$kH>PNOd)C@|U4RKa zsu;(J!=yP*j`-~25KfLB02#3v9GDi2p+3I-Gjq3iDRGG8oyPPlYeH$6eO&<6Li)n_ z1ipKAo3~T(zJ0RSAn7=skF%?0!z3II=vBqV-lgD3G}RH*;reTN^HH&-xUgcj)2kzm z$p;-(O8K*Y;6C!$BhGbPXb7p>wsE_ycus9BXxn4{d;U51Lrt#>#Q%|U*l;s6W)*Qi zU_Sp=JkQ#3B(r8dF9clywjs{`Vzn+|12(x9$lXcHXYU*k33=Tr0uW?@3#P*Hr*>*F3npbwDWK(MJ|#`($9rqFLj+2d)IFA}ox zp4O?l!x~%6Fhb|8V)J`%Ulf%+xG!hrh{%l%du|VBRN7zkfeoTTHnU<( zPYakzr?~haO^YkDXNp%xLQ0yoz=i+>iJ)g|yZsiE9I#%daHn!b;9&Nw)JgHhb-B-1Sne*eJioenLEv9a}(lHHYH!()t>RU2FzUq;EDECvx&h>X#4Fe5by=vS{ zta|(E4=|K_S9w&)_4aIB2j5!XC+dw9{-jG=e7-1!_86gw_m+PjEmzcpq#5Ghc(}2@ zlNdK9J%l(|@l`(b9i?b@Ju60MCLSOmVUv_N*Nz%92dd3LVZu%cAz>_$k;q9IA$>y) zFb)&Lk>aQ9o|^IFnE;e6e-8|AAftcq`)+FuWri_^zwML#w7z+Q#LA?W)(k{--K!my zi3Go6*3JG-g5xcL1%GTHy2oenXel$S8WlnQV`=XN-EC1kV|aJpNLtRWCdnHq!=#+; zOE0G#(|umI2-8c{q(!B##BBQX;Tw8`pC<|lCufgQe#x(v-1#~lQ_uHv`05X^)<9@D*_s^Aj&s#q zW6^Cp{$C!-_JkdkG^Mxv!;XgPzK?ttnH4>6$8~t(c<(;fj}1$Xr!i2Ei_Bt+CLF?c zAjnk2R+Sg8{9N0oZ&o=O>1n?+e?P>(#e&NY!;;*(PEe%|hJEQULT~jiJRG7>^nv!Y zxf>ay@a|AK$j6}DdnvkRdU=ay$5U!ABJ8OcOMc3l9KFxn2-tKRJym2!Xz|$0DK|1m zZ#u&n-6XF%KhaBAHCi7J>X-ra)*OpWuY4MfUK8!^Vk|qOjlB2+7h zDBjtGU};Dq8f5`+9X?tYB_0k0rL53oP#S!Lj+)~O$NshQlORVn)g-0*toM1?X{XS8 zN2OrywQ6|p8pr491A_dVE8;l7MlGszC`l0dTj`+@(TXUZA#|=r0-&RUO2vRpLVXxt zTB`4wW0lUlyzlid$BcNf>yDqu_n$&<1Fzee9ItvyG9sK-8^d)-i2le6F>9S(HMncr zjRn&kOOA5kKAr2VR$coq4#m~%!__V!qw?iDy_Z)K*y9^_*lOy0!racw%zqugyn;Ei-1lFDoZ#sl(EL zWO(XvaIWJs{E!mK`Q~|w zpdOF~MC}2dvHzsL_$QEM^gwMWmAjVfTZ*Z4p+ioCK+pS|oR1TCn_+zG0Nu>XUaK=! z7lcyQXE+u&-GAPlzQql>#l>!Hi!Z!<5;F9QGtk_0oaMr};S-+y#UO^SA{g~}cdOa+ zZu3XBDY1;@Jsz*UABxDWFhaE2PwSLB(iV6@0mtGAc8Q^jp=^q4ch3n3GE$>z{Bq=% z3Rgy>ID=AOdYX~U`5nbs{b{!Zvc}PIUIAHJr?*nzWZB@|SQx|FDglsA{S9L0Ro-Au ziyL?xI&xqRuyD+m&lXYNYVC|wXp;nvA-uX$^{6qh#miD|ACGXpby=5=CqyT`)TWgt zD4K*(Z60tj^x*G?sNa|!HnyBE@I@y_Re>VT+b1TZIbwgz<&>xz%eJWGgoAQ;o_h6N z%KO8Pn#F~jATNtB7@&ppP5#nK$38)O^4nGi-jvQMsQIP<-8-o7h|+XxlA}pyoQ0k+3*ROVy#T^gm@}6j^pDhuLXWT z6R#;kbobRmI-FnGBf;h`EWqYvS&ri$mk(8+p7uZMH8lwIYC`OGhORpi#^-f*kJcrR z$1r@=ch@&~MdCZ&h}sS&&}-9N!)cy0rTD^)iZ>Mf|EFXd(PRu&@%Sg}!hfx-{PWDI z_K0X_$Vk!n6@5d!a&7NnfSu|SzS?R5eu7nAG+b!dqs_#KF zA&ZPo7b@uKHL*km_g5~>mxNelIpGb>%6Q^W9;kHRS;pN+D!|#aO;9~;Em3ET{W^^IPeqklgg zNbmgQ+2@~tn3l6fCFM;>Qg{r4&&1~L<&IYWIlbQ@=`>3_TlZ@UrWQXrIfo!`) zO=&?Y_qn>G%hinInNmK@Ar*el0qZ#K(upB!5LCo7@%B$?!nbKvnmJWGU#;D?Xt zdTCVS(lMO{;cSptsiy%D1niFiBCtUvH&d=(FF-iIi7X)wr3wNZ&}Jl!JF6dD2TG0& zS*<#TR=7ahqau#1J!3iNjo;pvWCHFxJ#}CakZ@l(ChTF>7|A3zJ_w~e)NpzR=>nLg za#vp*{pdXWG>SDX1h$|fZVPOkVTftCPh<&p-*(5WqdgkM+Q7!PKtG@xa6D$gC@^#F zBeX(zC~8)WNT1oexFF#7nAD|YxbRB0Uoz*L5L0@K$9C7V=Fx4L<<#v}y&OmpKqktb zEK9+zFJieC4CQ_N`fmevI<7A|q_wVL^czCjc`^~a%8581`kr+<`y3&wzQg?wqPZ_ z!ml(tv5y=HABnucz5#5d)VV{)$tYS)@r-ED=vc90g)ao9V}7*vy43hV z5GBZv$VoO)+u82xojtUHMm3p$8B`)a&^2(XlD+toA8vLa_p1m7;fr07=<8Xz{o7=3RP8~!LSDTf|Y7_ zr`P{`^hYsx=uDw%`afEq>zA3Yr;9AOm5irMVPh#uBeu*X*mn!!;e@X&R>L0Ut$B!e z|BzQY<-2OUDbcArAlqcmw6I8RzqB(F4iFJV&NI(5VzRKczz5=g9!hmX`1PErBpKuP}yyhFLcO3l74Meuu zdRpA8S)CC1seS;LbzIBvp*f@eV5U+GT>@nz{o?o9lO)VuT6A9ItdYZL)N$e&;pVMc zRnQe{;=LHJqqopl1pt|0Et-G%5SQ zc<^GVvjGWx^!m!VcdPBiF30bUtUJU+4MWwh!1tPos)lDm86i~|736m8qZ2!u-2sgu zr`@^hS&aU%FMN?hKI{(SvqSVEshHitpd*qnyuj~;2oo0_CX85ey>7=W9$SuyE^;gi zFcTO3`sYzm=mN_peXaSzhl?_Fyj|dDcIiM-M*Hjm1H8z}0xWr~4rRRoXV66Ge0%tr zEooM_ZQ2Dwa1H8R zq1L1%G2cBMFnUobDeJ+I_3s_}!9Dx6#|uB#o#8G235_seV>}m;rXJq5|GKDx+gSDd z^i=e`hhk*K$4b5HB*+eGW$C#EA{bEJ%je#gAF(}0SP+$?!xb}f$yfehd)gDuG6(MI zigBhjLGJMwTm@~ueI;0KF(j%Nk;;*(3p%@$-F+h@JtKZAw7@M#>&~q$A&ExcZ)^2s z&L2`d%S-ZOOj-0u^~I9U_3^e?$%MYREcGAl39Z9W^Qq!TIL6ns#aDxmu9~Yqi+UK$ zmG;-bP+-ZP(}?HI(C9{pu&4`%*Xv41@>9TjtX3RgQI1p|sJmIYmm0D<90YU#>6aCK zRfVprgzciPL`J;)&a4CI0}Rx#JTHkFHHd%9{|N`G8;jm_4U`sxIt#prM9*Uvmr4dk z8ad>jJDp9dbl@8ieRd4k^6$6?9}n8iI7^hMi_1CDNv}dY^{}j3(gGxt>Wx|5nOVVs z%F~Bojv6WOpC|FWE!0kiVGZ7?Mazs~o^*V7gcv*4X1NES7qAk*=2v~`N?2&0x^}+S zp)F~bY(yA}V#6fyjKV}0LP`K$ZH};`toeiX_XjHFFIC27>1Bg;Ugs?_PNRV&ON#D_ z@~_}#+*0C_z_R#GJHuq>RD)v63Fr%S@_PyCh(?)Egg?dntbw&R`~Vto6^Bw}^QGmN zLJ7SMRSiDphp2{qKB=er^LIE2Gkc5pzxtkS#mrmVU1U1SGg;AnIUxOTjN0mp9$4HV zGn5q;xTh+IetK&a|11~Apv;P%hzq%DD9!&IP3*Dvd!99o_OqJxc(QlJD3Tf&vWf2i z-qg$8u2*j^t3OV)78V)K%HUCs7Ea+(dXG+9&?#*>F(06qodELy*!ay4i7M?Tn3e++K9;sy@UD(%mRG?NECw zw52c0u(a(;e~JMDTqE9Xi=OKvN#2X!#-#;bhD z>)qobCsO;4cOM#UKEEih8CPQNEfg@prwloeWL~tPmHE3JH#g~$?Ns#QEnd>ktKGUd z=m-_3*82P0D&XumW@O({NoN@ulMcf6R4) zDr_GWShV|dcck-4$7!qcKcaeaf3wj2W47T5t?NpcnY)}WpnJ-%atT|n8PuyyI>EJ$ z(+wwALmAIGbU4MZmAI?q3>ML1RW7%`Bgq-@d(GAODU*U4E`GaHC_!08}J_gjDn?%&#;pzEQgxaK{3a zG%-}}2#Fx;YXHP7<_6G-xYHC#8W~ZdpLFaR8>NN>k;&AeD%|sxUS2^8sZM1l% zW=^H25rms-j~X?DvI#uE1BaiW-w_x;CI+xO+k7y#kT^3X$YgC;GYunhp1tL0M7eJ; zpp0xV5(BRk0*4W6yg{Ia_6K1%GNyqUK_(>vSD3ukh%iGL&Td!vk9k~cc43^{Cq=X^ z)E*O<-=Jf65^Se1LXUr|p=oNk*-^%a9FR(p9_DnuRY2)tis&^(-Z@85%DPoZGTHt()LZdFx z5ra#R*@c*Ux?&dPbsAZ_imRkT9nKn6$uaG2OU-<8o^l^nre~$a#VZ+ch;%yTXRD|U zoe*|(QncE^oI1zf2?k4Ed(G8X!E5_6GIJVmllTiNPV`KbE*xR;PCiNC$cKVk|$zjVLm z<4Zlgp%h}O#q#nL=;~6Wkzjjh#P_h!@8g)G-9OyN{>*3=q;kGWEnfD!t(TEn_54xh z@@dUT_J`3i%2ug$2=ReAqJ*H|fscRBJ&2`Wd7ZM?x=meej)J!3l9orWIhK4=Uev+Z zLk%q~IXu}Qx@%$Rk2_1A2R;;_ewFh?Z%z01Q5oKps2L~T$>VH%^=Sz>sBh6R!e7vfPjehz zX>u5Y+qMjx^u_F|TB&7Hr%;`8vVkfa8KI2?4`p9HWq=nks{?kZPYN+I6kgR-F}bQd zNSKiICxr}Q-AEIV$3DC30vMDk~OZzx&CiX6SH-K$>BYh1wzxz%$WfN#$cQY)ym- z#0km#_H}J|y=S)=Z?SdFz4N*og2St{g)>g}186^Z+0ii^YbF014G2{59db!YOtVxG zy*!6hjD}n)tND=+dXo7=Il#jX2KIWVHkex6AsiSNV*p&m-=e28{+Erm4EX?9$tp3W zF5s=nwSd{0dhVcA<^ ziqDRajQEd>g^j(nzdvV5gVAwY0$hllxmJcQ_T-yzVw%^pv*z~JP)OErL1Wiv7c2K5)_Bu2m|MMCv~xR;AdS;g z8a}>hQAd_Jj$AwVZbE57bWeIG&)l51WwQGa5A5)c=?Hyti(87Ms0t-_>4b)yQ89mo zFcjX*OLGzRLS;TRQcY148JJN&J@FE-qk6aYR!(YtD8osFujq?9W)yQ3QjZeQ`hDRR^KiWCu*;)Pl>`~ z=q%G$cvQ2J1-)ZJE*sItc$3!RvcbR6b!QA>eyO$UtU5AC&+_p#PbidKRq?2O* zK=5WL3@H>(Vt6-;;-2NC=iaojRR8|@b`XNH9XeI z(Ib^Y1-3F4FRH18BNc5Gf!fS?N6n#jc{QVf8obsmqf4qXDPC%vIQFA8@1DL}?|KFz zceAOmSY@5Cdm0?bx83WUNE-DRmSZ-viMs+s9Y5x@gYQ&eXI@ZWK0#!q5{z!+!?pHX zP|rqP!#^3^h|z(bEbf$1Lh?_DCs5q+T;g+LykAXHR7!Wi2NQZW+|Btdp%KATTokbOPx!6O-c^2>wdciMSt;L$ z)gSBpEAnPIR@MdFtAN3`aZ+|x>W=EwGbX(W;`W*_d!6*b`L_@^%otIMwHZ4V5An{!L&28cxO{wg$+_+A?gA;eAuGHjn4jlt=jljv=+my z$k$?>eyKE_grK)R<#c>yl2{%Zw1Pf|GxWP-USe)1UF4e9%25KVkbm(9%-e>#_@Vt~ z*vxpA?F`F`yr%!w^Z#koI>QN}mheU5$n2l)ewj>NU4>x~`~<<+71wpHZW~QWRRdOH zb>sc;F3P^9gml?Z4pR4%Y0P}@4Yacmp`smaGs^xZ@j4i|E);52s}Lu0lNZ*I(UW?B zGqz-Uq6A4ygk1AFAPOrA}Xix73qHOmnoT45%H zSm*+%!Fq!r9yjr3rwTmcKsLtp;-2r<8@%M3$GZ2p@mxYl7CY<#9*CSSTi3L7i`zsl znzpP9y5AcMJs}a+o<+$g!;k}(tO_mWHCDSV*m!Q|0Z|9=Qm;6#$tH7gW+ruYV za20qjpk?EWu%+bg#Ac^=`?CuDZ zph;X7c1XCWWb{%=9BnD-91V?YXl?iIZ`rxPWV_fF!oArr9Gd5^=idgs{`G@hNIM} zrl>>V50VkLx{T&!!PVS$+!g*U_lBN@WfLH|o+z@^yE1jaDv=M|VSa(q=hVP%lf!6g ztT*C%kj=}zfGr<#5JaU=?7QK8rTOhrgvl)xn|MOzz0-G0TU2yLBCd`A@Q(`S25$cJ4}-FgSemci=JDNi^#$qUH10)qy&b- zvFhmOU0M~WLUBQ-@ECGx?&P(ic_EWS5}MiYeMOz85 zxATgb*yT>K?J$G+4 zF#v62B#c?Fg`RRbmI_~9V1t7h;6vk|R!lGUnD=FB|Ic%nD|ID3f<2lU|3}T{Gj-zo zyB~}!C*BBh-uNOJQ|MmAPGiSyz&#|-4DZvhRd)y=$U2+9BMb!a>FN&ZLh>ZQ=TFs? zKv%Ark@_#W{?0yCz5#QVXuS;pgLcSyr6ZI7)V+Y$AxzwezIY%bSU#&+=3m=;+@$eL z_5C1pLikGr*}n0S3f!W1vFkkN$VRHX5tiTn^(R3uVsYZCSr+b)vB@Zmi{{-c=(ysg{`e|-%YpNAnZ{lC)0H3U=X!sP4Ec%f9Mx{wcL)yUhgo};lp4A%w+ms9ja>ebZ_T|Irbjsn=hN6q}P_%mbv7;aGqfnGx-yrV?&TX zXQ2L^(6DL7jP)({u>k4@3YdRsEePCBhuWWpLBB#%XMSUcpd-Rr;<_`x%2H?X?i*20 za=ZQf87KXM266h|n`{3#VLrk>y$E^L%!rek+1gri2rpHKA5CD>@}rtE8pCq`kvYP+dRc3$eZpQT5s zfc`V{iMkzpiSojmPlN->HE`-hF@-lmB{(B&3)JN@<>Ae~B!q3=pnje;u#a~fJ{SKQ zq{%wR4|66RERB+Lb~*lAhxTM|f7Ro+PQIbXwiJm-ZdYNDj9pAibh=&40I8Ku?Kk$s z%2^)EA>F}HM^2)xDng^LNwMr7VDMCKZb8d$9=C<+h&oNO3Rb9nw>Hef%5Fg}J3C?t zr1AD|iYye!T0Py~tsDv`7d4UF5ag-0;@;y;qSST0L(YFK+V0Tq0F(G@06iV#SgP>F zc}5}~XnJf_nPvMdAH1OSF9F<#g@>b6iV@P*Ho#EduuDEr85t z5iT9*xO1fvTg+C{8vHsiJN;$P7v~QhkcGRPLcN@&5mWG;Ydi(MD^mR0OA#4=2S*y84>qK#2J0EtOOa* zP^_HV9F^xhO3Ga?bo`!LYYE;4(G;S)b>Sx30{wQf_sQnolD!6AkMrSoeMi{Q%#`k6 z&2@V*aUMOxNvxqgG#>!%isyQu6*bEi^zfL6O?%I9W$9tSHyp|fy%2a}_5c`~3nJ#^ zK!hFzg6&)k{aKQuc>KjHGz1=xtQVDP6)S@unblJr))$A9Qnpv}SPeSpeAf@~aQNE~ zx?T^Ss0z|I#a5n;Uf>D6!;r>vtJ{HsVJ7B9*j^`F=nO=8>v5~9Y*5a3RJHTPor&Vs zdPno@tv%lFZ+~}Rpd4=zdStG5$9nRUk^bzbm{X05Pk|@Xo*RBFa;5t$hx|^9vK_Y! z?rczd>f+Y#J^I#!HexHC0(%PQ?$2$sM{i7GgKTJgV##(2PTh@yQ)dP`B92z{S&H!U zy2EA1`l7bmbvngeWe0Yan!BaYQXjE|-fn=7V>Tt1mA0kH$5PAUZvD=FNV)Le=%tGy z_T(gjEWbvsm>jVrQLjv$oDk@EB->@#aB4q40E4%}YWp+GmWsvqs|{uWqQQ{V#jez5 zj917ag-z8P_34tAw7*2!JUPRl!rvXw9xO9j4$=kpz{&Qxv$pQQTPBQf&Cj%2swWowqST>b9 zZIbM65;f37eRGz()U_Mc;G&r5Kw4;J$L%5O^@pY!QMb)c!)buVaN3!MMLJxmSxt!! zO3M%;DoZxRd}w?(E=lD{(;U%5uoCy1X5QUvDi2XYU?%YMan6%V!gyeN^<;mgiG?5V zO!mc$4^iGV$=Or$`^ANY)=)RLZlRb*`3;FG=2Crk>;;)cc$zVFO)tB&r+)t$<(}|* zcI-e-zo47L36ttff2`x`k4y`w z?X^-mU_Jdhe%f15w$i!J&T5~Lblck~6tejnjtQF)QCs%v-!{a>McZ}CakXfpDiXjg zQ2}vk_fE~mK_Ii{*zC;NAggp8`Sy^TQY{Ijbjoz zu$o6Yl6=oy$@2DtOwHY`Kqpwx$T*#Qx7;AAewg9i%1JPm&{mfUyY2?;|By3y`QNa* z3a;5Tm<7?fPH{iQ_v=5U4?c&iM@L&pcwhZR6MQ!#dz+m3N6BO17u!AtFl(5;$&kTQ z<$UHe=`@cHw(F_n2>9~6;+R)B1!BUH!Xza&t?*{gN2UU?==B;l-7u~D>iM0{7P606 z(?DFZd3VONU6h})va7IqVM*^VtY{CwZAT4x8J!~!&bRCW#S@xRK>`(`$rof*be}{Ef%BVb*oD`$ZMDDkdm@3P@#O#K=2FSlLGmm0wzO-0U=5g z!mira{(A|=K#zB-tfye9?K^6izTd)hKNp?lX=nW%N|23{_NUiAPAmX^eECH*d-3d@ z%*Z_Hsd|3nSvk4&y}5+RgyO{h45laUf6t{x8sn2aVR+qD;lmtvqq9$@cA>O=UV{Zh zEs+OyupsiF#+;k>AMTtPK4nAw8dhD@WIdqKrXMt|8pX5ys|66+9{GETF979QTC$ka z|6P*AXBdyUACi74wVd;tdd)d76!3K9adnNVOFfBY5VLz}jL-_-7uCBVA2-dpzfv4hzt*%PdcmdlLoh6ET{JjJ?@H9p6*NXe~ zH_`+z8l&QbUrE(%+77Kh|4n08KZldsfX7G!B>eGh4_5|?D2T=O^*g!gK6>tHpj41H zBotA4aFYe{;T}yW>GA@ACHmSa8NrVvp1yuw9c<_}&+#(L0nC8&-v+ZI9;(l)Co+q0 zeV|}?SE{P^TG*FsZ_~TnwX#eev3Z$LU6-?HE%NO?EYF>;Eex$h7YNXyW>;ma(KY*i zkJ{P36QTxu8A4~x5na#St%^O98)FQ%)}8+qNG$p)Y|tL}RDsY`zOuOKq^3GK`jkKm zZZ4pUQ-x`DOSX>1$2uaVc;x+4It7Vj`?w!-Ep%tQO)lZ|x7Q)vkSZ zTlOC^W#ohB!-ev6#;P9DZxtekwj6BQ?rZ-gsIa3bme+ zwwB#9>}m)?_4kIh(VY4TDgRy}dUCpOM$^07A>9-A^k=+BzmgL-M9P4zA&m{x&aO{_{@1)LvYB#0+ zP|UcuxjahSgUattdcrZ6HLGFq?|It_v%URncU~YGq;&{=%kt1f%LWq{gC9(g!w2 z5L`ua`r)z1P8Nz5i;HF({tMK;T9SzPIGATDV4unUXDDi|TS^H!3%^6NIoV-8eLBUm?AI9=}FY%e_1KWPM2vYeCD z{+N6fNsWIY1TL6Z^W!bsF*K^SWM=wtdXc%@B$`_GmI=O9e)s?&pV%iYKI7V0j?ECE{E)v26f35n@vlt7`1=VIWNt)H4o0EWA673vRq7Wj-v?J9i%aC3rP_ z85(q5VQy*YbnlkO#BUqp3tw~jWY5}r)C|<;9Wlu3Ut8)8K;_Q69~gA}u=|GFpXK9$I+vD4cMKCA{l~GxI_QSH{V`zQhq$YM#MNpLM+lNg!z#ADj>7)i=X4n4Cx-&&n z4Q3B2p#Jg(lB6OW6lcNT(4Vt2J|;T0Nd5R`Y-Q6!1b~SD{Yc6oI%pKf>e9-!8EWhB ztT4eS%0dcXMB80%cskRkzd!`b2mPHVy2O30*rYjNJl%TvB@txS6tnJO-w9*Y z+66=gI;MFfsjq$){qs;=5g31yUuMJmew=A*ik-k9_mxmaJh9$k15_JnssnE=Ui-bI zSQDUO0cEEi?HM`n1D+!iG4hC>PsW`4quHrGUViEp3PPeSDPz!Hf-t`dHmcYfLR zwBvlb4>w&vyv!`o+H2#7x%%}W;Uw2tbcKNO$t`&{I@b1A4aLxLPU-iTlw@|+pZTU( z%4JVEi{3dYb~6QwOm`I0xjm53#xJiJ*3YfbUlF@L4OL8ok3sk#MVm{2~85OcW9-ejj_v8MsI6%@ph{L=d|VHh_t_`>Fm3YidLk@EW= zQ9x(2PjQEwlMvjE7{b}6BTI8g(nhLCgFP$)u9We|Z23Hyp$k>FKJcHaZ>%sqRoUBS zco|R59x6KSWTj|kL6rDIT5#;sbZHH7^!EGDKZuXH|Nb7f1JUWvV%%%>jRPbb^=5Gl z7cOx)k!vAZNhA-wbclilOGWEAt!j~ui6&_4Cs+r0XlKO;#`X6x=7}T8F~>f!ci>qN z*7s3IAOQ^xBAV^ux!>PQ+HC68Sxxk3;dz7PSE+oq2QS;RUm*dcU6Xxp=TJvq4)rB( z)Rni75+4=uGCl(IEpB=SDfE?+Xy~*pK40>PLrOuJu%{U}qW1LLbxLbK$0%52*GhNT zFNimiw92q3NiIv8JHh9>w6e_|^{RgtiYgt9tQ!a(k*3$rL4E?&qG@N?TZWV;=ybJn7TkxCL{U0{ zVmh1YBQ4tc3s5!Eju_mI@6<*H(-^9Xbv*eXbCi%afGnf5=~)BM#VG>VA5{+R}xEY7~Q=w5ip`zd1kQD-YfnG1Z6bJ6@GJ zsi5uDw(3K-oX*+~Zv{`&z}w`-P`}%DEchylSp~Ev+aDGZ$`ThkarILo?@cojThLuV zrF3Kh_wM=U@Z<1%KOSpjGD=Ao7(SLI{>ckL{P|7HUH(V@fNGW~jos$l9Wd@F>TOc} zL)r2NgTSDxh3DnWh=31*8Z1!s%Y;N&RKV~JBpAP_b7`#xd=*0lDd3KPgs|GfO2I1Q zc50#)jCFY&S$QR60A{i;IZ;~DHe>Phk{Ye-wb#|Bg}kjEnzO@Vt=`Mt! zi2iy#B?k4kM7`pt@$8!(Cc!w<8eEgxZ`|;i*$r&GMuyQ(&%T^AgGVkbtHo%oK^~xG z9GxIslgtXWH%5QXf52DKgNqz8PJ?$XzdZxVJu?UWw&~|QRCvjub3{1^TSw)B&mEjj zDbhx~4(|Jz|Rs|6hQGnC};UZkPKsUzUCkB`EC>UNGCnu_}clN+SgY9(yJu(_r&F#MiC#o z)j@x*4ar^UXXHQra?!tmnE$$URa;BIEtZ8wcdaJo?NfWLzMQQW{5om}9t-p1lNC&< zA>WvZ;G0iX67-^CC%` zUcQnWGm-K{h1~}ln-Wm5O37lhN~b8|ZsXgB24Mzy0vto7M+knfz;alDHy12GNj#*T zkgrtFr1hhUSt#)5C;Kc!pCRF+*Y>hufIKe6J22|(DhJ1upv3Yas`VnO^LQC2`j~Ll zyf)EwnNF2Tcs%#>Rn<0~Sw^V0mqz`uwzUcAJd^CjGhp8`ng9gL?#WJJ0z8UcgD4w{ z{pB zEMh;)*}G@8OCH~w>B!P9vt(UQOiPDl4<5U-&Dh{8 zd)M+DcygU<|EL~Bp7ptwIOo$<&ziz#vf-Ew88{FQG}mCC0nV(yFLf*%rU=M5XMB*2 z?oJ@5?PyvK#Aod<%jW+Q9I|A4@_+=JiCbehg(5NXW_6$aUGHiK-&X-W^L?Wcr6mux9UET5P|(< zLbfJB>e)<_4fT1sE+iV`2w_2u2SEM<=n)-D#M9sFc>3)${@=`f6t-G!{ttz!C)CN>vaer~CUl)`b0Y`U#G zbqBKf#zSmn$?Q(sQb-NpX)t2r5_YQ0@M-{k99|(1D`|cpeuvRO@Irqu5KSCf0uMJ-om= z!4+>&zb=3ix1=lGL5b_Rj_=_VcV*oU?r{Ho){j5$Z{@_7xiNtU`Yrt~;`c{xUpVz# zjTxJUnOyK?S1n~Z#H9q>^{8st!*utJwAJ|L(uR3yq&}Oe_R$2da7ZwJ64M!S-hM&b zMm_rZ4E(&kTaV|o+7)m&z49p7N>QOmK|Uh+@ZjX&xyN;;@)Dd3-sWtK#kw!EY|Txw zI*3=Tl6gNczB6^LkQBC>g-O#S;-8C^HD`9vaW7!(S>;2|1U#RO^Hu>eu z_~=Dr^$6iZM)4ChhOokB2Br=ua~350)6g)|x#cLz5!hXvp=f%Sj&2{+z zGu(q2yN&%XvI@3LRl70GHc9(e*5TX72X99AuMB|spLSYq&@t|735YMhDIr9X$fs7H zimu;wPmd>6UO=lyY14jP-S|~by2GuAd|BYSYa~q~FTEgoTUmteRq@DQ0JQ zbC`T|hV+K6rzgO*Ypvdibql3$;@$V|HvB#99C< zqzI%|G;eVcpD9B7U>7j5_d>TkM|PKi+H5t+H>FTUpl{|9D_r|rIjX@qRRdq>LSQ_z zblv0JHl4I=*$?xw1DKI zSi{@l=%UHDpjaieoA`bcrrn!hbH@_6B0Dp-Wb9P%r|9-oUewV723$IbXH8l$Ok zdYnfjGgKe*Zt)dJbHhN|%9`+v@DW{Wso>frI$!OQZhVyF^CaY&SL8&+q|j#DK<9gQ z6uqn&lk`m=ur$|U5J8XA8w3*m5Nm!vR1*_+v3ljPhYTd}A7qDkbK+;iL8nlQjB|za zogZanE`!cfvUes{&PpW-HK=r@7628D#{~x;joBc#?y}P|yq*--+sUAn^~%|9ocW!! z{#wA}=bH%f(!m@k>1fR8%yK03=k#miyrs%IZ`iq4DeQNq6&rzrQH8{xugJ1(5d4#; zLeDW0iY*m^tM$1H!|@VUPXV$1=!)p(Tc)xyf*e5>+IGI;kgak~6}0oFEQ(;Fo4}WpTU-;NT?r^o zZXT_}kjg(&s~E(5LedQ44;U}g$30?20bSj-&N7A`->(}c!U3&8W=qmw5y^xntoiQo zyV;%zS3{kR_id()ub6&RuAQjUwN}KP;4b^i+yd#2MDegofvo5=g&FdJTx@mR%U7oD zH%?U80uKiJNayU&cOn%6j;=T?E6Ileo|ji995xCUYdTl*nhFz+KX(|BR|c5qJ(v&+ zN^hnkfh%;&=j439A9+Do6r4`#c&BjEt&@APHFLRD#fCrbY$9Rp^j&ZK_0Rm2Vgn*W3zc43pH6tW`B}HYO!+mJ;j_t zT10rJg4pbxv5B~7o!K631cA_LO_D@Z<5^JCfcm7EV0k`8?2pe=HgQX(3kSaY zy8{sl%juG@shpJ3BVc5J9M+CC-~+&5)B1YDF`tLy;Zv!7uW`UW5^BLqOMXD`AoN$W zCI&g{?`{((%=YX3JuvX==94&hcJyQZ28C?WD8QD6k>WI0kp0Y_&qIun{jxMXQCDcy z#jWe|*gTkP;j1dQv~Th8s*zv7ANPM=nI*6V(_W5C?S}v|CeeMHeQOAE-xLkBq*~!q zyIkRNEIZRg>ta`a?JzYW{r6=<&eb*V(XFQk)qXzvydWI38#4ubYSMgd7C((c*TsC8 zlxkh1y&U@Wu^H-mo!p_bbQs&tpzS_8vWmxq*8gCzWzNorZscAiBRCM12xv+-#dMU< z(e5$25xdzR1CMnEiG^EN)(g^pt!5()0C<^?fuT0802^SsrGO0IUBDuIb&lJOPfe-o zp3XyHTr{FNG)tQLH38eRVwS5;!N8jsTz;n~vGc8Qd>JT>2m6gerV0`~d>2 z{*2$~BmxmJkMCe&NGMorG@Y+)0(!Mg5n?Ef93pG5vUE8QVEjx_qwP>BkFx>JN>WTD z2~@7`6be#=^l39U1=4YP&(?jvxYeZ_+geHEO!pR5bYs1c_MTyc0y%!a+_^e=WDRYbd~BGJ`lY z&L^)~W@!HFG*%X1=f^P+4LD(_191H!@ke z7#w#?}L#Xu(B9tR5k z{IMp7)ISJ}`R-Et!|rAq$Hmj#)e;+*ptx^~0n4>O0m?-EKs z5J(39On8pqA%U|}H6e(Zl&#Ja;MFkT#MI~LV?8CZ7PV2@Yw)%#MPaT6$9b7hPqzE8 zU4@2eG9TVw`wM^dAjR(s%jQn$lrAOACh+~8o)O!aN`qO*dJ!Loo?F`UMw}RuW@A%> z`^)~Vfs7z+k0PH$882nWOpVTv?3kl@3AHqu>eUUpHieI_Z^|e{H1e+*(DuA96wGTQ zc6EBpTmJ4=X|8gZb;yOyAz#zlWmck-`9R{hn&HRlpR2uc6`l`J?}PvA4t~2 zLcz_n)AFwuoV>m+Ue3}rg%oYO{{PMZ9sw_k0A5_kMlCD@>%SH z`*y#ZAztLIUzv1mTf@;9B^2|lmltp&snD4H>*?t^sc`%&Zxz|{nWPMs9J`=&s^uPy z`fR^ilkH1Ip9S{&k>iYxn*i;kTaxcFEK@&j+9ZTlexB+av&9bUbH9&kT24H9w#Oi{ zc9?Ai4&uIu?FeXYsPqi~J6f&}B43nE;iV|LP@o!M1n8D&``oxjanlebJR1~Q&vKY;qoy6WyR#;9l!2zYzeo&{?_Z5oz(uYNhw`FtT+TzJw>nE8YSFaQ4H(vS zjSni!sKl>LXlG8q=(cgh8eH(DSjDho^G}Gj>E7yAYy6t*&{MEphayyt{kf?NbwcenZH& zY>k_pUl$tt|D{WKYOtFNA|rA#D7xDTdkig|I4k!(yo=XM4AG{IXlOruuo1&RJ|J80 zRav`NfZDDuO61+hwBJU_l(I@&=hdT6m#+|f8Q?unJ5$39a4KGG+8z0C#;K6GZ@DPs{xlNsk@M_&@6RJj&kBO=CpD} z=HjqNYx!bqCR>j3ce%JWl|#8Gv{a;MUiXkx1J9RcYJ3{Rj7RnR(Y3oJ@d4 z6~v>mHXQWPO#4K-$& z=&z-qMfI-L4Rvcjog6Gk@-;ZC!6tIP(B9sTGwvEL4Gc_Wc70H8HA(c%dgDP*+y$UpqL}^#?$bDtFVd1JANtd10o%kUIoCvDB7oVpr;3Q#M9WQ&;C{fJB#slJCtcX9#?sYq z^=ZebP)6<2w0$h;>*qSiltkkAxy3QpfwswURNO{-d!g5a$^4(~7Zk5tD2yLC{=hGT z6q(XQO4IvuEQlWGP@tN(#cv=hLQtaQo71V0efTY$&>=osZ5c6je!vWrHij(VUb=Sw z95E%U&cWu6=zs=Jzy$~949X|0UJFCRW>7 zubL(>o@iPj-Cp0Vr*Yj75GkOU?kEyj>R4*>v~OeVpBfzw^oEC+R|p2kq0NGjB(@!t zmRp;2+Ygv}NcYSrh;|f>DauY11&^(NEL-kBw5oAz8Nnx*1Lv>1N{l-x&Te#3bgeOp z{qN)KAWng56KtnKS>Oqn^GU8rpvMU`U4l^QzsJNi=N7#yIS}*}$g_G%`44NQDpiUhj@Mfp3x{qGw;(=x6&ONJ;Ao(k z&FnR;?|A;+Or= z6Y0$)E%tpc->f!ie+Nir8G96)9@OW+LXNZ(VV!-xCzVegDO3=ctoZp3m|$-7K}O-- z{FxjBt`8xr#m4;SDri7r_B1br6&*eNu1DT3t+y2n-6VIEorAFlk_UHICHMG;TQopWijbJ%AXB>&J!zUe80CB z%+^1%;%+}upJJdXn<_Ya+;h8sW2}r*3@0G+!^Z#cW4KtxLRjl^syV4>vpKDVicq+$XfcuOL!J5m!E zA)O=Rz83dz(Xv!?wjPq`m#(X~N-@7Xrm((0pxpW!nin-ybO6+*2h7Y6^xRDMJ}-jN zWo3VwWT<>ju39Xvrh9>DXytV$8YPcz8Lc#?4eun}Y6Irvx934-2Mq?Kor55D=lY)sJ}sD9 zq3b2PR}40(?1#c;IB)jU(i?Wd%{$bC%iCtUKh2H35KW1Qx|-KVjm0ROC}1>wp9WQq zvUJ0W|3grehP2T|)r}I@S}d=3d|&FQbA68+4S|YTG4AR&ggcsuA$g@b(eH8l!JaRQ z7O7~m#IUVXd6iUFJI(L>x^WTDk6Y*6JJOHaWq%8214llUUl*MU^Z_vT_~|rsQE58l z%f)5q9MMuNSY9mVZ5@(Nz|Br?RIK1`d-6%e{F>+qAaIb@3xuU^I(KfrAM|T62NK3} z!pNe4v`X^FAIis*qR(0_!uFha!O(9eZH~wv+&uwoz65aR;4VKdy<0~k5dl*Cpco(i zQ{+iDZ3e7|=;N^~CYMw#M_WbBD6=!6?Pd(>Y`NNl^2QS(~9-aEq zjTF_P%UXBVq8#G9Fv*kL(KatOyHaAiYh8*@fW zfTFsZ|BS?u10431?kkzI=OqWJojUc!>5kw{a8j2&FcqQr@5B!{8Vh$}bQh1RPsomS za10BPwC>+Z&6_k__tjfDa5uv%&!$XD%zHjY(&6s&(&X$CSOPXzPQ=?jRu85t@Ly=96&o#7P^Sb8|G2g^U-0qi^ z0Bz$Hth^LYEQQr~V02f0!gD8c;0ehxo$L_-(N;J+EDOH)|LA(ps3x>6YBYo{y@T{7 zBE3j&(u;tK(wkIKTIekh^w6YtktU*|q97o>B=p`ydJ7PGPiP5&eBs`E-uvEn;GCV>+iky&=$5Y;+D!QDzK4i9eZ(!dqn4zv7Lo&IeS@{=Ud8&}itrA2TseEj zoHOBIKjE>|TWicYV7xXy6#DWK?3XT+M{(;*n~Oyzp&F>dujn-Ke9$0m9;{-^;s;nN2N64!54(9D8-$K>#t6$ANVNmkL5Qy?Fr-mhxV}09 z+MUJ)kt(cajNJ&51`E)s4$8JAtr1-;8Wr>PKa!p>Q6c=UF7VR**gke*)Blcd!XKce z*O>h4X9Wx8@@%6WeXd6frCrsB50QDb>_LVH&nt~Yx7fjs)H~dM-jVrhnErQCxk(<* zf3XE{HSqH!l}Sw>f;qy)VG$&sJjNx(R711LedI8=zasQw>K2lj!P3(4pn>J1kj)-a ziNf9urr|@ls*c$zc%G(M^17J8`90FZwQKuJ|4Jk2W+IFX~gSJuf_7KWF!s7J9q@;Ch z6Bvw?Zv0B~%<&FO33&cTC>-g;1m0+V+$5Uh-g9-p zV9t9+xzIZ}-sl5}YBa2-O~t5LrOA9$?$lAS75t3Dx`MyEgz$Di2Jy9=JOvJM!6?<&etX!YFX8Fk z=aD}-n&SI-`P+%v$lR9{wrQ_c?D9*|b9V({w=T(tj){KMu;)*??#V10ok7g}4&^7@ zoQn6p5ID&ycO=sdNv>$%YYW_^a>h}StN2?U5nWZjN)6%j;{ALA!BRPj(x(o)GA8=f zJT`%PJs*_x5L?aq^pVVK-IooT*XeT7O$1+q)N~JO2O`N?dr!&LHjB~5b^X(qinn1l z{*uzvL;eMCkWY<9zA9#FAWid572U;a*{ec~H}1&S&>-0O3~{5SmPTc{R=H)zAh z;%t85p6s)9{=)>9iKNTS(d(_{eGjW`8tf<7A>6H1&j`+-VT4wbqKZ@<3Igux3fw7j zb)uYOFo~`K&BY)tI)sn%v?_3K2W#GVmRM%0?S4`Pxu8cHluGh&0yQ4;V2S*JwvHmc zQKP6ZE)HRp{@{+xKJ%q#o{8bQAYrw{HgZ#B6Niv^s{Nt`H@J}v;v;kULV2Uib9(oR z9{-$;MBA=6Gd|>@A~tD!l-7U{jH4ka1;+XE$fOZJRzj&UemLZRTe)b`K(4RIdrL$x zUZ!II%N)t=*BL=Bp<>Zxa0s_Z`cK2NFlJrI8vJh?wYkLGEiMHx zU)nPZel++b7}-9N;hPR3T$!`H1jx~3=Z)S8Cus;2=7U}p6ZDwtj8aRGDu#D_&X%Y6RFX`1IG zpT-ze+0Sjm-kZommm@Ft7qOkm$s4${*^M>O(Hs4u1&q9KbyLoxFo|mAhKGc;9 zijtUFaEx>XXKuPAuM7@j{I6#k+@8x+H}J$@hMu=rr3WVlPlkU;gu4z z{PYzE`Nw?x>Da8P+&<-K_5|AWPBWVwvGet(FQGV_3P(E8q1~jO-Mrn{=l-)?1*!vT z;}e2u0xvG!ZSMxu%@E~TV4n$i)cjwwV}b&(SORfGMOPY@4MZ#s?GkcPM$4_^W*HUbo=)ni0^&hA!i9Ba|DC6k+^3_~F1ndZTxGK~rgg>f*jRkG*C=#HE zfOOWJ#Vv&5d48ykqDY`doRjvv%0?G!SM1&#OVmQrQ!XkNXuEJEGP8J3;i#3*`%~vZ z$Gtvc-S(vwPe;Fw?#uC{%hMWMP2u|fz~VP~zBmj~nfj&TW81}!%$e1M*#@qn-^s$= zr3bF#gtv@3)_qs>^>5D2_VSQg6GFwy)?UN`fU7-QAlwxb)&=QAHv;VX$2Ht+`D9!> ze+_902h+U$ww%_@MsdsZQS|8aPhY<%k|5>>2}%my)*tg9zvyP}bg^IuB9|@d_6fgZ zZ@3NVrM|+aFnuqPyn~{Q4n`T1Sqla79p>5D&nY{GR^=1Thi%PkbCck^n^tH2Nk9<- zVS<+|Ka^H3=`WMy^#BC)ll@hT4g4=}8g|#S%U6rbs+V|igZt&k+G&J2i0B;Og;H0$ zEQ`XFEgqOTmQ?XPu24h7`t)`Y)iqf}+(GmwVfj)XTzx1$6hTafz!WJKJ5^5~TmcxZ zT%kk?cNHZ$?yk+f)K3iPGxxCF3Wvw)@b`2S==oRdwf(-{L+Hl37<}1q$NW*aqvgDG z=;=?%ZN1fv(T~a_9VW-D;w3+q)7$3cxw$bdXvNdbr;ONH|!|b57UmCD(Rv z9bb4=_C(`+Ls2jr`4HMMlAYUAUi_|4gy&Vl;NP=j)n}UjP)^idSEjM$$MpFZQ=)@6 z>OxH%$~L!D)JBB^U!AtPiPL4KCb193i$?En8`%P5aq&sVS_4zl=LMep&6@70SH7b| zy?=8DqZGAQca{IWD$258C!}b;^)cU|y$WpNJHi(>OFccDjm2A1-A~H07Msp-xv(Kc zr2AmZs$U4rFwHftxdMmp`e*^?TasgDyE&2jGSm$MzAuVQTHaiX=g>vPg5iK8r15+~ z`b>SqhhcR;f4p$hY|3DHwttogfs6KnQs>jF-B$Hatfc>moyuf;Yu%K6!$=e%06RE?>2 z(4^9!WsAPl&K_P^ZT&Kj``&zODtq|NY?ObA&InK37>8V2vXJv!Z2=3a{hcQ@`|{7Q zZi#KysMeU&&4Go-FNnq1^fahUzACz6%OsA!g(smms79oS_6Q}nbW=h{+QJE{(k6?0 z#dr9~5{4FoR$rAD6CHaNyvZ+;k4&o7zapmcOg%KdG7ty*yy8w-EV5{HAh^}YFFl&` z)BzW+`bH4dcf1?v%qhfzNoez*>j2AbKHI5N3tyyn>=e~Tl#^X-S)o{1vNl;bnkLXq zpHav;!W=X(9xEU{tn5NJ|5w80#R#weJJqC%0%oyony@LA6qBDe4od^QxP+g@Lq0-! z&9WB}5!LZUn|eHJ`OBQ?ip84rZu#-g-*j>4Lv0jb=bdPAPrm$B9EQ5yH>p*}UIMp% zoLcs*^S_;tvO5GIrbioinE~af*q=!=;w}vFw${p2pvK!m4|vf$%{BW;UAc!Be|5)m z_tz!XWsWn|@Yc_j`VlKxafjf;aOFn>DQ0R?ZKuseUrx-+EA4Iqz5!@?+xOmCeA1N> zlhjv*Zy^R8{4|MO&%{i62*cOO-76j~c5AQqb=x96_kiMU(wp<^TqeBruSP8DSO-d2 zAD?&D;z1!6=Qy)Ip|)GFW$?BPs}2#IrHl~mRTO$R_RrP9s##Qq5XW{gzF!z@OF2+6 zaM?cr%fPnWrM@ILdS6Bw63g+6BPOD`MAKw1WJjbT zb54@(vOvTgxwZ_Yc;CiWx6lmjUp8oe4+o$@y7%?crGOVakl4$mvWvSax+#C0iT(-! zgK$UP2akxz+A2Ng-Q|@a#ZTO}HpqclFCg#gr2WQ;#Dvl0)#R`Rif?7%x=9qY1%Ax` z3|E`FspammCC1$r($(W*GI(SmVLQ)70i#)Po=IZrq3SF10a5G0TaJSqc%dny?Sd!D z10$+s4EPpc^Jo3*9?F6K7b~~&GOCWng}qfq%oP4m1ZBNn796vX45|O2SYbAGVvQMI zO#mj62l;M&srhNDaG4&iePC*FL9LdlegNKc@+9Ng3g#FaHjpQ+Ef>ggp8ylnb2vud z0+5h9Z5d+*9y|Fzhv8xHy|b2qtKaoXYe44inN8TVIY-FT5|kRA^yrQ|^TD0%y0= zlc=1QxgG5@$;(cm>0UbS2}_L=X&d@u=dhelLrA=rL}y^&zaWkuJ-j(!U>5i_JRi^5LDt3Q)bWC%tElE`^`oChLyM)qMTPH1U)Rsok1wXo z_V{%*H{HSiHX8PQi&XR;b^?rKZ2ry)y$5xx*cCI?nNU_mr3Z zLLL7ja5&3>`@H`u8|*E{6+71l;;#B7TF;lV6cUVnyzH-b>Gey!VlfgH&-meFHF@8p zra+97zBqrt+cFfk4nF_||$^&*(r;;UKtzE6j1Hwx+vm}_@6R36hgbQe#T z#f2kPmr<%hG^!_KRG6FcB-00-c8%gZ`;E3MfAAPF$Kwiqc#Jh7LawE5My%#zYcm9!Z-LS64RdHWdW$wh{{RzO+ z><2Q}sB=8SS0wam(&1{7Jnx%XVEFv}QV_B#6}*SIp}+LsC|^f6y5JI_ z7l}eXVH5c#M>iLvz49Ho#i!vnrzl7Vb3ubUqA|UO(_B01xcDTg;InFZn`JP!wDgyc zm3SVtkOOv=#p2Ud@G|(kU-f*UX7#4dYdw&&z$neK3vKN>;4uBg*#6=I>^9zi_ifS| zH@+rNLzi%4N;0{x#<^9cHiuajRIclMF6)@UyY-Q~FN@ysZw!gn=K#RRH;Ge3`g%M0 z=2PwHz?w^2cze)GbJXI3^fDBs3`O-&;qc5_uhY<`*LCwLQ8b)lOo z-{PbkiYXUTNAj!tAzPO(TSaakg4W!EFFv9-8@F!kmHH2{BZoiwUaj#q?N&6KY;m73 zs2?*3ZG2Zu%`Xp7U0g5P5G_o|>{hz;jhlp)R_^Nge@1KOev9(@`V;f~@AML$)TZQAnCuv6FD<4@gf+ptx~DF`Tly@g$`{`=_P~PPCiQ%9(7-DrCx8v7JX)q+;ER{ za}A$ezVRh)d%CYwIq*jZk0Im2kPC6n9VZ;*+8<)!14Y}41M{!L#$*&?Nf8$}*xa|5?;J zz?&)7Abxu%Wt5F?DAk#fXa%F*3RYi{Nl{qtiJG7j&aE+w9-b+z5VePg)E&|}iX%#~ zWgHUAi4q&PN6z278p5lM9P{eal`}I-B-Yc{Um?a2KH}2u3kwkkjhp5uGjV+`a}Lyh zSZOWISe}OuBEQ)_Zi&1npOP52zKLM+9p#}3HK?LZ(xf&#Fdn+kjNw}ZnJ2wyDRz(D zk0*v6#Yp`4bt@zWx5Ahn2T{cnQ|MYKA7TXyXZt?ANcucn!nDO!MPDd#Q_8sZ-$zoi~Z zK(WXHUlPPh!Fq5Ts!-(Po}Qrk?X}NCeIkv77WMZ{a&_DaC`jm~P6!G!cGOSJ)N7?( zIH+Q0S84nbQ!DL6jww9-xz8oNf%>^T+fBzoKc>I={^daY1#$QSqnsuuvTP*etnVGYY_{*}MID?3=940~Su1S!^=Ck{#w+ra$|xins4^ zZBX*N4ZH@HS}3w#$h0VI=i9SAXZsGvqx8CD*DeR7clv@)o#qf_8KfWXn@3m9I^Rr@ z^unWDv@PxsR@4QGzP#Wiz?O2HRegYf%QE@bAV1PVg3iWg2vMhnhK@jN@V$1Uvn#v1 zhk};FzI9p-)zoRvSs2@Fh2`GvRzoHMVkEZ=tVv$3z+83B{8-@(V{c?WWPcuF4X<{3 zkxB=5?>muj8p|xsj+hhMP66EJVC|vGg&W7(=U!xEi8GkVGRt@@omjXDS6m{WDa1*^ z2qB^un+0%ZO^xaT!hC1<)4HlSqOftwi-QOR07dxE@l9fN5UYa+_;=|+I$3GHZf+&> z#J0Vp%ZXgMpa7;RrY~^@4i`>ePX$YU8B+J6lO*Wb<7-!rAyV*Asjbnr8QSe(J96-Q za4B|6vX`ccNIXwQpKsWMsM3Luw*UEpYpK`IHfDYD8NU69!|>&1zC;$us|*9)Y@ zdDY;(_r$Sc(huc~#*4xL+DW3_<}&Bs4@i{7$aYb&TaHtqdW&&-+2(mkJ6ZN!$4y5&)oXt|N+ zUZjY|KSWzs{Zo)&q`bL52&KF%)tMM$b@{hKAT(hM_lsb=b!HZ zxYKk2i3iMu3U}*b_<7g7&&I{w7h0x?E-GhOXJU}dX-UodY~tikx_rqb>&RurDPW_I z<`T4n03rs1`p5=5hq<8y>Z}}|3Q^afJ%GB9Zay(vMor46ZOhw^U$i|9>_(VKNpmfQ zv=BI;-P+rUF;#-EU@mUg!lBArLM%uP1fdF0dv(K-9OPV~tzc4zD0-8>aEt;Zr8Y*1 zHqbuOE;tDK?Z49#`9VVASf6`bJ{{zv#|Mn3Fyu=KZSPaVdtc*xO`;dY=Bg~qVw53y z%-pP|a}>Uv(UQ^bC+iAj+po1mF(ryIyQvt{C&?q@ZUU@?2!8)f{LqqJ`r5RpnJhU#^B@WSR!9;VY?GR-MAp|5*g|tSmiM#cQ!~Qq&Hw@1Rv0!e_GYjcBf> zSyj{#b?sgeF^NNA7WdBaU#nbIEv4#ou*EnL+8u%PKKaN>AkXslvWwm+*2GgpZuWB$ z;(4N1*L}{>LwDIO2{{N)nC;^G@_BE22LFUi;OTRw!Vh;yL}+4IB+?1PWpeBswT*1< z6I}1o%aL?s^jXpj!e&6DXo;oI%~S4nDA8RXaL6f0|Cdw$c%=9f{ccFC8b${FgV85nP1ZN0s7M!_jdOdH>gak@6U5G$v4--g>fBr;Z9qs?PIjvI6?~I9EJb zSlj@BcHRY}mPE~fy>|#4gF$hupCUNKr|wSGg}N<=ev>*N-4RD=cYj3IxaZ|FhbA{_ zq3@~N#EE)`;O(t$7*mR;t9Iz#sy!gT_1t&zBZu5ynT$QgB#|a1#~u+w|37nV9<54AV401eS${b^Ee%-us#)dxY+rs|kx|UqN0; z_2pF>TCm|p(~S@?WLKmKsCWf@Ar9iFC3wNQ`%_l!P^wa_R`2Rhbi)(a2jiYkI)DbL zX&*AL+_%<9RVE#D-1%hz{i$H5nyKWhY+0Tljg4F*%)L_1iMAC!N?>aQ56$$domFMt zXib+d3(lG&iqS*vCPg8u37NgzQ4z#%=Ji0rw!Jk2tN+(}!Oa z?A;fSCoGKJ1cRds=JSh}}J*9hf z`_qFM#b|3JpNukIx9PLQ&%Vb`ISFG#5dF{hg=DQIA4DQJ;bB(mps%M>dUTu}+Z!*~ z&oMti%fKYe1DtN56YtTOj~aVna5H=G970F|lkD1*uO(ps5mmXH)4{^fW@L5&(oUmm^2!rLS=P1#IgpowCsj42dOu#Dwh%GxL4Es&|2Kxh) z+lJm@WLG1z`}QNBp}uN{ME)>tr0H4&I;F4L5U^db;I%=fLRU zo9@}pFf?pD)Q7pDu?{tm1v?PrIe7J;o+Cw(Hh~@3xbEFSV!hY8JJ*LJ2vp8~I3nk3 zVRC6g5GkkO&DpQ6=!N{8Yj2%yDN@kkrrmfFm{=P8Z#bmhk@B*nKX9?I%)o0*dY+9} zxGzq&E#m{l5_hN>>Tv&;^%Ln*4LX81Sr$)@E6knh?or*bc1jiRWXdruK-t-tG8ULsm9>?AOvR7huXRQHuVqv$Pgag z=I;KvO^nG}AW}J3!yi@%G136F$H3UP7>`_rgrc0=o}D%5uwkgC-3N5eEWChS?<)7} z{?_%q^DJ1|5T_x+gFJ$OojLOaTi1v!Sn6g z0%2K^UcDp-3M%RetFlz&poC@QOcT`>BA@1qS>4T2NTVLKW2yjzxj7>ljtI~%Kd!eiy%y5^p4#96yIyV zE{lBpb8<$&v0!kX6}9d)L?JNpq9jqL3*w_(20HGpNL0`hy)9-Mwm=M%=+Pm-$~kr1 zQ<=Xa=u=n#>JpS&)&;V#j{h;o-x`<~e7S*%l3mc6)F9Q}inFUbu&8d#|L#kAB-@C| zZAXRlHE^BOQ14xt_6N$9L}VF1;@3VkY{(RdYB)mfumR80iF}lseMc$|E$ZYrqSy zd%vdKA$I+^WuV%T4m5L(_%~1)wWA)xzO?tr0MO!GN!U_zgd- z(wwC+#ExfLL+e#eQXCjU6mXZuHc z7VO{UjXxUrvclUnY$%u^S9#cQgryjOlR7{1oc}2x*~2;5k~~U!G+2+mLY=p4s+KQ* z%avISOh1>SAv`OryS3NbT{;gZ%isC^n}+|ZI9J^`5iFJPLb*S1b_!q1kMsg8-3{Zn`0cr|+6iRojVP8fK z+tv{M4N?IbSCFWd%SLzjbB&K(6{T5eb|D$iBt_sq_Su?s95H=0z% zQ)*9@>mk#yKI4($XX)#qwG*sg^Y(Bx* z4(`f&QNeqMEu3k+c89nPmCNJKzCgEx3Cvv7KId-TP0Zs{dU*gheVdsHBf2=3r=C%T zgZ^r2@<`j=-!b=}c>O@ROOSk~^98V{d7r=S)-nwnj@e#&Prpo|ij!W7-oHURD^e5l zRj2S|KjhZ-BPaqcn#Kui<$JsNq{Utj>CC@L={qg`Qy}v#vIp6rP zY$fnZ*vEP8A^N^LIuMUDRxs1A$$i;t>N_VuDL)%o&|dw3|DW2=xf3sRX9PIR02f)z zpHu!2aR)CoT{e!y`pmUQJ;Pt^XXAIQM$ap}EE`FfhigJB#`N*Ql5f#_M>!*K9@v1q zRO~Y;&2@+j4ScA+M0dsLuXkd^XT#YGs^D1rNQM`ivVvPEm;2tDzu)!su}nIS@I$!6 z>U1E?@;r--TDwayf|##Qrcb3>`QkgHpFsImxx>}Y%iNEpn4e^v4=i{<mCT&BRzjv zzH5JOQ<+DN_3EV*&%ve)9-VDt8aHOP$_5mYk1n3kU48S}dL;L{lHQ=@_pjGJoplIq zq=o2)3!&H-S0h>nUG6CdjCaoeo)Y*6g>x1pBnpV_ovFF>V2FA}bb&VSMif>|lZQ`4 z&6#o!HQWCnlEddbVyTz{MirR&(F<<&9>kTkFExUtSa&tA@l6!UQc(m>El*|!6QiUl zr`4+X-!@Ti0*=@zkZ(*_{^4&$4v3W7#@MbrQ*mAS0sd_;g=)pupp8=aN6G{tma$b^ z>NDlvFud!rg&`Ys0DvQVN_GcgBjyb_YK~>CHNVws^B^hJwM9s4X8dOZj<8bjAN7+G znnfPS=`&#?$g+u(qf{9kx9c56Jd+X%t~I0kAO-BnH%wk2G6(2OdG`!glI4oz;h`F%fp02MeDH;xk$vJcWH&LRfiW%$Xu#j4 z%E9>FZ?N(VsCa&^yUqvTJOselVotAx21v43Y4Ns4z6RaBD2A@r7dxPO62(Bu&DRwUXP9gM*m(q_^K!(!sZvI1%Mlve>P>* z&z;_qOYxILlFcU{pwtje*lTS1 z7R8E@p3Me|i4>zI@^N97*;ude`#fV<&68BMhs)?<>Yiai%%2?}y|&tW?9)Laesr`T z{23aLwV~}CX^b5@4?M%OZ>Q*ACUOZzMi?^nWNIm$R$aE!NVLQV4Z`mNaTO z2ZI$4W&eF&RMA`BX`db6`_?ejAumNT@@2%YRuC3-8~nUcs_LFPg@n-4gT0qVuj;N@ z1$^JPBwswRr|S0#uKu=r8IJ?>I2Fm=nyXzr@X8OyqxiuEDn~HF@?pL!L}`egZH@7B z!U5A2q$v_ze7Ue`E;Ze}bXvh=5}gR{HeMAvKDOLSVzCc-ewtZn@=uKGg{Zj6(jE17 zbEf5lqjM@aNPb;HkRr*OJu8552`8(5> z@GPjx226~Q*gkr%)c8r!p&Jy&~giltuFuxmw1i#l1fsL@k-CLRg!X{L!DjQp57)Z_C54d-(4Tp7XI1Dh4eZwG3 z3zuXf!^s9SJ?p6pP1!Bm$ez#~$_J){iFor7PWJEPde}E}&2E2#R=JQ$|8li;;j^rI zL`M9{?kT%mtxv=#hm3ogWA`c_hSVcJhV_K90?h328^8}v+EpfCa+<)Sk*r<~@Oome z292nEK$zh%JN*aee!AmCdmJ?8sRH!xah#y<&?58BShJ_&2arFje5Jc$Gm)<3Yfg3+ z?rKg?3y6x%*hj6P^um5pS)m!zuG|uTRFz(9IZbm)@%d_fjnT_Ue?!NAR}GPF{y^N& zW_vfjLHNRK$qQw+(cH^NaQ)<1(f-GPkES1mKPj&2>KK&%k-9)^yF zo7=VsIHYtLInNU73i!y0%l$H2bRtxg+NW9QJ*PU7Sm%qoC&VMKN4@(t?1m;NA79}4 zD5LXtp1V!Ks}}W#{2am$$#dCrEU1h+lKbWS#d=|TrH{X(PL=R);iy8FloO0NV`l|y zXApQKle0hIA}gjF582_qfRUQJoGvP*1z1SeyqvF5B%_~!d5fI*_G%Yzek2S01ut0! zlfqBgbzeUwp$B=+d8NqwN@6MN*{uTnn7^NR89}rVb<d zk2|MoVN_-QTc7_^fBPG5fV8J?%?YZ_YS+VX^$Q{ga9nWFo0QUgjG9UFB)ID1 z`o(?nHS)#tXL7Q*?heZHWYMm#qcbc)v8`R_=^vDvWjEyi?cjYO8WE2cgpES`r=K%) zvyA&$Y|IubK|`$sdA`t=>R}hKKb)rU_y5#V;b~hokx71idd&l7=0rAx+A+8QK(gRf z^S%Lx?{V0Vm*Itvo?Z8iWA#lQc{*#+%mr07UgoCezG7tPB=bE~JU+WY@S^)RS}#AY zlM^(<;%7=F>Cw^`Q$M868%Lzipu=(t2TFmbcF>98W_5So)}R7oN2Q0oG|In6Msib<@2y0mUCn@g>0T z^*guVEfoSXLRvpDm61}a#!%~jF)5DT`H0?M>)w#w;A2$}(d{l!TvqM41 zd?-4K)^F-Q5v_|09bs832g?WP4^i_@#BeiH)j%ZP zkG#UX9TeM#Eeb!d5{#Q?YEfZwoWJDm+*BY(aMBJoQjEc05Xb`WE6AJk^Ibh(Frb7- zUn#Qo^5lc-dBM@_yDr@6syz47fLn|p#14x=0mNlt!OJ#Z1dt9(cU9jn1+PWO=h|eK zs}e{k*d4>)&{RZf4Om)FlO%>E)th$!|0qeV28F%k-nR)hMfG)j3TP$1E!F}U`%`;D zQ{e#2*=lt!3Uy;@SP9_pzAUZ`&mStA6crX>E*tN9IzN+CgUHwLW+R2_bdR&dQ*TsZXX8$&xYpWTM^(~597cXPH?322 zG3pnzTfBU(r0#*g;|Il~ZPvZ4;e0*}@hPdH`6$aV)t%5lHk39}FLFbQej+R1m2`^f zIQMn_;SmJ+)Q>_gkoR!#!EASG{^bVt=PG~INuoqsLx->^D_I$AvV6vPsJ7yltJ5*w z1pF@zGqCi9%*Upz%f#iKwpBHSm|}+5XET!(YWixRu17nj`)=&>wlNc-YKd~S&&iJ@ zisd&bwx$AIhbR&g=hMi|sx>M(O4}fUR{unIb?gr8l?Gl+P!5@A{--h%ZC!OxBE+nEV#fBw7Q7vois8iidGD!%9vGA{@BPQIZ>umx0eBxeIA z{OQ&x-o^KcL!gkq?ySb;@uNT|$)ulQStqq3l?YXL<41{dirTxFP0I>)i+z%AMOv7p z0Dd6Pup`sVISoY;q2odhC)%S(Aj!XLm=64*SWH8cnSVTw*!;vFQX=Q`r~lG(zU92z z(DC=m##0Sx>OIQeCt*4Hg0LR)xmo!Fi7`z*+%w-J)S9XnV-TXpy35CU+lC3*@tpjxwBxE3F|5aIXt4tw($p?o zj^%^4!Gqsg#?bulR>H8!JYS?w@-8+OyY;dD8(?SCN8@sMcx(kxxqSQ0l6LJTG^LU)wJbwiWx2z)O0vh{4KiH$FaeUX!sEI3Eqx}Jx)6s} z#d{4Oo`Tb0{twgqT@23*2k@S`QW^82zbyy;yp=EK;`Bh|(cpCrU-EV;FQZ4YDcK@Y z=AKcT3oQEcflRo1zj9vDA4X>b=wZ4hx+1!Y5cR;vf8kdURpewq6BLK_yn9QXk}Kcbx8!qnHNb}Y^w1&0|1wR(ODr5qkCLiyCP zZd0|<5!ftp(j0NQo+(bIy4|yajY!NBNuImy!=GQ%wi9wPk530s|IsVPB7`gwKl>IF z9Pukv*5x!P8y9Z4r)mK`E#wxk%roQ-Apf~{1d)H*YZgQal%5=kS?ZoVX=WDizYr1B z?_;n`y|A9uERe9Ee{KA!nQwWGbS2=gsJxLQK7K`l2&}p?MoUs}&-Z69jtMDT<=G=T z4TgS)K)Afy`M70x@T7X^ZL|pm$q;v2IeLxl@>jCFk+VG>xA->)7ZS}FT_-M3-~?%J zk^ac)1X&ythsUA9m8_n$%$i0g+`q`IvGpRiJUQO08Lj9uwL#VnG}l(~z)ezo|DQPa zuCdfi;?)y`CM1#$&KLAvu$1?F_u7#jS*f4@l!3>j{Tz}=? zp1CDCcx!DL^K)4VeIs!B3_5N8z{aavej9Nhk3KelEN4q-kdg|>w*emr{fgLatDhl( z+bCY!#{&!JKN~_{M&@X+p62}%)|725TbZf4_+G-9cM2c%JeOmyHRT>~ zZ&4)hCNL$46o~F~q*Jb#Zd?}W(jbLV7QUl{x#NeYxeB_#Si3I4M_+J${Y%0uOsT)6buc`y~zGmTy8F`j|`ac)8(85`p7GzW()~9AFY_YP_F^D}!YH z+AH6I>hPpyv&F>Kn%Csy`g_v*WQ;f&*`?BepKkP@;7PYX(*V9e1a0|a5q5`R({BO# zN%*QW(9g@`Q!W7dNsN}mC`H>+djGLwvJ}c)X^I+ui(ceEIG$}qAN|05WymO$s~)f$_EtK6!B8Pt67kY4hT@o+vOR|J4SK-Nxxx_Fw|8R}&3u&+J5LwF2tv#$+0f zJNk8A$5l}QSx4HK%6r)Da@+WGPh@OBce`x~o{ZmDn0@JtDyds_#^vcn>UJgEHMuPn zEZPD*m?&kQ0PP$bw-1%{J)d$ur-!fU5AdD+a$W(|NvV2z4*S^wPn?=^>;>d8WMzq4 z!v3Z!48Nii95*Dw9 zMy5>u18Q=7C%TYAzF>Y-{5 zs=r(3953hHTZ2TR$JN)~uyF&V3TQU|8VPu-C(_i-&y6%2lkb@Wks~J_UDIV++nxFm z`?crd-{TA4I&?`o2}w~Q?O`e=I2MtFvK~IuXIx7WKI1YKi+*p&>In5EDw|tILrmrL zM@E1|AgKjj%GcDwQJY1js`u2+t=>T{lx~&6CfQ5~IC{I?o#X^`ZE|}9M&)$-ga)LT zr!No4j&|pLZ*h1tn3o*Z5~lYz@qPLIST^HO%woH_h^44YEzUVaPL-oQ?1YLQnjxVG z@A>%7QhjiJM6D@B$ZKA27vpUY6~`YT?iKwHFUIy-I2$I(@5QAGxmGJBK?$R-@$@)|8bSfmY_(hJXh;* zA5+0M7D6s5X2F_CAEFY70fBo`46f8z+N`pN=*=D*UFalD&p%;lq$3S7ZUw};d7!ru zZbTP~x5Nw`DFs6n$$+!;m9@%jw=H&fL8BTLLY4iW!w5!*pf}|Pg*hKN5|P0EhIv|l zU{_>q{{L|G-O+65;onJ6REwgqN6o5L6tN<;M{BgVMa`l`?Y$$YDq`2F*(%zrHEYkd zh*&YI_6%aj4zb?2_jiBqd)|{ja*}g$@<*P0zt8h|KEtFv&e2w=JTzO2*sxRY;(5OO z-r`(ju{+?Ctw&%b_M+^=V-`%=$p*on|Q2tuQ zKWP$)+-7Z4AE~A8soV$^AbDum9_M_ZM%+j*GlJ#Dtpj1rq*Gt{BhZir{qcNEO2m>3B zK{idYFcW7?XUs@aFxzBDTE&N@_wG=7Zib3qIZVr+;T4TXs>u3S6ZgE$Ji<&4x#qo|S&G z>(7vBE$6tN%xFye^EzL3&y4b`pGxPMAlLHEr7hv*kP2wHu&9CZ0Q}wp&>cwZ#U%w6 zu8*k`B)9}-FFU73Tm~iKSwF)nyKOA8BfayvuldCrfq>~CNI*AD_iO`ZqhQ^Cs3zol zDSr82{c`1XBc42_!0xl1NlD~wXMi|^q2Xe|kCap0=kvmMvv2L5Z9Rc4SRzwl;Wh>N zk4)bEu^2_XB*no+?7s9b09cuS9FgX|VjA+X&}K7~z|ljsZMNA#b&J|eSMBWk_?gy# zVtIU+T{g>|Y25gR+! zr)OcEaF+WL*o>xNZhLF8KS~jMHeX51BoZQW7tS=rA^~($Vl&8=tz5d6us3*A61rcZ z!Sq&5^Dqwc(DrH2!`*`VevN`#Rg2${Yzs>cf4j#jpC`~VT2+pNu z<{CmHLb=1IovT+XgOm5`766M+ZPQ-RTIXIjcTZh^)wF$dJ z-hGVcIQ3>}wuKIub%wfF9C9}kCY;KqdG^#>v(&uz92ATC72eH2X9G?AkI)8iPYsYI z_4uVkNW}Fki&JN(`11B~jdQ3$LF`cVX_!9wk7K&kTY=^EZ&GfBe45*Lk{vYfn_XOw z+e>TN@}pm^A0+w3mc3#<343URPwW6M_FNS39Y^?C?49u`3}?8SpNM``sTCuNHHPh} z={fJ%#^`J_$alW+7znow-Y+Ywyr}#(g~R7IW9?WIwaeB*9t#T(yNYFw&gXKjf`B7YEIlrt4 zM>O)A=hEUWJ~Utz{L0wXFaOUoIBwCUoU6z_VJWTz?#AEcl*H7$_^!2Zhb40CYI@JrVt2P48CFtE zjQ))`YKP7lDF70I+;E<0t-Aevrz|o=@LbgGZI8g4$8FT@zDuLuWwKSTt>-?!x$8fP zq$X_m!_5_crA!&L^4N-;sy0q3ibg+J6#pq{&OIr2vOo1C8EkH(z#2!s``C#iMVQVij6DB@yI-R}y~GJ^-~R8vEqG{l@YO zg$weP`Y9)nhzbh z#0)Q$Qk$i(x(WgP4XUFw)`r$9D2^334xe6?yKl1->mQ;l{xV&2@);MrJs?D(1_>yd z?4_gORe+l3^?9B$+)eAncDb^l{)V1xRFwcmj?&$KjoV{+U-iy!%$hFmzj${yCoL~* zPiv^!nQ5A7Pz(N;Y4BD-?-A2__lnT)w}R3F9GZPlgH!JAc3ygtgs4mE7c~%qiU>W7;(GO=^_N=jybZJY166_(C8+~c5iu6A7kZka za(mW_E3P>)Xrcyr*aeDkznXn*oa?u;nKMuS)#3pcPV zOLmAir-6bKw~Jy?*&2>S0eP*ga_5<@)ovURc1^Q0;>NuS7MkoE!e62vKT9GRjWz|% zy{iv0+~1NGG+_;cnG(43@X+aDpQ}v`joeGQLtnW(bSIohEfB+8?W>zcy-%%tzBWcu z$^-r1p!ce^6qUHeUL+xv%#>?|x4ua5YG&@J6FW|>meM9URo@iCLVFzUddlxF4(h6) zd+l!Gp?aB)7Y?r>>;gV$jU98B1B*@6fmj=Jy$$7-thgC3mr3}_Op2Atty@~{r&V=itD*WP?Snfl=l2;7NqNrb3~c#aftoK;N|17I z^Kib9=Eyx zkq*Yupb}T0RUMp8o6MV;jt!{rST(c?$yy#W~Ek> z>G3KLj&ez7S}8(|(RRphugSyS|K$gni%}zek)e?AN*DNZrw)US6cciZ(qy#-iJgd# z!QF_rB?5yblh2?|#b3J_-(h~lUDqNp>k{$OJ@48obP!w|*apeb;4XQa9hzwg+QQL} z?4`4A?qlTo!ZRZGXwgQP)6&rrWsK3zaQl}6&u8b4^pqOUg`v4m=>;8 zT+rDnbX${?LVo`tT~HA+LYGIXc)cNG60X$Hk~La(+0A(IbEJ}|0a`gNIQcdSzfd>( zSt+yU|27n#3^xwJ>DSpO_@EP|!PlOaw7XxNG!Y;mEP+Kg?>?foim4N#7Ad`) zV5-?lW_Mr2vvLnQ5|=8`&;JnE8CY=ND(38>z;#V<7s3;f1+sm!dt@7!+*-yGf^1xl z3vNN~mA`*k+Tmx78~dC=f4N(zD!QVMOuL8ZDn6qPgpcHzbfWjUwpiQdKnGPyc&T75 z?rq6EvV}#wj72}CcHU})cYS+bK-!Z3htMjEwJh4sk*Yv5mZ9vQ;|2c@8TwEFc)0d7 zvg|e;uPf~I7o5yg=~jz%iid|WLc2cC_vUUwOhNmTD)Z~q$N=`TPP3&N&wkJ*B{TxmPDGkX4yT+dCtUliOvGM*#cnD zT?28ud)>~Hr&K43pi^hDtgCa(a@`@57j$A?K5B2cp+ z_X|uVH?`eKA{9ec?BW5i8ByZO*8R=4-;yS0j+Kzg?Jlq!=H z%NI^&-@8}ywgI6V)yIE9=I9p#&Ka&B>WVPH6Ac-k4Xl;aQGT2YI` zZ#^hZ{ifH;Y`MidQ~@Z(8IOo5+ca!Cpplwi(|2r4z;z46Wua%k;0$SuSs7=j^S>ZW z1RvIi8+$d9#H~$EN06s>x zo8~YZlrg#lV=MY|ByjtILW*Q3<<}+v~;oCj^Vd@|cU?k!K8wd_GCPV3v)4PR7NKRYr*%@ITq=n2z!OwIx?v-n=#x}7f@!ituZPCsa}PpY`H{lVL~W8%w8NzKd5UXgSY&?M+Idm41d zJiJQPRBub;rDr}iJS<I8_%xY)6V)H!jg$jeop`y?i3pdU(CI@sFvc9|*dolgja+f7&!>tU_aL$Q%V^J}XTf6^M7w zH*SkfJy~BiI&*d9Yt-P#af{knAN5UeP_wSFxWyUy?HlnP8*bkOx8#H^Qb;LmPvr+M z94WH$BsX1c|1{ZOC3?~vNedF&mFW*tuTyyX{k}C1YZRI1CjS8=B`L&GsC@1vsk{)9 z661o$Z;%17R!KK=odM0|lnQ1UiJh2w&Ag1PKFn#va)8&Lljf+r5s8&?Tn8En&lgEH z4g7iP#g?)t;|feU1f^E*h`x}YU4?pqR~zl(hBpOtSAX=lV$)<1+2f4u%R+wHr3JNJ z@8K`)pJ^`>0|ccy;icF6vcN*$VeQ$mWBPeo%MnC9e?D6@h{8n?uW{8jXnFlVf^&~> zSfR?T-1v6PlA`9^uak3v#W`0$-P)c%q3syCK-X?bQ8WiQuWjCR5vjZzlvZ*J2zQD- zKkhqy0>9SZy6WI)`;xSQ6b}29j_@z!s2{7%X19bt;E-Paa5!=YBSE?9vjg5bUShjE z(OOY8sl{#8SkV_vtwB#E%pd3rqF=lluc_}I2Aobz?p6DU&akgbwbUkz7JN-;PB^^P zYTIzBIQb$+BFNTI9EDjHgqg%_iC%dgkl=D*B`9=O){Y|gB<#=Z1__*poPX7v`lVSMetU(W>`^rmt&<@_vvK5I&|^IYcUpQ}?A2@&Ucu0Hw_!2Pr@>OXEZw{^KU`Wp)DC zn!E*IgFySvVqvjxalmbfE;{nZc`-q2>%_^@Ct52igek9*AmBhSH|5KE3DE1?6F%>2 z0%!*)#^THjd2Y=Us>2rONE}mi1Rb<)A=ikf?LHJ!x#8!-=Y*XN25eWynsAD=jG9PV zekUbKTbbWLtdfkAX3_sh`{nFk%EPI(Cw%W*ah*wb5r2<-j;w=K4YT1op`3%qHRlR6MWrlw&9hP?91OS#b3M{Mx^Nt zTMInI?y=zRoxw@=A=41_V&!CA;dSd#QR^xiWR6e}mah0@9Mt}Y)ohc;`zQIaR&mmW zIJAUQ8}0Wa?AGQoYP`D*zGtXdIk|B()VcZ5b&u=6#hubw-7r5#f9D0mPQ0(TN(oIicP zdy{k?kj}@sZ2|X^Q~v`Q z+v386W3CVB=+gg)GcH>TENR%Ef%+e`r9AX`lx}p`T*lRdkIPexo9dK3n97xgVxM|? zY<|VC-`4pQAK%P8 z#g|WGJ>v)>+9<`CsN7_7DEXAE&cwyDSzCJGqP2^i0Dq>stE4wSSIx#M)I*YM94}rf z+?46uafBozEojiPX`Wx20x_Q40xi$Mqs|4t8rCZ?z(1NO$<@{+FJlGD5Y6yEKgJo; z2nJt-i~`OHtruq^pq4*$N1`ZUpmp`Ew_GjK51UIORq%U}t$JWP{ncP246O{mwru

=W0MPQag9JVq8v2u(8Lm6P7gv^1{Wl?wd|wANlo#D&`pGf>{!sJ%dJqs?xMdb@6@!77u(QF4y_ctJtoTE zRbiLw8A(K3rxLrJUVNii(|40y+7sB>LlO}G`>0K8a|FOEE=X@g3104@o+b1Wr;;wa zDy@Yqt6YXU9S{Rp;Li4q4f5CDOPJDEDBF2>s;nOwMuR>=)S!N!3)`=Y0~yD{COf9Z zsddug7g03XK%a?LJgAOJ)VLVT1z;4203DF;g$L_!o2!sNJ2F>k&bRtTOlNVx6+H$h zNO6~gGkpG>?l01LMO6A(*;DA)%08=1E!N`#ItAU?!jle+>kU>^=Ev7=>* z*+Tvk|`G%~0(+ z!e3_|7Sz-sROj(>OP2==K4FR2DaDdPTmz5>I$Pj-L0u%$e?rQ{Dcv!A8B?c=Yn^b=NTYi_ei+btci@D)Xo8V=oX*;Mr)I#EzlMzT1T9 zq*ogAwF~qxWL5`$$_l@xkZ{^Ql6d62boo7!mrfq2u@d8wha`81XJ;Y%-*NyKgMK^C zbMkuD)ZJIwwKMpr-^NQ&g5Lp>?xs7u<>%+`2{UD2ih8Eosof7yKzPDaX8Qd{5mpFu z7+stBiF>{UU>pi7E^PW;MV`!=!8u5GkJ6lc=xC-+I`@?9(}5VEFR2&8!*6&4;!yk& z_1uXYpf`z#v=`15`jUzaTV{Bi{K)qyoAiy$KSr8{ zx+N6v4I=+*myvw)tH%SIGoJf$Js!GkQ^8BA#!rQwK)(GvO?s!1>G|_o-iHl*Qhf%gp@# z@6uTE5Fx>VX#KAr3fuv1UgoVYX1Dhehs7tZA|04N99Cz9bKdr$rn5 zGx{WN&QY7f8SYtI`^s3zK+9NS!oY$~@O03r`i_yxW4j%at%gk=$hrC{@;O0vOQ@YI zXrm&9#8PQ4c|7LJ7Z;GOmX0`DU?fp4(L$tT*Hi<-_Ak0JaKN0+_4BEKH64Tay zTOOaw)tLgicRgJLMEc!Dm1Wig%%}k`HqwGgalHnY67QQlLaVR*z9CQ*rST$O-%B`6 z{+ZgcEkS&Mo8{NcH5nKX&lZ(SNRlg8Ip4N4e&PoZXo9(}6TO$Uf5R|Bd{u6&S~&T0 z{bx|3?iM6p5Z29$jN%Ghxw0AoxTC;$^wu)uVB|@o9thyI(&JsU-YD*C6mKgu&;gS^&DZ zlx;OcG&pE3%NLOlXbh^~iORx?tyHX+?5rvWnlP{d6_tzaiztz5=mj6V3tL{}5O!G? z)*xTBRh>i|xId_6Y`2wdpVXdiSF3pSxK&jgn?1u{+~P%L&W>cX(^BM2&9y-d*&;SW zCJ}pSQ8jYLfD_xCSb&8A6(Jh}hT;gp+#7V^z*0z!S0`{?PG zQ~Yx{OpVrmCeZOGXLx9T3GxN82KDoe}XBfAK)WK?`AN=Bt=yHi3FEHA*air72r8({EUKXmW zl)o`y<6IR1q4p8Uw@7zLPt5ngq#d;tgUR}SkCJrT>VcEiDA|v#2H)S!k!s%2pB`*7C|UjW z;XidG&I(xH-5^dF-q(C^CgY!*oG0epmaIbdx-X)$x!FSz1#XjN2{&Llx=C*O#oyZqyk1ry{eH(AstMuRV_f{S_}PK*iELUu@Y z+}8+(j>m})vb*f6X*TyYprly>Sz5XjPRzWyCQhqa#qK56NaUIK%hBa@9RL z@t^kcPA7CrLAD&IP7vk;+;KYlq=nP|cWoYe2q=a-HeKywP~ti^2pg^Kd_{!< zliuyN>NOKPWxrq8E`IFqOv!zJhMEwH-$4hz}nraZNYq5DL9K_NK?ySaY>h=CM^FFMt1-(Zh5!x?P~wuiIw2iXb{lq~rfI z*P_dIvj=?#RKTpaWbLxKEmPP(MqsMSJ-ivW+1-BhkG33Er%v0_tmtOLKdP;oH15zA zI*p(`GaIb`Ct9sB&X>pOLzgWGg<}5y)(=!^A+-T^JK^?qj65>UAC)7s1 z(xN)S5?mZ#raoavDkFEQ%gg81{?D(!zVHJc%r{;2*Yte&k+$wOD~-1(Pz>>^@SgsC zy>8#B8p5Tl`Ett8^_|mBB)QAn<<`r{ieg~80l1$?eRqz9f}bo!z>$olUlI7>`da+= zU;Ttp%OIBnMs8Skwt~nUu|W#AyRU9OzAR3ntW?vsObb=j9SBiUTvYt0s84& z@|eE0YMX;r&oip~+!f5ZQL@us(^XrG3a0U1&wfr1*1w%`u85bv26)N-EYVvCGri=u z`c+@^$Bmp0=yp~WW!)k9_Hc99#YrWw(HSR83NY&B&bE|0T%QFKl43{2;w72)%djlR zA{(D1XE-2kc`vejxrp|ikbY7%9s2xM{&&nI71BEZyFO~$Z%A<|al14q!NAHg+6C~F zvTOKwvxd-+a|C9N!&lI>{PJBWNJuY}3;)(!mK*$GNmVKK`#4YI4vw3$i?K+c7y&>4WPCXX&Uud2X1tCh3rYy3*#B+_Jl^bHf$>nRYUg|vBn5LerD@AfR%O;$mgI~FJqp zi6-@O$u()1Qv;R#ECSdT;(MIYQ|Z<>qTeD(+ZXXxX^Yr8%(gEUG7~NakJ}-|@Aj8H zax^l^Zi()6vh?Gm%&pFUTgvK!6+9`%OwkZH8-#SvSF?9=iKv64qnRR{MPW;}OAZ6X zj`|)!t2tSpwJffz83V3)d&8b?m5XCg<#P6ds(ktcodNkY%sK@|ZKBTb3R!4lQr1uH zh=j8YJJV9Zmsnmw+v4=rlb%p)iU~_xP;LW{Z;8fg-BB5_ovt2ZKRt6hY#HF%1RmYf zi?e|Kz0)tWTHJQpbTcpHAJFcyC0=L97jcUesAzm^QTW(HAWOV|j;_40YXR>$PU?@v zCyxmW0;Cd)#YdP%Jl2H`$Xs#4I-wo#*zBp~Zb%{W>*u2E&8uWBbEW?_O!lM|&CABD zu-q=C>+XpNYNj73zJw-Po}Ycw7U>MII)!w@AK(6v2b<*2DaU6`(?jM8a=xyGe2!Fm zkNukyET#>BI6}HB`vr>d(^+C2^sc$+64JkkT`}BD7e1 zQ60CCE8rO28*vGSmK=u-SGNp%{Zmi^f2G_qD{{L*4>?1S!4HI`@TeMmkiId-nVnM| zGZt+<1w1fnY~&}5M1Mu4P$EEL%ht^C_54-DwC3j$1>n`WPObnRgg#9z2}_Xo%@jjL zKxpnpuEa=QS%0Ra9MK47dEQ%zgHZ<$oA2akw+aKFi1El2HE+SUOlN@#fhycK!+U9) zITr?`$;uo_zCsv`mg;@gOg{=4%dT|-ZP98) zp6e1sYS4WE4)E>3wflAW+axTu;qDY+^}eWtF#9B+C8EfY4{bB1 zM=(}LKbiIaF~aSFU}Lvu+jDaaV-m1Sp5MLyc>cS)2xZ9ELy=S{&q^gOfX0=pkf7e0 z1#n=`^85HE@98zB{z(i%<1j-uWTl>P!YblWb^VJ9Llrnx8f{Atxc*>)b-&h)kF+%BKvq$zsy=AR6&fMo58X1uVm-fJm zVU%1Z9%!0Cvqvu6(KJ4+KRB^BtLG38$acol5#(iL#~y8&i|^at1Lk_H&Ue@BsHTMJ z;e3qa@{{5`=uF3iey_#Nb5ro&+70qaKeMr+E042ttExIlxxfUAp%(&WdpX;u0#g}U z*b_^>KpXNcU9QEmNi_?0}zckJUIN~d2v&kk@7 zPUY{(#V!*YRJ8_Gx~kS0r|KTgW&~lC*8= zJMUXo&vujpl`m+Dg<`Nr;A8xFvstsy40{T3A8aM4n@~8s`$TrNve0g|%30R(+TT#H ze9dQMGb`2r&Z3z5trFJu^N~7NPWPMPh_wf_kMUwbbK4i(n|T@gmssc7K6C%FsVMD0 zE&c0+QH)p|c4^y8-U;&0yjgBVU0&vllfuu3{snJ$D&!!C5*atDOvpC*xa+NidmFh+ zE^Fvv2ke7ICDTLlKTP+w>ECz!;TX0m9+m?T>*s9yt|s~AmEy!n+vtRFIP4DcDjb158nStI!u z5d|0$11N^SapC~U&du=T?c+!8rsM4V}V-0)T9#fHuy!OaL5qz0NLn zRz*G8%YYQ0h`yY6|8b_0={&unSV@KGjZ@^;f%Gh;z_94j_m{(njIY=uo%DGIx6Q9R z7a!6j>twjinCw2Jk=)>z73_jBV}yV`6V(=vxZym3Vf58)jwGE1Ds5Ns!TWWu_rGROc{`PF3!cP`hFQ*PZkEsNw6t!=sjiFq3aAtjs6iSn+jv zqIz)kzG(jiQx9*2o(lEBrG!Z{_=7T5Su+4L8YKd>{RA23-M6tb=PHC`(kaKUkB>%8 z7FWIe5A1ymYJ5jR`Td0Z0?Pi|jir+Fzu>py=XQ_INUqpKk`c-L`GfSn&XHckq}WwR zP1C*sP(`N=gKlj8{v~i<_ZrYu0I&DeU%ycsIMd~-Y_2R}23!eV60tJfe1qCtbHB!) zeB>z=HeGZ?^_O;d%8sdB2$*%18GYhQF}dVw3<7QY-4n13U12UVPMsn^ z`3xWgkUy%E?!v{n5LDaVCL=86W_B7&bK(bn-i&VWsEvHBr|1cSS3l&|Ruw!65xQBM zI(<47NX=r<`S$qrWc9tX53uGnydSvo?feO-| zWy^J#PZ0vRK8ir9q*RNVBz$_JYpD*LZ|^ff78uc(EKP60PJd>7Yl!3&-9X$l^O3|P z9j@S)?@boMSH9OcSxlo-kR^=ryazU;&)zLw*u5Uk5|jF%hfWvWu^4f&$dTyrkxGTt z=2PHGOg1`PlRXq=-CJ~H{hNH_xWm{P(~Z)rT*l>Xl?{Pzr`Tl=V%z#mPdxU&bfw?CZ*nJFdtS>BLrxdTfRtJ{MPwBh>p}uB=T@@Rhw)rkD~TV4y(bO#`ir z*;1eBiy3yDGN;XXu^w?2XP!W42`Xeh`s3{t8#6S#%2n|apLyYZgA;+DWH!dz*pU`? zJ(aly@r#t7@>Xp;W<<)GDtULR8QXc`)UAw;>!Rs;pbc}O=^Y)SH+3;IGNcH{|h5!$-)VAUEfj`8wg&mzvIbJzBFD!@K)mOx)!%#+}df$gdR=uK+ zyOhY*$R74pNeYXNKad43?zhTEbbUE?D?Ms`?WX{g!yQ^^Hd}i{ba32IrvD5u=fHV4 z0!KWIeeK>g+xv-@d1%!v>W2aaE59u2cB0=0j|ff3y{*8XkB^)kc*PpK+tz3>cHTYX zWNk#hGe4#mn*p&|$y&7Rr*{U>7|Q+mxZG5s{j%V!$G}!4<$kt#@3_NhE~ZQ*cd)KA z&g*Qzp8)@)zseQ(pwY0fLO<_Q%5gw?^Tn&#Mbf(GHqa|+rprw`p>wnux2Nmn|Hb%p zwi*)ht7cHn$U0PMTDm1qWl(u%zoW&j`)RRp2YSttb7w8O#2`{>wLyF1H@-5~ZTyq= z1KyXa&c829#*>-IyhDhw8g~c@B%zyS+EeweP3EOw%*4OiuYVEoY)ukGsS@{n+5SJH zkeW0S(IK@7#!$*2vx%%y)Mp?OyAg`XN|bND2Qg+i84J1h-l*H`PX0bFJO7ajy9bmi zYd)Nao9*&7;OD8q4kfL&*(&Ow2sI%ye?SURbReM#>WkeTiL$m0hDC;XfpY$ouJOS> zUx0rI*mPDWrIWu&a&U_rYcnNq?JnL=0C+YDrY=v8*|PULNai=>*W>>U;zptXTCmLa zj7ZYHNOXLGoQl-^Yk~{4Ki}d>l%@FTb5EZ3+NOiX)R@5`@kQFNG{kFfAq3?5YklcA ziIO20Wy05FR|wCSmcn1O)$h~|qK0oyFc1{38j>R58Ok-qu|HX9xQSor_e$|tfd#$J z+7&7xX+(?sYtZ>l>HP9~i{7rG{obGOFWzr!QiD@JwO6vknMA_~~Sp3FI&D!p>f-Kp3mi>y)=Q_VuXuFX;Qs4}0 z3Pbe~dOr4`YO*k3*>a064%t6r3Z8dIvrrSGKOZz{K)~$*oNwc^XyWegqUq|VS)9-3 zbM*)-+#E0}7WVs8S-DDK@`07{Z=c$C= z+5OVvoPO)Zu?p{)QS8sj9pwfTZk07VSJnGr@NX6DNg&@z9X|KNa7^E}^R7wWTK!i& z1r}zBV^-PNXtBe`Q+oC5@6T$t2hcW_7|(is!!6bG@Ul#|uyu^2-=N8lf3Tzf#N~Bt zm!Q+g3U0jn--|6G3Gvz)&0((k)9pz?{DuT@12T`GhOK6_3)hJOD@(%!LIvXaqsh@& z+Nc{3c17o8cbj@lx{SlAbaH*n`L@`2kQ_k9#rxu2w18;t)>(T>L_15eC?7UieJ7;* z3gT|~RYYVLFA_dEql!e3Xh3@GXr8O&fVb67CcE^oFaY@TsGX!BZCnBXK(a*nl49GC zG-`p`B;OBcR0Jwx_xSCRekjNHWZ)+bj9Ys|N*#xOPk#m$#=krDgrw&>Gv&kggQ%aF z=W%&CmSzSmzo(G)RuCV5Fv_==6`!DKm=xT3to7#Y#`cTdmi7H-HGuRnB%y(TTEv+$ z=1)K-B)#M<07td?d?1b*hlgjWon32rV&~+5Pm^RS8-Ev>Ju^q7P;f5R!@=PeWabNY z(4JNP0h_B{)4c7cH>V)hK`XC9dg66!svd()@A)2}AzS^w;$0_oiM>-AoyLmTW6I@M zG($)ytoW|-=76|V*UbDxF5Pn|`-Rx+yH}36yRxS=2Ia89izi2+0S~7S?A1d*mk=us zPNA_ngOwMZ3fzw_HdW!5w29&*SgiAIf|OX)^XZ|!IIIOc%&YAI)B>t z8@(X?q}TfBy@&ac)6W=&yQ)Aj9?=4^REh1b05NOzN`e$vN(@MMRFU;-OW)k{+FS&G zOIOwdch5gtpE2``is6hWX|Zr%1KH!>ZscWvIkBIBPuhzb#AhF^9?^5BM4l(`6U+eR ziwqV5SFyl&|7f+TeitD=6Qx_3f{sP5bvwfZgvtaHP+q_(5qd{5LXj259x|Jex3tX< zAa`QE&4UiVH7D*mZ=}#|4r0=2`CveJ8svm@5Nmxk(FCU#N@XX+v zm?iGVr6?+Epyus3w`2S`@yeqic+nS*sP7dP``~*Ss7_#Rh$PsY-DkBX+9xpPoW(~m zlQ%F~f8F2Ky08M38+Wk1QB^|_l~PEsTpZ`Gwsu^Eb#$tX4^P(Z7-*v3+_-S?i9sUSJg+$T_AK19{Ghuf*detrhHpG1W$+&Q5L;POYHbRTl-0v001;@bxkInRlYb6xwmE>et3* zAqItVmaKq+C43jIS(TkqS=s(CMs#^_ln9YH5LV65Jh9w=L^XAW)Y_kPiR8&T9D2O9 z*jM#g4U6^d4}fCU#tTn)M5rfAyUb*I37Zr6Ea zK(v@0M^Yr7F!=*`bE_<_me=L_+cq*9Wxoe^8))-G=OdY)lFzl0)T+iRUUS8?0HqME z@QT;P_IfqPrJVW1F34ek-F=Pxa!F{nWQkq6dbRU-Nyg5-jrZ%g&JRq5l zkfIK#OhVps^B_)-X(8?@WJ6JC4XG4b)zwoNWY+N)^lV#s1n8RfyjNsV;c)oy8+jMqtK_ zDa{GDQ=}yL_6t$Ao?}slO)(xj`**~5A+L~CMsXc#WPkOJfb$&5LV2Vp`$*2v3zWQb z6G?Gqo>;|7Kt9Ri{lo`jcc2-aciy^>S}`j!NE;mzsDt+!43yd3ujM;5zPKeTA3Dr2 zT=PF=(;m!|?UqCPgVfsuA{6!A$a`^4!_uWEp}}SAN1qLv^hf7<2rDe=IF6tXWvh4X z=4Y8(wj*bUGrmdh@X79pJc<8R4;;LRJ1t8lY-}`(N8l>RbuWzr|gxgP%?)fVC8l;Np}d&|f?vn0u-nKqj5 z!os%$4*#)Mkwza*81q8FXFNU_2MrC+$PD}X5d7Hhhd z!!|gvr@HH6*`5aY>a6H|d_NaIEWK($;}2S}kP}>ayy*8RE^>c$b1O@!`myJzqhOsq z#n0rVyr=g!#vIrM4(!Ks$oI9{B>S~0$|IomZtP`s@Gh$gM7d-%S*`0RVWdHF;Nqxe zmb0}gcsgXH`uJtjE+y}*BNyY=mvN!GV}Q4(sVhqFUgHVJ$*_tsQ4m)YyyWVLpS z6X&J6T>k95(&ou2FRgCI%nY$_NVz6Gs2N`m-dqZD_HFTjDmC`{Rz1P0DqRq`IK4-L zzf{mA2L-b?7tqBRxbKKr=c(8XJ|`me_448{*<}(>LBTD-x^Ql@CrTKr8QGbpv8TPo zmoVjjkL>hIXvQ)*7xSMVPF*26e+)0s#rmFjejGwkAb2~zHPt(foXyYH?081}N?Yli zNj{S7Gf*jcxJ78tW3<{Lwe5FL-LA#YsM8Imj!DifZFpJAH-doQR?ErONKB|=M;IfF zI**1b9X3G6d4^N;eVu$W8R0~j7Ti8b%L#cCy9QFCrXS5K1s;5#?$?H0EuD>Hn$w7$lo(n-7KqoZjZ`I@=qOPt*lEu7{mXoo9!`-fbYVa=|_ zM7!UOGkOM@zrx7sPBKwckh@z<`$~2@h}S}AoHF+BQ5FeVla|=rS==O$#fmWFDb#eF zwJB8`M-!MDKjI3wW#d$-QinrE!at5~SM+&5v!yj?J-*hQ*wU$-5TdrRAljG$JN)A> z{c_dw?0$OH2fi)}L&si;`V(1gq0q;AE zPy;=S&^5b2!R8*{#ge&wx?R?WX|94|u8jo9%*yG>w{=Ko>si{o(qOY}SD5~*i30Q>nGgJtqx})=9 z_SD9W^oW8wm`UXXVTv!gD=mZKik{)gMn)UOvUv@!BT15W$`SUO79{RyMNUOI#wW#| zq6vr@y`b9=DrRu;Q&%*^ctj_i#Cx*6utTO)NV;U#=#f#vL2trN+Y}D| z?%$H@>CkIy0-2k6gT^Qo^g3^Ol}Pdg=oYJIbq;mZuJt=vLyP=YU5NZ_jNNT#MtHW- zaljt4A-y{{EHQpPuuWC=$BBR@7OUL&#&|!VJ`AXykM4jNaZT-C=a)#a7ZTy}lNrK*GFH76em$*jvaHj#9#w0kU>9#P{N7Cp$B3zpvRFbTE zr_;`#{)L%FQnV+*|927SG*Ir^qQjlyPU}YC{NDh>E9pk?@Gpo{8jz+{Fsf>pIWpBzZgW%e}(F?=8(gCJEb-i`)Du!D)Vi~51{tb&RRWv zN+4cHzc@rKp?Lnjl8f~3DS?kvH0a6A#DX*K&tH|nUSk3jn@f>DetqsM`79)*pSR^W z7}N}{Lpda#Bz(-!I%wYfK_O9t7xOtOoIrcQ&)NNPFuLQ7PtQR4MTRL|iJS~d4sWCHR z={UuxjB8lWUoo>E&xY&QF$FLkx|qNZi_x)EF#zhlo6VBdvv`E1A0XMu*g+EoB9fiFfx6}KXKqg`nAl3y1sqQ3@;t(lxVat!z3Au7phDmhr z(%V+V*OgN9W?1!z<#>DE7d2z3?`qi-=b!2EMCi$kf5vCe4D$+2Lx(^`drLmMcW$VV z>o<&ruP^I*&c_19{8F->_dA3u#uP;a%ondTv}E|l)UpSdYR-hl7c`3EtF ze9pf!Vh;Vdav8R=P2ZEx>I$RtVK`I6>p@yIzsG{F);2W^-F3nfCN{;q zC9E2FUL;&#YT@)>R`0yvLT$#naUC1Qj4LuPU*c5t=qKtN@^aD{aV{92?i?z6iCQfA zYA9iz_vPZ+K_s-A!Uwshtj!gsvqbhly?U~=fbopD@5;OI>Zvm0dUe<5E>_crj{MS#$+;lx~fC*x-{O%i7MuU$cX27yh zhSIgVWqhR8$$?E|kF_xP>1yY3R^3~WpW)-Qy7*MB=hoCZx3u}>*fij_&CHeC8U~)a z^m;$ax%3aN;TNp2O=})x%X_9R2t)$zseJ?q4OD~moD=`{(HO9*6=9%IVN+!4GS7O( zSQH^RJM014J~Z^b<~V#i#<2+6IkmB#eYyV5X&GGXm`0a4}IvUkuxmy zt#!3layI@bL;Za^^S(9ys13l5^EE0xSFQg(22KY?$R@{yy2?f7qN}kld7^RuIJrAIUC$TrW_N;b2)Dck^&aT6p(wFs%<} znMyeNqsX~Nd!V`xPCe_}ktKLOV8S~`C0`is3=#8_P`(Da(^9lka3CTaIrW=9o*1)7 zca2=a=X@=WE59Fi-EqDHhw%JaMJ!o=eT$KoCmI^%NFHC1=O{bGob?kQK;E&$dBjLK zG&Bhu^BsQZj4S+B6L$E$*~HPZjBV~i<+GYcFl~~_iKlpkYV;$hmbb}+d}WK4fxe-w zEQ*|;G8-vBce%;+2U+k9a$|BTKQJ%$+QQrN1M4%`*kqo4F2}rE|0z5d>34aeh4$8h z53y4$e{0c`K}J1&K`mejPy$#C>Sub&IU&v@9@+{}0`RhZM3`7qvX}OQoIvjIbMyYv zjHsQ!{wHZ)gjOrZGz=I8c||?}<3?*lqh==7bPXhKTlPS-v|?q8mRY$Ti$kx{v0bHY zQVgD?WL1fIbt>pTR^9c~E-+7Kh!!2V-lR9`2Ae+;I*-o?firwW*U@`}m7YmX1S;T; z157CF@ch4FCn5l4*y#l^jpjmsH31en3TP?YbIqZ(S0V9Qoq|MkMS0#EjjaO61AJka z^E-%Y3#V-4-fVm?kJes|<%aDeq8*dZ36Ou@a-Xer)wYKj*{N1E zl>q9@3Pk$Pm%^A%79AG9U#ix@TPW@_C5k%3okf*_FbBf5z6oFQV1=bHVc+>WlSwUT zkkA(K8}A6RzxLwaJ<9Xhce@-LSZcHNJaoc`VJF_uf$z$oKt z?ALZOjY)Jl-6sF3KR>uIOzy14cE>~fCicdQuhnehlwit4<0}W(Z_QZB0dpR*()AIQ zM|#NRTXQA3ZamHBmv~${IpXN$#T!jXb4Y4!o%P&h8ND!3;zMkYi{%_$i1!8PLuIca zt;t))B3+_pKIiCslb1qIX5XDOZ%CA{tz9`j$UjOPMgKoaP2X9%o-r7YO;P=rpzZBt zXQlKyDH+Mo6WVOJ;}I>sWD3{`s>o_tzk87#&Q}TDKCf$Y@0{Or!;Gq&0Ar^b6B+O+ z)7|4;f^Puk&Sb&qrK2C>l@PfHW)tU3cBZ6IU<>W%18g75UyGZ5;%AhD`^Sviu;)aM z9qf!bwuQ_kD=PR(Sg|J_4*RzFm>f8^rYVZ%e=J3&t|vSd7NWIu;=`pA5#HHPTHB4+ zz)jQbf?jxEw?Rs_FsZU+PRS@QR&`Tw! z6D(<0tYN3w=UC6}$ZX&4E`dS{e^cvO`msFT3ho1aY|?db?)}E`Hy2Z|F(=2zJ||x; zL{mtq%MG3xtIK2I92XMKk<>QcTfH%nk$Ksh0Pf09w)CVnn6xSQxmcF+2Debkp*!MT zm*F1gkX~~+2W&Qn@^Axz4t`$E{O1R>`&v+jo=U@CcYKs9Jg}+B?5f@D5kaotlM-K_ zq&KOr{-^G|_?qyQ^+=q1EA*e9<;J#tN&^1!^;?joE!i_GwA9<{&_FHpu8z*(u?KEh ztmhJ@4r3wC+vg?EbdR@33;4|97Qf%(EJ5sI@qM_v4Z;KIvLghz3&1DPkKG*ypZz8v z+6+Yet$o{SH@y|Wx zaHZLk*-oiWe5`bZ_?c9*mgBx6+Yf9c_4`2t5woB z6#gEce~NX0v?p)s8&<6`&v@wce{KQn-iH33${fPx_vMVJn(qlh#u2 zRjXl(=>D63xIt6jpmH%$W9K6m+}dKNbm#~4T4Aa?J%4IsN>9T{TV}@~RUpl8G*DcOO;{14VBKtsa~QdMx(Wb8C{kS0YCwZMP?R(X}p7LAJ-Aj~bei`~OB!+Fru^ zdyrfiOsi4oKt#tAEM9yq{o`GVC?mqbUt(V@bhgo7-`#*%Cf=QX7H}H#qU{P7qqFRZ z!S{hf(VW=^1U=x=F~^&Ez@)Gk8-44Y=8?x&)3@|)u`y`__a4P0J%QASGJ`(UIRoB` zdUY}j8+z`Jd#B`2zVX3?GTS#x@5ae_#n2}K9Vg0M+Thm!PwEcqjoL?AYmC71F~#Y1 zuH7jwZpfq;xut5Qx=r22t-i(2SS@5g63693i^tsc(bcl(myjV2NHy>?YAgsQpoU@A z$9Q*#zoe);3hSZ$uS3)j`K|F6$_3A`(yhq1C|vyrzS#L2|7EORAl>PAA|~QW<>F5m z057Z`ZuzUIz<;E^<#^?*_r!kvc&jm^#%EUiIO{!mjXjE36x#RRT`eNI`bfbKy*Adj zHi0#x!!g|8-ZM2w_-UKu)c>O86f8P)4uEM-sZHr22Nvqziiu zpX^>V+%+_{^)AJYPzZpW$RxS?Y2iYovaMtd$<%v#HnvSwDR!`|(-s*jgz zS!Ue$k~g{oOaexD_PhbGuG=O0XQpXch^Y27AM{cI7U_0!!Ui#Rw=@xxz!(d~W_Koo zg;nwyhX{#JAk*K6dS(o+zT9bxcLJJ{hW)WC(!|uDwFxnDyX0aEQ1*tddLZ6=tJmFR=VA`aqM4^SfiTb z_?->%Sq145t4h*umW~(7nPRBxX6+ghbKQ004aTROuNeU~);(OELm^3jj<9ieQJvY_ zr*UdB&K2W=#;`-9gL|tav(9G`GqeSbhW(zid8%Kw`jmMF9LuyPy^4^zL&Yk%Pz=BD zG#AT~nOAI-45Yps642Teaa<;8NgXRm+&N&HtDY9s-g@N9O79pEJ5nuIR&tTPg|&aI zO0ozEeZXWp6t#jYS(AzLv`aaqMdbF+eL{Od*u?%e|FM*k-a4hTZYr=VEyNiq`c7{&?m`%8LBvywwf&Sa zK=r%E_&}8vbWF9HfT-*K(xy2ZpuaHSt`D`mC6Rw zy$^Y2&of-PZqjbC=LaO{f2oS5)C>5q$%&gXj^V6+5eX}a*=g~gu6%`vQjK-d+fmKr zCDQa-_}UXj&jm=l&N5E}t8Y+`!#F(B#L($9EH_IVEPLFF>l@Z=W%Koo3N>ri{0Bz{Cti|ER1fuD+Uj4&fq z|EP#KIP2OG8zFzz&DV&vCThwo)?QfN}1WEV} zc*qdY=@`~%vMthe3iptQC#?D$7KI5ZQha&4Yu*sprr1%{lCotVks28jR2T@sHOBCV zh1M8ByJK(zTnH1zfz&@k5S3equd9k+%#0~Br=J`c!kV%Ffgj!<(0HGxFyXY~;~6Tzcl>l2|~ zP@snOy{ASA9Yq+9atB6hdTFM>QeH2@*av_E+sBHqk2O;|-}ezumf0un87ybMc~r{6 zuM#4$+|g&-bmf(P%kD@}x&l*KN@!Gk`%-3Kt?0yrfw+e!LzM70pD2uk3%{2p@UEpx zRoz8ScPL$f2nC{dkQIo?q?t0ht-Qi~BlcZtFe zZ|^?9Uuh6__9RzCfZ6l*ZqHJQADGzGl)E-CO|LooATIXb){+@r1j1NZw$@y%z`=LQ z8g?&*hH+cu2zdFn(}Fj>?%mI;ML=t?i&q9&vW~97zN6G zKO)ELvDGB4HwW2RniJ1@0lf?jh}3;-6nfp} z89aO^Sb#xexcCfHjOOKB?K>huWqLm5w=RMb)9wSJ{jl(pm6RlL`EKy9DNyi zxP{Is()+WEkKZ9HPgJahPPtc3W`-;OIp*W8G*dzjn?|_RU_ontUz9CK_${#JO7}k3Yv413 zz2hq6a7&lVDNG)SK>0R>}=)Nidx^+Yoca0e5ybE6R^=;oKCWqWh zyOS>KlC0K<&ay9$Sl(G%jf|*`e9@M;lkOWZXBH9U>vr=*(3>zzy+MEK%VzQYO(C-xM4<6Sk-+0y8; zJjbJ8NJVw$ovwh_)sNV}7}5ftWMwFif+zCVRCX8ltsGrqtX!G%l8u|Vqsdlp9B&3i z?k!qt@+kNe8-2VP6el~mgHJxUPtN>5HHb#ff`n*tqLiEa;XjHsSWr-?Lhq?a4l5|K z)c97>JOiSbfcaqg)WPKop>ZpMAyG>yozbiFllinoTZ1r8GqCvK0qWGoOe%w3*))}c z_mxKP|H;2}cD2bo0?~ztL`Wm#;;&b*xMoNI%hYKnI4`u}^_&_-f$3)$+;L^aAAsk= z*}+L_9bfL0?xJH)bP`@51P9|UD@*?%qK49)4DtkC5sr7iW>v6}<#C=1n>QRra zEzEdZL*J_^b+u!EJCC3L^*8KNXk0&M?w0L;3ehRLo&<{ua4h&W01Uz8pYtq+x#bpr zvK-Za9%fac_q-1G78B0^XQJmvo;f?1xx z=Cx@Cjl}|(*{@2>Eu zqT}hvLvnXa+J&&g(JjHS$H6CQW0v|73krM%<` z8x@^kmTbC;wk%HzV%{hrZ4g8up53BP^*(IU1$ot6$hV(Ect6a^caV9XQU%61h{xK3 zB!nh-w1!m)aRU)z;TFp62FT9tjxskDYQ517A*smCM?E2aMdX0eMPbGXJ0w{gH3$LT zI5}*a8*TiRT~pHMtnPJagrA70xw6|#;X4?C#By8%p6sx=n7%Yp*@X?Kp1PU-K{0mG zfBeB52XH$>G?A}2`U1GK_cXu8{KKtn=sW#czkz2fwA^LQ~cO#$Q z_D<%R$zP!fLaP37&aN|EAG-jmErW#@#+ld_!n^^SQ#5=NQ`YRY*K6ClKDcC?|32V# z9hz3ssQuOz7{G3vcfFrye4LbtWbN1KjNnpe@~~CzaxV}?yl;=Ws+)=s(tCd zy${WJ@mrhp1OLf?K$t8t^UVqoqvy11qih3^aA6nXh}JjAwg74X^#M%&-~i%lkdi0- zJjMph<_1{^Yz53?@13`-EExdoekW-Smdw8ilZ2ZXP_$=0t_3EznJwhNw4P_@ra9FB znws_8^h5z;iO*wlg5bkC4(-9V`Zg9G&j|Cy&xB3cgG^h2V|ZhO+%kAz;7H0gdJNz7 zwBBTIJHdcEa=FA)1fH)&-#gE!UON-^*H!4G?XyK~{8fk%s1Y8@gk&qOVl4~!0Te$| z;MM_ns;9O)q3TQMj@WnH%v_7TVVli3!Y}IKW+JC9#-WAk81_)8FoH9XDu^l|7q7@V zV|oqBAIp^9bHFu&&4J~1Wb|ivM)R}j9`UKFq&du-%mZe(tUYo?~8)7nOidUdl5>x z3yrU-bsb;5f%GRUC;8@GB-U0lwZB}Kt$so7XLLDnzxPfp5EO|CWq&!IT%A6?G5L;( zt`0Rq0F@;5HjN~=jd%=ohN_iEjKV*4WklZj4!nfnrT$dkU(Ou5cQ=e<%&Yn~u|i4m zPXo)h$8FRsqI}y>Tjg9;7ZUo8uOpRxBP6SfZ<|61&zCwua^#ui+cQ>W?7S+4J9L$} zFy;$Gj3s=}w1%WIj&icP!bHE3l%FwM)nmIdN#nzfjcVU_F;a5?my1?OveRT6FD$QG z=}#KqJ_^J;?#Ihc>jw6=t%myo}LWI>3Fkhd+SB5pw|W%h2zMV{$<63sqXXi@ohDnY~|=j-mR z4b?!<@`+q$;1wIyS{8+(3bsav;BoQ9@| z05xP|HeSGI`4CCnM{-hk9I1gCwd%VFyGJb*ns!4w`r8|qikbaX51YsGQBx&eKV1bb z`~>9Gjcm6bJOtE_faNCBTvyH(db{w&fu;-Hl*7HC20ahpc53D1O%Mv$qtEMm>pTEv zfehETopwuqc(TzgAkc<3ZxzV^UE4?80R;YPTgxv1e-R0u5z%0ltz7z$1=z$5x{08) z>w=H-tbVsZqIFg~u5_))JpH5<1&+cM(=QaI8$goEZvfWV+76N+JqwfpRKwZ_GQ?~k zY}l&F8{` zv{Xt$TM$V4&F~h$;Q<|$oq|$TkaxeSGkWf7hMdL#EgY$OtC}P#Hq>=s+!=pna z;W6V+3jk7<_G`^EpA5wB40C=%Hli`i>4J;0<*SF8(6X@+UYrvs!%jd*%-ZO{P#JYt z(h-aLvHtDFIZm3MwEU+wR=Xz=U29^$s~Q*ak=vt=>o67Z?F-ZCutrh0)T#a(1nMz?Cq^+wpaPpC5Bf<>O>fkq*IpeW z<^#n{O?qsFLlZ9FWAL(shCEuk+f!6H@?b;?9sJM~eZ zCQ^e1oU4P%boisflOP@^cLOdt-nA+W=}d##rUVp`hCIQ0jLr;MC>O;}x}6&*^{F~i zTA6^mkCzW>e1}XJ713Gz>zcc8hjWX2bv(vql|7l~Y?L1Y}Q`G2zK|_HL zPp?4dP^@pl#C{37xq7oXwQ`w-z1mV>nt?c?n2g6QG)5pFtwxqkk(Gb-Vb0H)1wHzr zEi!g^1L%4$$kY)Ec|57DvQyxZ?O)@OS1}({%zy@Te%u}kJm(2&k9bHotuwOL z1pyr1w|*=D#z9Yjevq6u`f1i|4-q)F{<`Afw~5U;v<@cp#d9n+%Pq5bIBx~VDZ5*y zD!d7t9;cPJAl9DWM{Fw=_7!jix21Ce;ZHUW3Dn38hRtUu1X`GT{@Avo_PXPv7D^Uo zRP?(5`d;k?Fo;6($9wIx8h|e9qc=7zFcn}0q%#NjLi}`N8j_X+lRXok(irq*6_ z1qFg11%eoSpm!c?GE9J!R}2lrsUowGGi|iMulhnG9*Sy2rO> zvoxWx!YZzEI?fikL_Xm&R^{PlZ?6J#b%#iyy&JK||08$&rak$hZcEh${$nU|u&#fV zn3a5c*Wf`Nx5m9M)@yYPy6`hB-HeHPiKU;VtZUSAb#)9rhA2rdtv=(&vAY?HUn%<# z&}$*XJNMR3)n?){IF;ZB0$^wtE6(p4&HK8NBiVBWnSr97O!}p|7GB!$`Q*c# zuj#zd5iv=>=z!<>=GW4rajRTzZ^Fb(OI0mn9FK?N_0)uJ)Qk8<9bR6jFj};5(*OPq zJSgg%53r9d&+LECcGOx#&@cFVz^Gc#H<(uK8sO(}_Ybp*KY$o&?NRT@Fz=~)Ouk#o zyhxcRbZgy|SLozsT*LWD%P`3lR+H+8X=s%tvjRcSKNwG!1lld zl1@TDBYv98>5A6QoAg1%f<>_{AE&dXK~V_Ia&YXE3Bgo)Ea{C!j{FjAn7UIL)w*vK z+BWyhoC#m&8W9&*oars}S$|Rw0KWhAbvEp-lqkXo=4bP?gr?E-FCe2%tRT{FW4RzP zXE!zQH{!o~`xlj~1I%W-NI10@_%sE5M~3aOux zhrcakt;}QBDYhSA*@$f11Hi3&7p?8XwCg74Rl(cYa*x=TA;OI1YT=?wy`Vab6FCd* z)7sð#P{jc1*6Bd4`}RV{1%EhyMI12zE9Wy%-52g?VAbPn2U*MvZWsp1vaqNn#g z{VbcxdPSBy&rzm7-?9;I9+yJn;~3wRbIo(2U$C+Z0D8iL^!9qVH7(kjK0Dw8r%s~$ zHG{WPPqH0BI42&G%VvT+KsQO}>|mD`A$+S>XWW5Z+;bx-Y8gL1nTglE>-f;FWJe>} zU=L}9%SD|uQqadZWYDFAZg0t8C_9!LyV-gm=ZE#cBLb1R-y5Ed-c90Q7gJrOVV!YZ;mPI`sT&Titvj_cgo^ec&YcvV{^(F*@NZNqy;jmV43QT--m>Yy2HdpJwv)~x|cX-uX+CrRAI~^_5}KK zmLv=ed;hl``>@a+hrTR_O`HGVQ-nc%e(} z*FEg{W`yjCEJF7xnK;3Sot@)~-`~D4IQp`@QT^J`Y=5gmBAh9y9 zgKdSs9S7o3%dl>q$_woL)>u{goeV>r+R5vm8|Xx!Xk^Osg3L*r@pmu0I!&X_AEW4BEDz$-pcKzM$s$wk^wyC)S^~hH z&GWYC%go!KL}#0Ks}^Kom)nCtlK?WaYkOal&LvfK0TzznPB!T_mtVq|1{*oqka|YX4^!v*9Q>@3QBY`Zc8)%G$cSca^GfpW%kZmK4DUXn`Zoro$X#eJ~?_yJbOT?j|z4?@XrnVb8BCUO>>)s@%z85c_jv# z?M2&Yx!wH_VC&he*Z}~363I|?&mag;>Mk7b#Z(7-K@w#IUI2j*j4W>PK(IK@>riFb zAYgC0IYD=`Q_L$?Xz*bKD}EadR}F-Pm$cS@)o-Qk2n1=0yEL1HC;6&}Ggn{~8e=8BAv+EN2?{n@xzg#>UP~bN7tkN0r#Wo6!56%q5|3aUihN8G`^2x4hn`l_1s{6)!p#r!2d=AeF_3oEugW%U(( zTkC7=)~`M4y~hY8G?u;0MG0_2DBK^GM?@L5dzmSDS{^h<)|(t7e{=1kMc1Z~7}t{v z1Ypz-->yh@qX+#9(bB3;T%tQx(YnqE{s4VrgT#;FR{(%X=n>`HTC^@7UgQn?#&}&T z^O@eQT}q0c7JhhEic_Jsi=N32DvbX2zF4Z)UD$I1p(Lp0Ef9HyEg&l@(VF-OExwwm zw=hJE-2*E}MgICQ_8`jQqN}Y|8-mau{w^W_9FB|JmHJT06<~}9D2-q@s)_MjF_*oC zIWH#Erg$k|YHP7`x>DWIHfu}0pdO^rUS!2!iGEcTKm9n?Ze1taf=rvHgS=Ry718Hy zi#ZA8>90;%B}|b-J#}_1#kk9|dz4J1$SG{#>x-6qo07`aK^2f>GNlH@Fk5XjCy*eRz<`{CnqQEX8dnawj zldRR7m-eKh%Dx_}r`eN*iO{>NXP9Y+@D#-}!4d4+2~j-|CQIR0trZHB#_l=&YC5-U z?e+H_8^r#mEojz%8~SWJA1R#v2hC&$s;xNq6H;gv3e(uW7wV_a2(nH4;$TI7W2935>g?s8K?ujeWL0+b1W<_m z4u)x08eea3+%M@fuj`@+if#dIl~DXN>I**N>12N7fyxE)(8j0-OmSw zPvE1zWaGy@e_+|3yZTug-YXl~J%rdEjzxrnil%elgaFT`i}J%Yp_)1gMn^rto8suE zgQYt@R=e`62Gk=T_n#5>?y64!U2*=5F}Wc+8%f^n90Mg|sWrN;8!KyKVK#@G2iZ5> z^ZPQ$sAI6rMcN}{BifH%!)LelSs}r)9l6$+8d!7>%NOLVhtCfkUuhG@oLcF!frf z*mGpMeap6k5gpb~fU}@8DnNl+y|Ns^;t|uQ97{!gpbK6-R>jO3=uC-X)!|Zac$R%& zww%{*(d6Roys*lipySBhytwEGgWC2W2T!@p#Mc=R@>;0aNsCtS(-(3RsFhj@I36ENie(W$F4s+DQv8(DmK@-x}8DkRx zK~QAa2jmC04m3LxLP}>lD(DK`4QKrsSW@gKH}zX`v*O>wDI84#x5ZvLhBYD0ZmhP1 zxr)1_Dt(QcaD~!&pG@!wPCAnHw?N%M((y}4GWV?ZZr6Dm)oI$4vC+|6+C>f*)HJ>} z+`-S;17g7YX7;Tlqx!NHPHfkx~T>{wTbsi8+@iH%BW z#hf;_3%;?iY;YuroU>d_<>D!;_8}9b@qOoZj!otg>vfw$5Tov$+^&LnFV{R`@?KHz zc3bPWHy+;5$t{T~*|_S7N9;FE>Qze|{tMhEe{Gy7G*nUkU&7{$1lu~RBA~z3sg85;Y9)FgbDHg_w1i9;XHBPHN zQVtc~cu_12>j2o_to?kN&AmP^h^9I|Sg`&XWIt|Q$>-tWP5j8jy)07wA);O}{TKYc zDHZWh?&%djL&K!S*CzUES=;_<;9^bujydXQg}(a$!fnJ>S!5?4^Tq+x8R6sd;#k~k zfr|h|bfMu5xQ_=_7ntN5-NdVe_nm{J2Yi*uk?^grvg&Up{4&f9iT!B!VZX&HQZ0)< zXEhmRqc=Hg0zvq;DB@Jt$HGx?u~loHk!+m1L^(IU z9qlxuenoAHD3wLl9lXcxzjHleVNoN6l}79tI`uCcC37ey@R3yA8swIy7f~-aYU_4o z+HCFK?LXKR;BToYzch0_+siw)O7U?&DxS`zM)7-pc;F8&5huuio7l#e>a$Q@;Xd?w z`-ObJfn^rRAE_&9Z++PnJ*!=lWuT6`S3LH#Yr4vgk}G$bj;rPr_n95No~C2TFG3GJ zsfudP5QORXlviO*{Pd7iC`X&B@;cO2$#pGDFz991FvPDQ0HTs(bmz~*#OgHjL(-4q z)hlO{Blh}mYhxuX-SU0Ek(3tSXHx{(v*hr@?0OUPN*5!U_g27FxmB62>Q*bWmV-z# z-pa^6i7vU#%l9Jbwx+G||BhR{BzEEGAM{e6KR-wOPKQg_bFdbhg*T=xz5%JQ znPX1-yPdKp&IlMsevJY?R+-$rDLs%fvmvGi<^d-GuIh2Z>32oaqpusY9)EYLY|~94 ziy_X>I(2?$ge2}9w;_=vetWZACq&<-(Q|>|czmXfqj{0S+?+LCcHEC>eL}eVir3SFy z9hWbH_X16tsW>Cmj+j5XY#+T__pZG;PnLeYiAp()>gcWScD;A07&&WIxDQWLSNcx2hSW2Mdg3H2DRoI)^mLJWP=21BlJI~Y` zj&D-~kxxdXTvU5$og<_p2^M->cd-{$wkj&BTB8~XP7F3P=Ak>zmOIp)V`3fp@zmHJimQ+V zJ0_D$wft0x54Q9FoNLmWlKMQdcHAiJ!;XSN1PYvYK#wxv$)|+>+)W+>1idz@*)o${ zt@dz4FTHZH-Lyx)ZbYtPTRBLJ^-*Kpc+shL5tNee-_O5@|2CLZ$xBi(oJUUGe;PIya`a`bLhCRma1e(7i>|T1BVWt}?CcG^KZ0xPv)DdR^Mqd}{8MJ_ZHouM8nhuH^m= z=HL{F*dFLao@|#GZN9m(ir6nWEt(Q2*&`xsn)Fc=!47%|L_9Nh$`Cyu?9*b(QYZ&%ULiv%$=d>g}O}=kIa*<&wYZB1&Tm7%WQfk5>RKV&ps$t7P^+ zxLKg;Oz_oFCz*n3btV}WHLlHBDa&H}7Jhtzp9@7{!ww_8YJmZhO`OoV5@lGA=}{Xc z0}`?SJ4caNVl=wEm6Y+unVOK=zjH#GlZWOD&kfjxCLLpiBh2j$W1e3aSnWDf-*WOvW-wx-xAdl3 zLquui`b!qIy_xF87URy6$`{@;0mrKq&?60%kEvrNoQmN-%B`#HvZK{!ACxybC)ISF zr6$5Vlc0l<4vzwwWPOAJN$w25gR>P{Tobnrz3dI zaph3ToS(j+hE3o6BdyT57Ov#jn@T#w3(#)I{;ZuqzlXnEz7}`cdy&Wr{W2JmN3omq ze~Q)Yyd2?&E$@2CHzwPU4$PnXpHWxV{~znmV$}N+`X*;`duC0j?OX)Q0PJ#1|DqzJ z7tVfelMN*mozF zt+fgcr9aIo1K2P-N0r8MNgx}bNt$29(gEIPiq(Q)_Q}PVD3I8sW1Ff$>AP6CTTXK> zMzCjR^4-W0tYFprnAz0HY>>;#MIK07xoajE4f`O}^sWnkbIVG0WBz2cs5wGr-3=bO zU1MvL-&i?)_pQC*c`p}6<);^&)mPKWY(W5Uy=ZNCkR35>rRPUZ=}JO!PVG1oj~Nivtcg`!B); zm2myt?aa>e0LO4@jSFi|5>sXcs$DO9WFK9-znIvG_>II;92se0+G$) zTtD_8i4V)(u&^~F;RFAj3G;Al$oJhn58w1R`|cS7M_QVZ*7<9979i@E7%UuThyJov z_orEKb4Ui1)z5v$O9o%edc0(wq3LrVwKE&uu%X&Y!9#<8HBWhccpZg+uaCE=#R=Wg58#0Rke z=ol4>@%kJc#cpYCCmr!D0*SFPbLsR5CXmtpN7i@8Q~mevmrivwopUm>oh3;{Hkl`m zaVQFzh2mtBLpYQaCwt2}Dka&Q>^fGFy^cMykCAb3aDGSk{r%qe=llDshe!XtU+1~5 z=XG6A+XPSx$o)-mnidMJ76DG%xE#e6rW+QydIkG_AECt(RQ;(o7P3^2-&j9ZOuGs~ zDtiaLKSzF9F6nBamtvr66x&QV^L`?{8Ai@Eyxx^B7Z^3twWMYOk90!RmQu8bvP

  • d&`ZHTnDKCaw3IZ!WsmPCi~^TyyN>F!sBXWcHgu1jmP7TQIGj1J6&caK4mlEDaNi zlUAP`#(&}|Vp~(GHL!@U?M|mOP$wHOBqrb9A;|#)g)jT`gSb zZoe@nY&TmumsGhbGfTeFayO~L2TSAb_t+4 zI|UGjYO34mtN>Fg(%d~f%TbOeQmucs`=Fg~&Yeh~=-Ym07ydU1;zrlgL|-P1SZi!q zJbW3-HFM=>Y{fK;pyZRDFx>%XByP@1?g8EJBB_Q zY1*CaXY-H^cOzU(f$d?ZOeZ;WOU#SPy9QM3aUlaS!zC_%Bum%n|Fbe;D3Ev=zcQkE zl6L6dtXS>_+(>i!or)^!Cz@pHz%-y9IirpS@?wuLKT^IOD*~wL>8V&+nq@xc_u3Bn zOkH3wEcl^7ml5GntY-``sz+TOq5&nG%!)uP=Mut{1tv#b24R;$8jldQ)1}^v?)%Ql*P%$z!tx@^M=$gizh-vwdi2ra z-I##a>RCJ9`_LR6=UW?yu$%FWUmIflK>@PQ$2{5*?3RD|SxdOqFf3(2sK5{NCf3u2 zXzyX@vTv~7HCSaW7%6TwmLBU_S|%1AXWGrXbf0JLC#lA2a3_~0*NCHJghrv~gRr84 zAUT$e_M30A?7z0$DlVU*u$uENZcm>ae+#Bj z`~@e2sN1brIm$rPLGDmGD#!6agrCqn~s zt(nWIsvz@cuA6PQ`C`ai`!AB-!II#`pmvWzDUS)a<)QUh%Wwj}y?mcidYA^}8FChO*Ra zr9@GC_OXpA%--NK&AX39$_x2vr_d#|{K$}H-5b8)o_crhPxqO9`z@L(EyF#hlF8|w zMUEbre&M)|X(TbDy_5ITum76C|GCU8mjZg0n@8{3F#$lmEB)O-cvfe&rs^NBx~&Ho z@8=4EP=ZkAyh3DePDqKYf&V+yCtB3723!K}hVYD(isD2_!nxq$gcbF$ZpI%1OHwXJ zGF7$U^)dXCIv>$rVwOH4YQ26JPnYgO+vm_4c}wvZgzsosNbRt?gx>Db0bT%P$RFgOt$1#@ z@nKL$b159}qQXus4Vu3^YD>|}zTcwE!GAI-;y3s59Y1vZXLhYbj-CUws`0lP_)Rmu z?pb$>A=$?k^4gY+1?zdKZ#AP{i#biuF!!^RNd@rq?L0>*wI3c>(rv{@-+42VuQ4^k z8{HbXqGh&Awq}q0Ro(!FB-sOQ zw_$FFG_6kkT=0G~G+qkcTO9br0rd@ng*svJ9}^}9nS3iP>$i$Rwidj$E<&W@Ao^aZ zEFPS4oejf+H-9EW8}&2(qbMl1I2@L1$==JjziUb^WhJmP;tu>K?*aZ%W;<^Wl3 z0R0?wrVG3`EJNlIX?nU4MvW(}Bzn?K3KBGk@= zU^^Q=tiGD1l{d%8*cCC;*Pe|naF(9sqxZi?jEz4TGlq|7t7!D$7i}ojsrodvZ)@R} zF9Jpvq{6UcG7W^y(es-Zx1!R>_36RLGaA>wvo(^V&DaHPDK=tJ>6rr?!4`X#H|pQS z?a4QAxSF*VE=~q>m<_XfHymquW*idQ@yG+bhO!(D$h>YH&RFa`^hLvtC8fP1Djy&x z?`@W1h8Q*Mk4@5s=7q1d5?8lde>pf=Fnf==Ih7205%CQAj{v{qtqUES*Tg}F!Hq?eP;D;+*>q3n1kY~n&baL3{&t&ah4r$=3`)`17IMBLOP8iyI+aH=SopqU zxL;Yb|D}dlyE^5w8#6ihZpiwVMeNo~xJQDwAz1`Sr0}TPV$?79H4H((JB9Bx zI2J8<0)*M7=9c}-Yr~qa+3dDsTdpuq=1aS3icWt`sL;iOD?c`HALx<-CQGcVy=NH) zH3H`aJ!3Yi9r3yAH>tNnx@Le@PH7$2r2oeYz>a_NzOUAzuANp^b=lCqY`uS zayiS})C#lKco5byGw<-N03QhmLU+jwKIga>Ma)&ib#yxe)@!U~K3&kIgL)Ep(q6Tb5 z-X$JRk1k*3v3ftTm2lI<{*Ut1Juk#CLky(_m6UXOW#$x)>Bl$d?S46^R#42T74-4^ z!SCD+T$}aLc4Ov>DX%^7UrX8S9;jQe zo=&b5RqKDxD7D+g3R>Hy@(VH_>_%p{$j>GH{xe&Sr!%zP^zf;*uaPBcqVXiQ29*hq1T@I75hkOWgPPkF>3aoeYdWI~9gX0$Z`eoX55{+hRH(x0 zp(Jf-Y;Ufe7O#7DdU3xijEdbFJe5pSS|xrE!KC=9C~;%EA_goG&Wg9{w_wo>F^V^G zX=q{6d@XExxW7qnZOpfsnIN6z|WL$5bw7WANld-M% z{%Gb3wX)$h%8%J@_iC{n!EgTxc+9#0St35uQI=5NiDOW06^n~dep|YW7SLgQjXjaF z^q|3czm>g@j^W2zPDwtj1S|fU=lyyfa%?wMLvW>mnaM4 z1vx(dlKCqjDh~vr6ST^;E@(bS+foy~h>_=ur*}2eyf#0l^B2hObDy?s+Yfw^o^ek{ z{0!U?K|?s?G@C>#FB^6>FB~E00d7zvenwR^a|(-ji(C_lB>%Y7(&=_3=xsBadCFXh zvRNE-iBJIdPkDVdY7B2uU%)2)^z6D2e_b3R!DMY(@On1WmdQA{)6NvE2v3cSv;48U>pX6adR((itU=X2x#eeBlGeg(PT6OiF?859a6)!8 zJ?VG5*5H^}t*)@(W`fOjx-S?xXk(JKyLo<>jH2aQR+f(SmAVfTXAWAq!dtVaTu#C@lv za!4Z#c|D$`91Ga)B)89LCGJ_ac38`b4&88QXR)l6D&A;K8*4k*Y5Ir2kP)ln9EUx@ znY**ef8Tys6#xe!Wa0bS$wj9yg;Ax|tg=8dn190z{T#hqPQPwscvh3FuqRg$ZfdS! z!<_CMId?@=6ZY63>co5;|3z*K*A4DA2Mh;{H}YcjU>!ZGdG%>{vt+5t1O7XWVesz0 zLY?9VZu#L4I+4|HhAJX26Wn@ympsglf8p7!IHDofVv8Dse>D92o-7smh0n}Dm(Rgo znMAg{Mx^dUvc}pEzqvuCSNbenSau`IUI^YPN-(h}h@WL_NQAW_rwBEf(pv0U)QW?G z@cS)x4o}2RuBkCO^aor=753JqGoNnC9V%$Q_DgX0wnK7{Q}$XFzL>)E{e%IyX|{x6 zq+-((L5Cp4k_Q=X=AOS{ERvEhc`ubaOe zq0g__i+wJzgz#h2YsxMbT(pWZd$lilE(721saeHu-?c@-ALQlB;B+{xBxf`jL~`(5 zh~@UN`-0y&vPx36YTevxa&LOENFN7mE8&3-{&6Ev>akUt>@4?R13F&lZD^;a4D*t&cc1Wk#74wvKxfAf31g;o>vz^; zGHlHJ%z7>QRHs{yph7)AGzssTkB>#7F1HBnKsr&d=xX!(D(v;uNOf4e$yv{^gbFI- zuxNJPB}oCMJGWTizCR`yd3)G)I*FMj=v--09PncF$zs)78!gyvTJWHObrLp zE-VHyTMdb>o08osQwKfwf$|Hw1|v95gI&(;xvV66^*s2c$xpqB#a4@DW_eQQHv0r> zRI-g{w02jJfz#o?+P4l@2%(-gLWps_$boM0C+fEn3 z8C;Og#vj#>8c8L&6 zWWPn0Djn%-G%1zv)m-_SF{l6>7H68owF90;BW|*D7;d&|dguO*PVLqvwdl_6JuIOn z)x=kf{xbuQs2>tt#v0uM|Hdcx*%whuSMzgkt2(M*=>(7FsKo;cEGfi;d7`AT5AT2UW7f2FS)&1>oasEhbi(30<4NuLGMAB;X~jt6GOw4 zW4?yHAT~rQHY?P-M0&}%TDOP`0FQiNu0vYg7KHN48(;@v976`>SQ0V6@e0b zvI$3gX79FeP{GfH)yke2bLa2=TIm%;Uv>#tU1`R1vdtQhs?8}PGXl#bu_4eIOYR-2 zADxO>>JB6K)X?s3+}Jt!O`$7#2C}KFpzhnXi!%bx2JNM9wk#308N85(+;8VAfxe3I z%8T8bLV#yOT=sK~C&(=oUL!XglCLwOwXV7XYcdGWK!A%F4KE3;b`1C+$#WLT%1YSIN!)ZD2QQh zLTc{bPqvg7@ukUd`;WJ?&#eWjmUbln7ZvodC;C605~hSBDu}{FHIiDIQOQ-&u2V93 zmqX+WU2{kevxal4ItWWK7ep?;0bU{`mWE2_S__{&35UE;PpH>*ZC=ce#44yObOe0@ z;x+E3b8%xC#}m1kGbzYd8UQ&5gcSr;1-`*_>(Bl7L+K|qBmg7E4~8#4N>2@VHd(n8 zW>m5$&kRPocC3vtbhTEIJ&%iW1)t3(Vs0zfX&GaVa4n`%cNc6EhoFWP zVv5-+V=a*OK*TNlzq_7fwnI43z+2nlZzC(shy!LVsHUgU$a9X8Kjg(TktY#MIZ>z| z;AvZeim~1v_nUO#F7#yYseQC}79H*fr6bHhe}crl5McPji8%B*flxw#0`y14Jo5B* zSbApfN2Eqajg6iXV-i+_tU@Vdmt%_N(&VEHYL>@Uez9 zgn{D)Zx1e9^Ld@Tkj{oOZ1Ma$c0+1VJ0#9i_mD4ZtawZ77ty41*)F7q?qzqZbD1m_ zWn_`f1B3mdm=izNY*vC&b{~JkEqHe^zzfKJrGm^FU9yv0Qxk1YFv4pnJW8%pgB--K zbdtYV>Gsl7Kbp-{Dhkvo7@Ik9Z|}?!71_2dRe%pqFQRfK!iIS3(Ogzs87xH5b^uRF zHNG^*ayTYCD{3fC(ib`_7RTq!1-PpEoKM2H%gO7+e8#r{52uyMVC2c*wqm+><#7v6 zzw$F9gOdL-&}2NCvo^!NR}uNuZTT<8v3#8=Y8ugm$lotr`fIt?B1p84g60flM4kqz z8RVusQ9g1ho>6KFfi3~^j_#L2_rfo!0=4#@)<$JJzmazKW#(@XML2*kL2qLRa5pV~ zH}A?-)$}jur^o8)Vb|N=(j!e@7I{;jv^HPPdxu*wf2+$9Yln?$Vb}>4?KqNv4y<-Y za!tAZ$u8CA42ujyF|(-3F`*2^W(?V_U&e;U?`#hsU2&0)GBLIM4)pbx_o44orT!dB zj_oB?#yqrbLEo_fQ*Pl%tG7z3Z+6e>GVjFbbBZ zEw3v&R%vj3kWPK@}Q4sy}*!0VEdc)@bH(oeE8H;a+_lo_zf4i4fmWVUTOWokUs5gj81S&!Z_1+s1 z#B!#(>-PKUyKiVp&4M|<>^_$ zIJkRlRW*h`-vHZtb`QrY`G85i5d4XJ{!W7JcY|61ETi|53VlA%3xG=UwP1qVf&2$> zwuG220PBGz@t$3nNv&X`I+!4|_j(l}>BKb=oJAiYRrk|7Ec13DHP;Mg%1T*3^cDoj zJBY>7=FA=6k2Y1Q?I-vK-AuUfYMT1Q$w{fe;@zeMQdS@WQ)ObflxpYaqHUkTxQ6X8 z45|3hxXmpGQ^=!1&3D-OtcwP(`4ji=?-Tf+gp5mDWmxPl7_uLVcs`|U8<`JmEJMmI@~6q4z^DL2lU)jW@kZ8LZ8b3^~tMb7=_QJmI)U z!zZ^o{~fBu9}?=$2`{ow$UhigEn9d@o|CA2q6FJ7WP$LffS zY5T`)XQLy)8XsbxKIuIvqS15evPM6H#zST41=;~#Ogj)vjdpnpEehEucp^nAgq5|5 zzss;iL*c_gI3r5P2v*7=t-NVjsujz;2E=okH^-P~RGemas6uX;R`%{8xjiKm{X-<{ z{T(1Je;B_*42&PQ#=iC1c9)4L ziEHT14e?sJ$1aMDnj~-#)Z^5fLWgKjk@wf0ZR$J#x40V`jJ-N-zrjuo#f) zPNdjpt@YqWX6VY}xdQy=us;p+sU>})5#`x_=?zb&_K{gq6F*2x^HM8`k6Qm*x5w}7 zdpP>}I!H?VEByLcB8JtT8%qQn**&z1sXCVu$AN<0lBFiq76UX0&H9?J(dlYoo#3UQ z2sG?#d$aH~-nxbsVjaI7=^48hxmr+oUPeuw#hW6>w9HT?C-wjQ3sgH)`VPu z*hdhP1z#Y#ew3J2vc45B{6sjJ5wO|KAtLv)k7d^&-I$>1t&hrE={L`x^F>S;kAycr zX*er#r;N8UO974AN-<)l9U@256CcX-<*8H+;bOuuOzwlz3>`y~+H*htGYy?((UJmmP(QY5j^2e4v7Cr}}-KgD$OW4(3P4 z00?6foeJ)w<;b10)mq!Jz6*zljoQY@uCRjU`p3pIdv-h?3D4IU#)df9XnABxVXqZ! z;J1gOiPNbU_8}_=^SXZI2-+c4=hH}on-#ZJyQSC&3Ylnj@jSt0gL8JEz0~xviItU9 zFWD_+2vui&Me*TOMzw9b7v}Z4e7Vky#N#h{(>78v%N3ZA8S;O(GG~}(Uob**wo^y{ z=idV9c=iQ2tT3M zvU}NS@mFF8gx+56ygaW8!}MA2%ZeTmz(m~1{arR?;Y}$CRqN8a03!62@M9zP3|s{O z&_b3I%kO%PA|1_dhBmwESWAjNOIz=rO(>5I2xj2@-uDzILA<~VRMz_TgI$Vcz~gB- zycyTi97gVPg@IqJVg|hO4kX_lsX=t~mnOHrtP=nkS&2c>KKy7EVJb#b0GY-Q#3288 zj|vMFH*&!r35Pd$T34*I)Wl=DeF$xkS&v@yAK$s-ABc>~oNxAHz6XCQOf~J^N#<%X?E+5B zR5pJyxRb*n0PG1zZ@rdtNZM}(Q2bltJ>Td2Df~jTjc)c0_@=p#(=YqGeYtyq)4$cO z;cQ_FPh2z335^%l&$)eT<2w^UB_h0E2x-H=Lx7KKj`M>MbYP{0AeXk~zN~G#M`}PF zPFu}BK3xr(c%g^HU(5be67<^A4Iot=HH@ffxWw%2VwNmm@SL3wYi;dvKGWI~;+nu_ zm+1M_%x|cMzCch2 zd|4MD<*eU-732++rX+3q@$ePYH=-JEWnKHjgNyGuvD7L3l5bs#Y$7=NY&Ipdiea)U z<-+9_3Q>Jty&)NQNNu~}i@L(uM5smn4ssChvREo<7xjq`TJyNsmts_eao0I>)GBFL z)mWP?i7#@lj)meVUPDd*vN06QYcQWk8q8+)>X6JwaED3l-j9 z6c|~Iz!Lflc3~rkkf?m2`P#`ldYbzw-I`_rY$Bq+&26cc-(Sv-U0iAAR5Ip)lqYXpwWOvr(GzflK!n$MnHq+whtqH0}HaD+oSq zTRUXdyXNCf*hRU7{5cuTDr6&gm(-}+kB{4KN^JA{XYPDiH*pAlI~)Ca`!^&R#1M2C z|551E2wL?*zvhB3q_~&1FL}dq2N`T#5oNBZ6{0zfUh0UN3wkK56gh1mjReat=&)EW z+GM30DgNf17*7Za$Fx{#7^~(lUU5LvvjQboM>j3JeE2 zQRC1$C0~X&v$HWha+cSs+c%wdIjANk(9d3o+WtS@kIG}it}E44o+sEd*$DF;PV%RP z`C#;MpAqXal=oKs;{m!25ntMgeDc?()8!_?!`d=n5;G;)Ur7V523wwWxQEMozfclx9Tm;%+-G*czt(VI_AY=9Oy5BHtpz`ZHRP*eItI$x+BeJHp{Xs`mu~nBIvu37 zs}ED-U;)vR*h1bRNBeJnhn%%G(rL)kPs56UH0Th+ z91~~+rwVYYdcJ;ArIQ52T3#1*dF`-$aQ@$Xd1$?;n|1S(7lf|41mcjbbW1JIFJiqhQLXYM8-*ZP(E0y&1_UK&GAv^jKwo zh5+|qtR2|*Imh~fDg2zgx819Nv_-+(LqtuSX(^Sye%kU;Y40}@ed#(hQiSdMz=Hc5 zj<+kFXtZcx!uI5z*~`4m`%8o?Y<63r?nBJas27-xEvzlsKtB_!0_H4e0xr%!bOC|5 zF7-}^Al&a@{qdvx(pGcZEcU^W0~K0|9^LtiLyID(E{9cJS(!G6p!nNxC+n>1eY~sT zJ=BrWt`93>_d{*tg{bF`PMw=tA`NsU@d9hb-(3gU{GA7Mfu6i9@dO(J!>(uc6UuMy zzJ`T5mts&gDqHe#!4slyqRne;;D4A1|2miQ>2N>z2}Jytu+MoCm`#$=bKBo{v(~bH zF$ImXWwy@~gPULUTI%}ptcAPfHuLSx4~BOvnj|kOx!#>Ln>@iRETicaD~Eq8jq4ZY zjKmm}LhB+Z2kIN$Xh>cd^@~}HmoNIT<=O|HB60A9l5(qng7 zC30~!6m1Y%*_ptHVXIp|J_+|N6h^~xFd zXhZgQy>aL5R8r}|nJ6FjKuAFj>mGW3m(Msm)wN*$*nsw*)lNXQ z?_+hphN~nsG8bPT@#!UlQ5@a*2gNqzE+0v^@LuAI#Kcb$J6`WwZyIsBUOrsWa?H3Mp@%cBB2BNP8 zCW~hY=_?_B2Z8@_!98q0K!@Q}aPO1OzrqR1lye~-Rs6nn@l1{}W2S=|MkInen(d$h z{&xxTid!Q5G13ZOe|6}n=qJ z0y=5uJu-8Ol{#(yRdAEeNTyP?BFjY-imLC%6Jv*2-pH_JyxE|A;d8@4?_N>LK@r9w z--!{Zmahy=1%r4tErXG4tE3jZrC&5Z`Gg2tg62RlUiGg!{+4^*=|dEr7Ab=Cd9o@46WX%Hn_Gl?E) zxfZv+GA%hZ_-4$(&wMKW0Eivn;u{x!jkqqt@K~)qSZvuiUluxXuYcBWk)&dU^z6zx zRU*~{h$*Qx8>*jeWGzwW_{1UIy;L%O(nwe4@{th{Y6e^+qIq1AD3{GwsDZCiNh$}3L!a}@=~29~460OuE8j=y=Tx>b43H8d`}%%5B6w`}0m1^#!B zpvOXi>)l7pAtBOA*$AL0{M_0zyqog0N5J^I#R_~4+Ra!@F6jeEw|d6NKUV`G&Yfg5 z$(uN!WI~xaKzttrL=}zo6w9hJ=~m|V#@-&x9m=I+K6Uhg645%$p8dGZ-anu z0UI@NKyVL-*v>I_zsYnA5~IEPu0fx=mcs>c?t0_7zMS*ZH2q~hrlBd>Xjg47@rsUs zC=W)rIY}uS8?~KkWxx;kTXi>3Q!sLoh0{ebHIN-->JJ(}1%c$>{Lo3$QK`z4ZpZ&y z4D#<)U-H#MT-tF+yX{}y2rB@7-sGKz`VmG;k_|eVJ}pswP3Ayc>i1?DkPjLe)@p&! z8*o|t@{EW;V^NMIM=!)UV#4Hi*zjnEbe^37yKt6fjpmiu>rvt=Dgu=cic_Y?W{q`a zA6H%liZZ-q6x5SD!A6mi*Ym&J95%*p4C3{p^RB*)=tHs zWyf4yBn%c$AjsI+|E5pItv_qqWAuqf?4l=vjQRo)oMFQA?<^SJVyQ7})50U+unpbN zV*`HsoFdCorQ;cP4E80y#EM6CoeTGixC#6BBSAiUMYGlN_|p#tNyBT5pn`F5T?}Wj zI@u0FDK^D>{fwLiwjWJ}c_D8g3-G(;Ki5)n{@E*Lgfr&sj4?I+@Bf;!FC0akWokL~ zL|916yc-NP2m+&&G_ZtutL#c9#Y(J3j zabmx%t8ta7$SOVvw0Wo9V5xgqy)cx2IU*k94U8rmS7N=Jx`9;k(HH*KOQC>KvNn`e6hBB;)Kg>dkGo)?NMjsUj=LWn3o0A{?nB~YKxJ1+rrpkqw2r?Am!OuV8ml@WAgTq{@eC1e+8m+1= zO6B&0gvzt$EoS_1>f^1c@bC_>`k#mMc)NCiUK^h7K8X~^f+yji1N{?6t$Ji%=BI?s zW*~n5T{^yScbImuJ70F{kDT_TmEaMLuZ6zOv@2xJ5l?}#NN4*?0s8c7U{2G* zt7bvoJnN|TO13EGJ43l=$c`Hah8_VfayRp|X12_Z$H@MEu2$?`d8EJAV`UVX08fKQ z^nkj{Tz=H!_yUB_eg{J%fCvQexj^-zA?IUeOXX@gW@T3rnvO$}MIx@8NT z!AkhTeNN<(LwZ@4#aQ)W=HZFeH(TmgY#y9^i>Z|}P;yhW}LlW0IbXktzZbPPj zV7RBQ-kxHhW7S76{#RB+yU-D8`HizIZ9#m-5sl|UNDfS?@X`L|ouGxL(AsautmD42 zxG%>XBil7Q`vdPE(_CN|^&Oe^7ryJJ&u)|n@yD1!@LcmbelXDHjisKOlX9*pYICUD z7G-C5Z`bzKE<}!0=*}1q+z+Tp+4FZLD9GobZ@b|AucJ29%`-iiPm?cVgQTrt{aE74 z++6_k@q<873Y>7r@rz!n=Y?9&3v9P9)<5HAYSgr1o7itsJaAE69&&IP&KJ>uJ)H`O z%P#!+t^Ko|KeyxZoWcs$x0~a}xiUP^BGzw{SHU(ms$~0l((3O@k}a|J=U$x!!iU4J z#c4r{8MX%(k@?`8VD`IxG)r|=QSNod)8jRCYAt151Y0VZdxU3^~`2(TPy|gRO4I<-mnH~~Oph2jT4|)vDd(07VbyWapwYgAE z1CE$Qef59H{0EbQ-BhcqH+|lEZUf+Wyx~EX$Zo3ILA=q?_Qe>9NYq$qH}AWEh!wQ& z{(U`WCVev>k3ZtvC7slmfsR}o*~+c2W_3I?%xQpm242TODL~-i1Xj?m$qZ&X#_)p( z8_;0{!`6d9~u?kiWWmXtDAPOS_a^%3Ia+FacR}$ z7`T&%``2C5Nt*>ux6Ej60gm((lPBcBiV&@wYRB`_8FUR?Y(- zqGZ3(OM|Tg!Y0%W@C^M7*Um<_u9U`!AWQ!X>M`GQc{0k(%6^(#>_^(8u@09qj}?6I z`x^S<`I_6$Wei1$Uk%Ep{ej2^H(^#B8wq0^hA=ODVST>L7tNxH&@e4?+Kk76%I@U< z%orh&7e>2q1KnVh?EXZdMpX|9jVc_C@7i6y45w5NVj<~K3(Fz?}&j3QGIp&LsVt!u??2J z{#)(}W$nP2yAF#^OTdvU8=oziLhVRIhB{`A6H!>xyQbrY^)R5r^px!>xo^H?_`YDC|>Zrw&1XlYx z4QBs_;+jkIsun#bIPZ^u;(7D(Zie3Np*k;49~p}Ul5erh+YmSbap-*;ipf!BeC)RQ zG_>vNP|BO^h2mFP$Y5JljWeso-eb###pog;p; zDtsh?5}k^<+Kk8OU<2lEzTVU)TT$xn6E#F zurGz7l-9>c!*v$2gn%{~vv=L-?PYvn<<-KfkxaAr(Mj>qigv?-toaP3bvc z0zB_Ot^T!xX-Dy3Exg^_XRVgPx8E9;-WY1Ma1c?hvWJsW@Nt*B@g;6_hv0@)fb^VSO=o3H*nQ##Ik<4 z)dqr3Uy(g}{9zYCP{+vj*C(Rn$zk)h^RyofyEYu1@^y=Gt%!4Nr>0UZ4e+p5eJ`vy zl^@&@DEk0ToBu zzT8yf1$OYGK=&)JmISKd4*hXpOxnWAOK5cZ$vZ2mm&gHgy5wrYb1S~xvqCzsacBl; zf)X}=H;~DeO!+!7Q$$gDY~%mb{<+$h%8&I~c?_TBiXmi3;WJ!bC~g}M7D( zG@Bj5V;7;r0}Ro~qUZIiQG%eo#s~h>)CBWm8#fGchi=5vlDlE>ggLqO4F>yV)o3YV zNm9HIh_LFUF*aTmeWR4h)=KYY<~gS8lU zrP92zF7oJ*KioNwMACV(vL+9HI3jWBc2U&vpn!w$0rh1V2zS5djv1&-yb6wDX{ySD zOEF!&Lsh-Rgx&4T^)jlEV(%#;H}N+OY(Hv>}yL zN(Ukex*-uM?ZPGJ(DQ1f@}s}xU{0d6O6`MpU!}LIXwS^IdQP=43HyRgefX3|UqLBwjqS^I7$4IbouqNm$PIj+ z)iMwE&mDY}ToX)Ls?Q*Y&dN-6ddudFOZt$EGxUAzb)@=1`?ENh1tZMWwv~N@SL36| zsoqbG&*oV4Sw-lADzi$T0S~gEIPwagKfRbnzhJ!L5me$jDW%FtRpY+8ngJUo)*~e= zR1Oz%!kEB6%nc6_XZ^#Uy8Z$M%kaqtCSPG{R)_z*+#Dao9&oBa-y30A+VaHOS&j%A z1L`#is>OVYlmRvQOEz-=sz3M~w}>d`d6y8jYFa|wJyUU|(@Q@k`+%$rh*?r&wMH=M z9$pxhlH06`N}~EY9+%SZmAi z#`_NJpOam^fH4oO3$2LzuP zD`qX7ZGm0!qz!UR@L^YP88XA;r_A_(wGTwY!yqrBtoLWo@H7*xZJy2*H8N*CCwV0t z9NMO8-CxTZUL^JD-aj#tzyJnD64+y+-~TuNN5;{^e#WZ1-B@p9`a+0J_k| zA8c9Yie~K%Su-8*)iLxQ(#0#%{YdBlmCN6#$b=nUbXD3xF0t>wK}F{`haKmeY6XWw z*gsn7NJ~&TihmF5qE*UUPRzg2Ls!BWK)G;^8xu}ycEN>t-y9llojvT6oNX-d7r%k zHuw5Lw^On8*?^QOvCU_MBjU+W-w1j$UcCpBz#1{M8ODXnz6rclxU6$8oh z@Bj8oOYDY0U$;Yc^pYlGrcEd#P|5d(#ZLo;o$2V%eD)Rz2GF#n#y7OPzTrhHHXi2m z5A_h^5OK*KGbtY9OI`Qw_327{I|FNzm@d>P&v^Bx_+R?m|0&et|124=jzsau1XrQH z!nh&nL3N^_OUmW^w}U2&Ks)p3eV`CL^Z8aZ>blxE0{R*md4QLy=Gh*-J05m$!Sj8| zI%9=95<~(X{fu^SAOK1~q^AeTec1~2%TX|+nYq49DFModSNAOfLvWLeT3c#>neiUc zL^MAN5fujfam%Z_@(S3Z_|eT5?^9IBuR_F;K>-)6roU7Q9iITv0B^eq7mEXebXvFT z&->;V2x|bA?D~zCz|@`GTK>8dgWPfgp?Qq6C-lK>3qi~Lv*6Aqi$7T&_hpa2R1_D? zcDO!PY&EvH$X$0{{2(-=B!S2o-aI9Ntt+=Z=^$CIezs=~D4jjx zo4pf}qo&jyRkhF{vMHVV@3Ae zYy!=ZIMf){3fj9AI_~lH^aO0|kPO1Vlas|5-vXZ>(X$Y{E6DEtGh0VXbwQBaz2_*> z$*sDc73J8XfFSUx@M-h>`=r*t={jMDX3ybyx*50rW<7hB{2?DiI5%^!ZF*!Z2#wZO ze)Pf`dOOrvAh!S z46msTPD$N#Y5_J2AARbdk6M5k+3t(1z2Y~$`v1uK(s(HMzkf7pB+CpT>zEl-wiL>~ zO~w)`=S21`k*$)EouP2DjTS|gq*V5GvaeanzReiR*g|H;ZZHP_(K)~S{-59d$SWT7 z;Co%)>+^Z9)>A)G$5~QVmuTgHXqXM}Bkbe&;jTl&Vf`jUX5#YBhl-8$m!ky-kCKLqKWUxtcm)<>Ox ze~%&?Pah3pCmp)~Zb!rW0{}ly3mOh^I@Nq5(+q_J$fE0jh(EHT^vUey+;F*Oh7}Cu zcG88tF<)&v@EZGdm|fD$rY2^`jHIWt+X3(7XgJs^5@S#P9ugfp8*A*DHI3?Gl>;5n z>SxQ{nX6IFK^$*DpVy8Wfb^VTWguVaH(@EQW^*ro;AuG$`;|pvk^XM^7fH#8oQi%96#GJu%p{XQdM;0 zA_K<_kU<|5&qPXaosIAP5sa%{SO;2~eceJUJ=uynSQReGxFEmo9uwMWJGy=~bbVb3 z-|G+EE;I9ZI>I%7dM>!ev^&;_%4R&LiBT^pluEMFl&Z?_rQ9Xu-oic!SF&J>DfcsE z6ZhmSIeeZ;chOrfOK0$4ny@}v(~7=f&`4XWFp(BvCgYe`Nv%96(N}JNZISGpX+f=$ z&-MQI$?ZimCW$?ql==I=Cy6KF>7Wh_K*cA;a$C>NUmxPgt{xX%peKk?*5P83G^dEy^C^eUpSrUBQ#Kknlb zynU|EAnr`lCSk37g4$+|3-8a(45eA(O#qznr%dABoNJE_e-H(cYP)gpB(Z;me$LUi zL&NFR#0L`rF_tyfpQEME&+T+R(j5GHeLJBa`A4Y-t<>v+EBur=^_PLm!D}w7h17Ym zz=>|3U=1%%(Oka?@4U^R*ic$j@0_&zj@k_!aEb_h{%{V@^9o+m_*7t!E^m}9c&mLc#$rvXZGWRc5{0(4z z2STJ-Q)oxHJWM)Kup&jD;a#@{?j%!Dq`~-3S#WI94Cj4nsG~hdF zQJgxd6cKc1^?=UYdNo%R`01%ce9Z6MH|Xjl=T+Kug1@Z|U=r{Xk8yZA>%ReWVj1RVT<(eq+5B-o)k?h`l1P0;ZBH;KY%8GXWdooaLg>+k-9 zykv!EWkJ?~Is}&%R-7<*NJu+)zB*Qn| zfN1Sai?IFt3O3v{+wOg2mSBYqlaxFA{q1uGKqkTq@%kDCBAYz$OP z>hP|ijoyldk?-I>!1R7COUrS`*6Oy_5M>LzSn71*^;(pYP-2 z{SOcJyAx_hvO=aw(W=$IE(~CEc z2H=EKxm%~zo~S@Snp>;>n*T^DHNy>}HliOmX-Hsu;@{Kq4s_K!#A6{z@mH1Hx!iXAur1Tmk6TAsZ=UQ&eK zim{B~7yKM+AkK1LL@%NIbe3J@5L$sn!Jt=9zJnoq13XP|`^}>Qv`&q^blXm!s%_+f zvLA1^A>ZTTZ)IfxmzX7i$cV-SzNkn>r(cDHfOrT{S1gomY^n8xj;4!4DT!%PIpE27 z;uQZM7V7Agx^4*-0qDiowQD7{7fjurh?JTgj zzW43%2Jc+QIEJ#}J&YDV9Gs}|Jr{oYMv-~_@l1&upF{hQ89yxS+L5B?Gf3l(`CrXc zF*^>TO6}qf1)r_d`iqloMuXjL-rW7Q60#DMwyWKo2#;k<=#8pvaY`OLVAs*-yV+ax zK`@6ISC#vkEv$HduW2PJnE@Y9hf_ye{JZ!4q_*oQ||K0`CLsH-bGVnXbd$BTj{FVG+hSE_lo})3kt)40kOgDhpmuHxU7ECLX6rsKE_&g!ns!@Q0EVzdQ$@5Yb0ZF8K!9FJ_0CTbGYC2gpyz5nZ!8_ z;NQ+Nn}SZs{9;x%;A}k}U)!UpL9K06ijUEARR!r3TWQM&5WIKB8f{nA9eE^_Kl!@U zj$HRxsS*xFQ&h z(14{5Cvtt~H&kCN=iZt#H!>1LACBffi+>@2G`7JAEIeXGg@}mldMZ`dm|gbi^Q?H_ zl>Y+Ka%+^LdU!^9^h?RW<4Z(*#o(svYdUz)K*w#Z`E9`6ot5Oa2*{mgC+ep>gmA!Re-S}$+x*pD*QgINvOkJ ziJPiU)eyD#)|jtI1MIUGeAjRL+v%Sj(1bNYnL@hMA^~$EK?Ku#InvFaoF9VhQ2mTZ zlpvuxj{z^7KL3q~IDNz;`bAl7ez}-RM!5nbQQ!pg0y(Z5qC%*M3ASE&V>`ek`xU3* zBCa=sPWZ|erz6AJe zp~}zu+zxtwmE3fSW5SjUafSk>dPAvY?afj!8Z_I#pRnZKl_#B+JL*z{)R0qn((oe5 znJ%=~R-ix(elW(o8~mI$_r9|I#$0$HVZW>WaU390-d5wTjcxDuo(;^=k~h!$mN5YY z=S#8ABDdu$RpbL&4WAJ^$IR^j@MpV4%a^T#ojA)B0>lOIYdfzcke&_N?wW1 zO-K7h{+VI?|L1HTnfQ=LGc^+TVK9p^Q(r*=02tPrZZ&n#I5OvHgL5%)7kCkTOo+z9 znjO{`YW0tY8Mu#U62xF6jqrq#l*Dp*PNO6ZaS?FJy_0MD+1WrwJb>5_h2xp`q&c1? z;IL!dDi{d0Ho7=(3pkHjPT?Lfbs+e0rzIbA!#Tf?im!;oq(gf~?XlY2eXiKTXGwbZ z5nU4>HO^Rj2*G=KKfQQ;{e~DDTSv=+C`bvVh?E;J%gdvtSkBNUT=m2ZN_AFik6;=P zByB)%J?h}#UZwfNm;*!-Y{yM4V5{Omo2Jo=cj+ISky6-+=8B9q|C+e?+cm|;i

    6o%k9BWU#}-zsSkcT;eTtzpfX|tQqsyAu z*B3DE7GGmoT5G>ZGBnkwzHk}TdAK;)&@bt#>akZ38yXr4;QmZz$+ziUnv-uK@BXig zm1Yc(Fgjga*2)e4vQ>r6Uy<{2oo@2a;d;YEr2&&Jn|hsIopm@-LM}e@>7HcMx`~&3 z$&^*ARt>mZ(P=%JiKKUpScj6yFdM5GIOFu5$jXy%uYURR9x6SUWB>CP4G3&i_kuele%y;=WzD8Q;FZ{N66Cx za{@jdpv|zKAsQ@L&qkIGLs8|?GvUv#KJ%ZmKa;f~jU}DMU;MuEJugY%| zZ>_K=wi;b-l*?PIUieE4meL^*r3`GxQ)wUv=K0ko@ig8)}V|2etZHa~bwy zi*Nwe(TXQBikFi`4OfLND{*h7VnX=vIuEjz$q_Zf@d{T(I%b<+g}lGL`F~-7(D?c6 zPH3tNgrI2+6Ih>!sN|NFm7!E=9UUDi@G4>ZpOBS9_*nV*`R5netbN!UvoI1mlS5fY zbFKTw=wEIlCz47p&3cyl4epZ`U=EAz%tZ198y~h}r`=UMe)RV1i#z_SL%;`rtO6}Y z$2Ez0OE!~|O)DRQQE*r>baQWJ&LxqZgox+aAL3dVh(&0G*$o~I<~Zri^7PoM7cBnyR5jV%#s!e<>z!`ky*EyjfoW?GHES`w&i zXn?sF7KMeDu7ZuTK3{eL+`fU$egFPFXTx%oggt38t^0b4v93qKHP^VV4_uSO^r&ng ze@xz?{t0dnwS#=+MEa+Z5TCLZB2NM#?d4k_QhtqNVNtT0>rVmdanLX7)?L>B!W%H- zqz|4LJJcMh7#an?w6!IA!7Zj9(D3W=CY(P|Qc}_iq~1@p%3~ZiNxOT?hA6st|Ni}` zC|;jg#R<6KZ>IOn<(h}W=A}j&8zcHEF0WXNi6$p;k!q7*L5Yc03f4dGx2o`K_ZlCT z;J+8x$)8)w-#ORfJW_R4g!0@+Y|vr zJ_4;^z=a?EjeuZ6(ms%hWOGm^|3m@6dRP*9Xn(kM--HMld4a z!95Ud)s8Y?&8r~6X@YCerre^*qP&YxEo|ac4Iw^{P^sjN=nUQvRk?d)1wC|E z^}!`CWn&U%EhHPMikY+kqM8tGi;uBoM7&zIY}qg{j!Lga>vuii3A+}7>4}GY?6@Fr z!*ra~)4Md|uExjatgz?chxiK*`*OJ`xCmhI^*_3~K*y$6MPSwH)q|nLo5Z`L1$}X1 zEKNTm@1mVP5AH|nx(QCozJhrkPY3O4N||5|Rn3bU!DUNev}zSN5LxzdX#jD_W={*{tt4{29f&(S;g5;kw47PDoVi&)aXfm-?`T{G zfY6E@&O5Z^4Mc2ucfk*2Zcnml@bN0GYZ+Y7wu;VL#YKLVG*-gzybN`GkN}M(jc?A| zE%}|W_0VrZ+-ViU#ic>E0AXkGW(=*;x);uK4#WKXVVD8AAA?zfas%v>CL((#Q*r{fx~JM$ zG}U&gy_}bW1Gxz`J1wP(uwvK1jnVS`EGkMDc85Nm?yF{O5W(mPSqzq5P47`03SP2U zCc&`6tt9M4buPSlwpByB*QvKkLV0JKT^8}fKJiL0Nm$~c+4%sr2Zptrd3cstw8Tkl&|v0S zTz1!CE;B0&iY#N_x^=4{V6tzY@OV*CQD$sxYynLDR<5vgDm}`V=Tv!Mt+e5uJ9q8` zmR}q@%1L1mEE=`Pn#E%2b-Hcu&HL-%)P*HB`n`DcHA~!wmw&`0O=1EU1PfZiL5=qI z_9j3JT;Z!%e}!Ew)3xmdwD354c56O#%8=i|V9lo9Ugt{-GamW@6V8vmGkduc5+dhg z1$K6MaeXWYu{2c<^x#&nU7HCeI^qdD(gPH#68rZ*xrp@lA10IYk8v7H1umJh`esimVS0H1p#GPZ7VUcK$Am2YFwn*9Fh5@U z5y`dCuiga8IqJCd!oq@bA<*ih0reHXd)j{G@L|f}bauR530!dh1RWL?6?L95H&(ul z0VdNMa4Xa~iRKN-Zs>;em-VffGz|9`mXy7XSFu|+3s;NIa2yeg%vsnPiVOT^2~eD8 ziA&rf&E!5Os1Aqu7cE&pz#O|?ko=-<5+AViH;zNNnr>@sytTk!Ol$uIDVrPZ#=NKY z#tGnh+ch+7wzN;FL(+1Mp<pbM4hCW|kY-YjC{Sx+fm@r z{Ma%3JGTJwk50;8U{|$u6h%-qN5!3NQ>u!9HcZlKXgi`$yH`YrWL6S>Sz+ zjuj_eTpp~OK+T+MFW4fbe_m;pM$3dM0>^6$PQAZv2>0GsmMe`XEp^)}3jlfahyKf4 zwif(Y&NJ2~fX%?4^xCg+E zEZ6E1=atQ&-HoaDLlaysaP%%D&7mZOvz@{AoFNS7`|EBva*{zfuWtzrd)_0%!&?+X zxzrpb4R6>%9(U%Kajz;qsfv^J0-iSmy!i>!D}lO;?_7N4-Jx7~ez-|JNHe7L3_o(t z!rZ)RV#0R~9ggN45|NU;2)C+VkCNtQ@`AF?RqHO0g?V^n0`~6Sd4x*giohme4r1Ve zW~U2vvR*m7XH9&2>sTHStA_CeEy4qIiuikjMI-gWW?_wujpxCHjtCZB?L-AW$>OKf zTYjnX^=p<&#mbQ1ZQQ@X=bt68U)q`=uW2WYeWKWQ6r9OK1_5?#Bh^w$kLPsvWMZ7<$=?Wn{F>Q zW>|PQJEyIi=`O)3u3$g~>O((ZLo?)3xcur4?{UHMAe!3iV8*LMRwKF>f7t62m;t2( z)}<3q$)2pcckh1iW_Pu<>H7P(`{ypr(ls)G9%y|SR_8WGZyY}|kZ2ryFV!gMt_8-A z2!Ta>Y@x`z1M~FpO-&adBbw(VwO z=gu5pq2uEPRL8$f`qP!1%{egfXR{L*VM(i!S^4}n10X00!xG6V61YSrqKTZ;JsSEO zF+MwX?4Y!CW{>!|yVLuf#R?{{!LPooSh@N#r4i%W!4y2=0GxIsTi@u;zOj}|lt67IFus`k%i3kwSe z)#^CGt8Ga*FL*5?hy zeSFN5;2PQYVijB@`{9nBgrcJ2^yxo}k1Rv!b9TfWwTDO&Rl$55>X<3yqM1XHpbJat z*;MRb`t-0V!OqH%%BIp_;Fx)RZky4CKy<%!bU>JByQL3NfRntBUTiOVS=y2FivL6< z;l5J)G#XFWxNj7sBc54m_;9Xmq` zMl<)LPrq|YCZJoa7!1#Qiw|5$rGuyVirzGNL% z_MDDT?x(m$2qc^qh*JE#Uuod<8J||`dV(il>(k6k#k_7fCuI#KosVX2%N5W5FJ451 zUdFQ#y`l9iH0d4qW5b4exs!{Wpk~5^Aj1*)*GTFANrIOg3MD$X-5pQL8)9GcU4+qD zCEA>fMn#qGQQHFz&81~!(<8x{Y1Bx^jr5^4>F4$=S%)WEAei3Mm#kRNwFfSVr`axo z0Mj%ua7tQ$M!s*503?f;_pL#Y4RWa>2>*4vM+ulIfR^m;VNaLE;c6k3ne`4Oi$LNA zU9{0Vja6u_13E2GlSNT`cBC%_U7HSooWfxIKb-S|f{6(e5{7;G;fo*=l=$>i+<)7P zypo)p+zeG-95DLahKHUw|D-f5WMht2O%l+avk{HbhVV2U&4Ql~nfASNptE83fl1W*CWC+$#J39Qr+1juox z7pL}0uby|Kn!E-yFf+O|^Tcm2zgmuo;&Ke-a26(KAM`U>0tq3@k?OgMBM`A6HMO>= z=Q_mbD-B?rBG1K?#1zJTOWdL{2&ZOY-vb z+Xg)%=Z3)_N|2K~yxWgFg|Y}Vp6*P|U6u;@Co3?ICOH2H{d-pif$&`D5;mQ>h8TB`l>DuU8O?}`Hk94IWhVBoV z`|+OS0o*EZGrE6A4_+(iT$3G_j;3@4RGKYoG7Px=9K(CGl2fZ{QL*5idE9j8&c3YRZmj@m0OLfWi=(O5U3^)39pIAkm($$Tob zEz8(R1`2G2C7_jd0z%5@;GQxfSG>cW2#sHG?7x3;LL_VoDs8y@=#)@&^j((MU)k?f z=1aN_!u$$Gci5-0EjH?J(RkrU6cGIaB1|(Ul>ohjb3nEUNR21N)Rv~ELnWt70Rh^J zvHdenZA5OeoCaxwQ`N-drA7(mE8J?5OlY4%DQf{?c>TgJ<*$j+);x0UNU#s-owq65 zHVZSj>AQb&DkqD15!5EGc(n6%&n- z*N5i97m+8nVeYG~g94&k`XZPjIShe>4ExvQ=*cJcwhf$o^JrJ?yG*{?6k4-s1ypbv+GIayzzRZbp~n2 zZ8mxQ*2UJs3eC5PE&ae8>UW~O`?rjP0&;h&!|i46aE1#K(X2jcTiT-}UE$h6p$8kP zn=`FZyy!bhpsP+jjD|hAyQ?w;odXxC7ecXUQ!m7La&3(u{vXE#5;?&&>Gllm7r**D z0&=kMGj>0H#aihBp?h~HBN{tpa+m@+0_g&C>!ETzrTuNgh7CBnH4-Y^eU7u`A5wCW zBHxTJ0=dd57*KyjAzk`iID5D{1R6JLMVvfY3$yc+4s3h6AD44%2DYO<`Q}0^7uEWp z`m?MCoD#n}fNvO-iOp~eU0pN|bHDim8Z)HzqTMi8m>f-Le}}t4cUN@sDO^jHdSm=J z#_DDp~ZLdY#*l?9nR_d?O_2*iNH~|3w@lyvcR};O0S`OzB_-LQTuG`E?udtx%Hg* zU^mVYxL3?`)a@^DrfaH3K;htoWDInvs?>T0t6ed<76IN4#gZ7MCmlw-?92-J?R-89>_xAoIipg%H==~!k<5l6wd7DS6e;*PJTx4got`t0oTK2_Gg@fjf2a2>$#O8jXQiWtRf#u{jzAx zv{qBaA|%-*_Kh+sjq8ife!UHJ5{JDpzwKN>hg%;CmeuIe?e&h>*{0#d?2ddDzx`~&) zkAAe^Y_z#)(5xDGcc@Y_T(0P6eR*0Umxx1W)R*vtMPh7F@R)&z0;h=!kP^s^*NIh%uxl>Is_Qn{^GP{YK>5Wn)6~Om%EBQc?GPdFfKHYtHFYkFAH5F zMH4X3flyHX_~VZcCv~&FIt-F=_ASzdoT3JFyZL}-LbnioOYk9i`X5~&$%lSlHKA8# z=`2Gvs4Fhuk)xT_vV=Qv!(qi#$fN^Xj2WMT%QFM0+3n|GAJ=-%sjBU*+Re{Ztl;H; zrV86=Os;p7+lc|-fg9}=85&}WOF;JWX7s&L53q=!N@RD;%*uzY*yC+W@9g5y9t zqAi7l*dzpxz7Y5H3?&-76b!Ebv4?RVBTJXY{h@q657;;Y&0?=tJxzaRS{ zFb`r{#G>H=;RoVxvEg`AEX*930*$LiIl!DP6(>9s(oo09$TB%Ep4P>Nj#Ra7InKdk z5|U|`ZGZ4-U(IERj(CA^*!zZ4-4Ax{v_eE$uEaM4&AI)4_V2ti(F0yCNorL z@gqs3x;j7M0o1rfm35l2v2p74P-6W2!GZy|3xt9p#XXrwX5fK>QVJH+Y!Q6+v?25o zB}P<5Gu`v8eVw-IuiWh)W?xIjd30D;krB)40-R#3Y}yJu4(Eab$YfI{(x-E{*6K?b zxR#XUDSs})44n(2vLcF~<60(ag^a3a-GiN&_YsrXFWeq#nPi8?{RI;fOjDzHG}rgK z@}eu_LXD$&i1i4Yv*qT@?!KTc-9viVK%P1!1xX<~+=%n)Q29SQ+&vwjIC3{GmiiA+ zfb`5N1Ug(F+3Q(j(Q zHdET1preoYyg9Uir_l(s4AK@KQUqTe+*~@O!;F~k9CGW!DlGe_!~El5@9${&O!2^0t#hris@_VT-F zooF4L?P|jG;hHMM%kOCmA087IL$&=eKu>FUdW)D+52D z64nG|(}cOtrLJ6r?7#c=G-u9NbK`j@?)}1@U%;dGLA#k#`>xSg?k&VsY~+0V=UyOIrM)#_}6ay4@CTeyZ=DMe;{H>g!=yo5fh6%wYkmu UD3L5n9{9J{?7;3^Q|Ivi16Y1UEdT%j literal 0 HcmV?d00001 diff --git a/doc/pub/BM/html/figures/figN2.pdf b/doc/pub/BM/html/figures/figN2.pdf new file mode 100644 index 0000000000000000000000000000000000000000..ecc4f1267e20401d5a7eecb176966ca21d60bdcc GIT binary patch literal 22919 zcmeIabyQVN6gQ007o=0_(#;Lv64KqB(%lWxNOwqw2qGbf(g+gL(g+xopdc+Fp@7AE zF6#4G@3VYkt?!TTg*9u=nX_kR&))mY-ZS&Nhgm~LjvL0q2V%ba0$kYwf`Or6H%mv5 zhzJ;>VPR+E1%?A9T40Edzq<_>B5mPq;p}Dy783*6xLP9#1%Id@>*p=6?QP+01Lpf$ zNzTpH8_fIlTEoKI+s4xs%#XYUK{Pzwth8;s!A3woGIBusHh$h&ka5W9+vtDUz!7>3l27P6w3x2KJT3&=13hRJXpx4g%V!s>N;TKmd{ z6nctls%WS+-LwR_SoGE;99Y=SQKt(|Ht%9Z zHnXGX1vwo|dTY36y7vP@q%QEwr}@2lZ2UJMw9B84H7|b*X+NIMVSf zo69Di$n=*vBQoXhvY6ij%ek?=d14#k-|HNddY@B%5aV7#O1_B#>5?<7ePQa}5@+~P zGPhkyhmqBR-iw|`uW)Y*7hmRLkG%KlKsr8dekJkITjnc0L3Y|nc5amIx@L|u@YQ=d zM(~n8vG~U0&I`p|1SYnved6(zh=BVVci1y=%$ABHBhMcr%)C~cl|pK+rV9jF^9kvz zv9OS(KMt9=!+BrkeN-+-V;R%yZp#PV6M8AK13AMN>SO9CcuMy)gF6WjDvjzBPZb}UN30$X*L z(|BSVMJ_zg>g8cv<&+~Nr!5Ijb~l#pRr1=u*K-G5rE%WUzxndr^!qgTrMt0k<{mYX z+sP(AF@Q3su0gX5>aygrNnd+$ZPzQxDHqF<;OWk)sn-hNVaKo+N z;WPx+Zbns3eTs)%CK2~83qO!=5TRw@H*C6i6pcS-7jM#yosrUYC6O?Yb>Mt!j?Z)? z4Nr-c!1+`5CisU1YVqsW}v$$F}X!eTaI_qypq9<$NB4<70(Hp@F2TH@W9M*^*8k z)|+u<7v6v7;X#o)r`nL3#A({wJH+%cqA)3fgBx|%i|}c$C(Bl3iKVvzI%1+h)@s(f zkG3xSI@<*Y9jtP6IP|hcjs40yEht>`^>)8k~M=DHT5u4$)!AoPO&k>`G!yI>E=^T&3ZO>^` zB_5S2WOwFYMwcw^J2}s%n0^1e8oJbkP3<)O3su1&6xlG1*sv}olp6l#fJluotox>9 z?9chI!!5zbIZOehF((%BV795w<)sxT>VUi9H@R-0aKN2YGz}#+vXUR6vu{tN=%b-a z`0<3t!U#19GF}!WgtTDJD4f4}{3zCno%3}VtL)wrO((K#f}sVdZvi}~HG z11QObC<&-$o|hv1Hg9J%cw>M$f-EUA9W{E#gX{EX6yEY|WWUDLClzB)?qnvk>Y#{H z_l_fKLGuhH5$nQnth&YIzHlz~e4#U5kmM01Jr{eBr+;nQB(p&x1XO6xQES0 zP$g|UozR3}A`z)0r@d-G%^cJVORvY48>+SNpqt@zmr6WI zZ+~+^gOS7i@_oI^EDR`%tlAi1vlIc$-a>P^H=Ak8!v)P)wJCb>1v!1)_!aZe_upXUFjwivZ^6Uak^!3k>l*nx3~-wVijk@yIGW`N|rB9qsH&| zs!9{F6=TG74U%q2Fb>6wQY2Q!=Q&e~s9i@@)0T=1I7`3 zp<>vMo+bqcm}JDaZn3=SBd}q*38UXNOO8sA+C-rc%u#>Zdpsf9XATQ56@GXZy4q!V zEASS}b6%X)E}7uoo8^O)UG$k{U6z(wPi3gg7rKQ+4R8?&D6IZy@d?~+*6$9TLmOCT z0w^x%DTbvVpsmL;U_cH*`3?6K^<-5{k7dR*CJIPl%@$hA%LTn6@f3W~d6KEb4Iegi z+^xOB^Yg{bQ?8PehrhbrYchwfXW>tU{j~mI(d7mBiSEem7tW&4!`Su_#dCL~r?8ol zH=l3VQ5p00H8rwNR6JkTuM@sZ>h=b2hfq3rj)`x;DPT^EkWWMMW6SCJMTR!F=Xp=d z*PE_HMZ52O26sx3KYT$)r7_D<(kgC^Zma;?^JJ<}n17hiii#V0C&0PoHJ%k)+*~Op z=H=NBsWchZ?tqQPm<=c!n2V6m8s(3dOxqp3K|HGCJ4}ON=Mg z)_b~_iuH`Wq16sDUe}7T+pM++hO8u!2~`>Rt!R)^N!T?q8o!viGj%B-t3d5;!3gOV zPqtY(oh|7llY6;1nvoO3=6uIDJpmm*rf|96G;%8tVCv1sZp*^f|b` zLYpiuN-|C&9%&VJs>;G;FPKp_8!a&CN?34Vavr%U(@|ee{XP;!!tLtBmzSy!D;rmZ zG&vs5Ug`9`o%Otrs8_ywD|z>d748smAR{lcdoaIwH4FPunwpbcr{(9Pd2g8Sh8ovL z3E|fi*Js-F*$?F|~WlN{A9Z`x*?!%HpZ!+k{z5s;x9AiUgMHrOw)YZA#_R`58X z62Ao>Q>uSL2&2|3URd`MgyFdW$vbk>;ACsB&5B)-jJ~bHg^#>A*G4Ko>WCJ`;Xcyy zi7pVE>}(vpWnNADOd-35D5CPTahzv=v!`9yi8wd=gS+~aYmuk=mwCcOAML6BJ||q~ zjK=u@a^7C{3i>NP3i_75{_ji=uDxO#W;G3(M>#K)W3)?&nI~ABSA0}#wepVdikdH1 zdwP%GyK!yGgDux~@xmrWpO>^Zjta$4Z`|IgLZQL9eMB&IJqpClF5J=)gF-xAO+TiL z6L6~ftPG-0Ekn$xQ`UihuG??Brd71!kY!k)z?YjXX)%UbHc7+SCigcxg+%w-EVtgcWKvJC$*sS0P4Knk(P;-? za_fexClZamX{UrRw%aGaVWv{A6uwX;;2B548vohv!NUY=eMN%i)=f_T4$TL~IF^eN zw!L9~52RG`FD)8o&plGX^dn=P@aG*g8h3oE*Z#uN@&M(!4a?Zav7snWxKnnerfEG{ z`;ch0CU2ZLH1g7G&lL_eMw;!F0J2Z)vNr~Xb*oAnNmpu2s)EC^r=5y+s7Q*)7llUyxT~FearFs8*ahp=t`z1oe5`v-`K24jgxxH+Qq+M6<)ejl!e=PPMvGx9S|Ko=*#7>zXIoyYw z9!i#X-lBsmn!gf<@&h~Ix@-^c5+k&ws)ljTQ^95PcGuyKFC%M=0$5QVd zZl7ekvVPpK9Hg5!wZWMWP{+Z~BZ#JYTeW+ou861VX8jz{n-SQWo)G0t%b|9^BUGYfr$-%EKljjm`LfpW#YRx?9KMGR*_~*)Yc=Fux8<>u3}*0iKsp{Z zo(bIT;VWYwQBl#;Bjmgv<3Z9IX)9Q8?D?86@g=`s7n@*w5bKlZ;rHH8i7b$5J3QEU2T>fVYME=p=8`?o8QwKCz~v|@S8&Q0F?Z~~cM_@W-tg%FdlciGbn z_>icUQC~7&N0Q-kAeo+$mu>AA$lWY(a)}*V549}XM1x}C{p1_E_og1#?o@7QE!(fq z%QCo;JCV#E&kmf5#O!!gz>}vCArNM$WCtP5L+YpU7x0;99Q?-Kx^)~Me_As;-7W0+ z@S(}+-H@8&3?M(5YPyK?3>d|9N4&37RvyqQ#KC$Ijk4_VCCZ7L1o2Xmc&4yMp zI!(!$7>+uZxKh*&cCJavWx&Dj#-59GsXGRb)u^&@Cg7hZn2S6DZsE161HADmQe<^wux!$`Ldy>>PRM+LWL}RO+ zi4WrYFvGH4EQ@Q8!3_1PxS}L<(T?|X+u2AlB=cP?uPB{JdfQ#6sMZQ0ct2x{fC6kv z&5dMbfg&^}5cPwl-Qd`UQRl!4C>F$pmo7|)0lrjTrWL$A7<)^ z%jy&`CWJA->NUf%s?Jylb>^j%tIZSpz ztajIWDqLU?Hcd0CvA}burkSL89Y6eK*ZKrXpx&NqJq|6v^%vltL60Ibs%g+CorJD8 zp=st`rVfQnTGQZ4b_QK!PSXrc9x$rHrWq4zj>PWCxh*C2A^h{t=e8U*INDfD4NSUlfIZ8_Avx4xcd^0HJtBi{8Qfra4UH4Gj(C>osZ>1vB-t#Jx%3eJKIJUe zjx^8?%`8`l3Xg$c30)$kxn8I!;qPnAayciFVo-9p5Pf3l!>r(RA(Lh$T}Zz?%S8gn zbJAp%t3;gxm&jt4i|-n#96Q{F$bxMJ^8v35Svec2X`SON+qfp^;o5A2;oXu9I_t#- z!x3f<+yk#!t}EKW_Umh+1d4+RWndwI0rLsfLou4Vt@TxdM81uyo0~59pO>V3zAd5_mp5 z%M~XDv_Cd0NE*>6KWOigHBJxEIl5#~FaUJZvs~>m0NotWjx0d8Fv~R$=yk*0B?|%Q zFV52?D+I{dyCLc)hw?w z%_{Cm)>&R(d7~I5dC}_n%4Z-C2XcJit=|N4`nqm5_PWu1O5p7`0dfY-#X`J=&~t$%>32q z2qlFL3i;>A%xf_Jt)EHd-!YR6Y?z~_K%3?Ej(gteyWQ)L2 zkL^vRPS$Xf=#SmFI*aI_b2hcrE5H7&SgGp&1BV5i7E+-Y%(ghA4xBM;{CKeDI7&J|2b)WQ}=U6Us27@a)E3 zo|k+;2WMov!$?=v)Ee8CMW~Wo@GdGkp@-Hp#D;*Mcj3|j%f~Wkh(fflhK)lq?*@~O zbQ>Ld@m=%t!a3KE&1f zeMnPNb*7QP3b5FCd;RhTC->2CmzTcz-r|dU4Y+yEozu08J56OU8LOi4 zqgL)fEyrCw@Bj=1;_$%^_e3Mbpm`7jrf$33^=gB!gQm`}tPAY)P@K;EZ71p0V%<%N zW9`Kx^5^d8SVhhv(9WuImXYua$6j+a1{Z}Q2%`eiP@Sz%P<#f?>!49zc>RE%tc)S` zt)zI>qDc9L{-A0JzfTs!OdmwZ#31_x7$p5K5h^wjlZZa_CG5MM(DOPahPl>vF%j3S zm_|kGn9_+Mf#d_KY4@<}-OjzR*9)h}3v@l8gft|EZH{urU>{zmXpqPgB+eYf!@KKl z!6L!OnfI>Ah~%oxGe>hYKQcm$2VNjpbhA^!ZrjxD!)R zDRvTVcZsFXcL|e;duZxAVhrSu>e%B_-t#zvj3lMn{5D;w)Dl(tH@Q=eY7}`jn{buLn}VJ5{X&)6jR%*Ft;1!!}*Be2HCXGnJvcA2peJ z#&<1u@%a_85a@dDvbjoRG4+Y&Mifx*Bu3imyQbx^FdiuSwkbD-@^$tq6gZbasH_?e zc$m~eMnZFNMh zNw8b(Ti29)^69#opzvzG%X+g>&&YxdJzV!Q0#x9|Ec#{PUt)syJ0SG|il`PedzT1(#S^h8hZgk^u1xgj*HrT$HYQSdzI-Y*JF$&3)pc zdSxQ;5hpiW89{rJhM)4~DJur!hy9##LY;WsJBE9-7OzR1m|ohRTc(^Dv~DKV=^6^t zeRV)4r*iKN@4<1FGYbYQx3TLrw638No3=@wvU6(>78t3HE+!kNVl|S!8bjzkGMj%a zUE4OGhLf!w>2lYY-zhJ;OSdbsH2pD_T{#*Z`5`YC;u>1q9m8?VEW&y*h$dML0D+ zmRV-%gVCp7mk;Xejae_i$Xn}}I~#?Gpwx$I-iz)yqJ(gne+e?;flD%sr@zz zC;m}qOOFtU(>DJ0cDnzw)>Q^9zQ<37*0A4o&g|;#3QRSnzj$Byd7kNEHeYl2Rezz4 z;~OeOaQDyrYTKWu1dB#)Uj=rT&8y+AuMY|X+DlKL*{)RFrn4FNKvf|o0?Dszv+tVw z%q5>tM3{Tw2E)U<#&j5t_~st*>=~Z9sVg;{>WVW7LGSiRp0%ZH5|Egp*Y`QSZd9sW zBY02Ou^HWh^Uh+Lf)dLQ%K}=kl1wY?AeJW*)gRjNrmsKSq*a*3T=1&dUh2JJ9JYh? z3p4(16I3@R)(j{s?k03YM+EjBKFYq`wm`QR;D2?6+G8+nv!c5C(;Ng?Qmnk8dw9~Z zM#Uv;FG{s*<-KNJ7mU?6={Kcqisvt`ggEYbNP0vk_5p+OAiMB_twBr>?=r{jr*JI= zW&KWqa*ERA6K(X#!)r!n{XSQMt{2U2%D5&a$j-feO#7rq$lr8Hw&iN)!XtLOn}pGa zlb)8Q>OnM^C(%1iZ_0`5FKsw%K7`&d^UF<<{*XDMEf*M+dk1nMu<&V{lhB0z*g#XZ z$6L}Y0;>)A8_{MtEC;F4^;3;OBi;hZ!!5>g!G?Xqf#50~P^-j(P222El{%DfS5p&TA1c8v;yB))3C5=2_%ou8|YcU{neb2&Tt z&9WhH|FJ?#3WUbDJ1RMDN!OHp@m{}jfBo8P3Yigm4g;d>g6i(f0_I`Caniw(XW0eY z&1>KXjH%>rrh%T zbiT@^+F-(7mo=3XAN!=4#i`Qs>NS;!?7LjTfPq|zg0}V&KX?#!8A)X)!-AI@#Bxv* z@;*1F&%1It7c_Kn+D$TDB+%ayWk0on{qd*H^QpeoL@yf}0?AhCeML~8i9G39Ud*B{ zesa}PFNH<)%6^8F(4;@*%@k>!cOKlU(MobNvRgR{*9&30%hp}+mof+j36WiXVgJ?7 z?^3bWnHV7(IHn8o9k=0E_D3);77XS*O$_oJ95yn~!soQVZ^0Tv8(5ZOOb;~}RG*ag z%DhD3J!p1tP8I&RnvZ?u!w6Q}fHcRNEqhQ=?ka9?VFD%ihSHMJIjt1@;F z<$JS^^O!n1m}T2!dHvoo;{!Zug7?}!!WMD0tbxJ{vIR7|e2@*F=*6ql`YY&0A!Sn) zf&+~o_oi{4EGey`Y6o7BNPGEYrNFadY5zc3#LRJjglr;3V5BoOs78A%rHNwN%ODFI zG9sIG6~yeMW(wiS*=7_UvefQlz#sBwA%4EukgiSpwAwiNZhZpL9h7VN{6e?0519L` zS`<9Zr{ztmuO;p*kqeEf`@1TNp)g{SiMg?y~c5a zx_PP4joW|qf!(L#D$vELt$1y&IZm1ls%hb+_fKuS6MLTOywV2!TTEw|qMniat0LAIdq5wt);c!=6y@4S+#wK(nbh-rYvFg- z72c<)a`|uT3)#WY0!#2s6+%CyUzt!AJ=yV^V)gHgJ14iljjQ_pG1o2af+w|`xhS0! zn)A}@Mrond`{1#e4?!zoGa`d@bUXD^_nyU_EZ=Bf-df%~ZrqCCEccgb&o3yiS9(RS zE;|K^-CWkqB==7gqI-f9dmg=F3`YW!^?~EIZgM8~#(-7o5vTKH{Q#TY_6kEbD&3aH zt8F(5L7lQU>L+rcA zPl`4JPAVRoF2_u6F5h+xAf)?nJG)S6RW_tN4R{Cv9+H?YW9}dA6EB!8V@7uj6e`Ro zsEdT8(%qaJR<{OPdwaqlE>V|VXmHFRj#8an*!-U1>MebU82zSf$gA;$Pbk+-Ll}k< zK7}T!g`A5~`;rpY)A#0G7IWiFibvMdbNFANtXg=qK@ic`7TeDvXpWWcrsw;zPBGu& z44uZYDYojB#8$Wa!ir<{W=UPRTTfgR6zzE2YiI2SI;d(06?^5ht<)1l8^24qkP`0Z zwr5jIvc26at>poyN^9{egH_K<7#PKG-#PkUdF6_$k{-Pp z6RhQC*F9YMXhg966!9?DA$KKb=`-$u$wdma1+ijOu?OCAmZ(FM@B?OtjKs$hbLyj) zIfB}-f`Use?xJd&KAFsFYV)XUO6ckr&fz#8f92AP$H(!gACAW6clj>ge15!i>w&1e zr+=nK!_$Uv;#~bl5*l8u-VKvEP7*G@9rG^SH*21AU+6wBMQQ7#2O{rjqRe5ag<2Y{ zb#?}s>7BoX8yoI%R48vXBveNAf=<(aAYyNj{@xbo%Xtjk^Be~cEu^EI)>n;>g$>NY z$ZMNyPB?TA)7G0pbrX4)i9h?Vtl!SL7rAVzKEL0%dHv1j(OMpn4zK_{luVb+F+6u3|D4X z9K0R8#8!+`RWb%*(9N9-NH)=96p9b9A?)Nxeb-StfxQhd$znw+#2gA8nQq! zyy8dYzD$%tREhnZ`DMyBqZ#}A3J0ePA>w&rD*Uo51fI3FN;|Z#0!}OG5($i0WLHY= zy4F7Fx-6e>E0-bP{Zav=?>fxPvhw5beWhje*AaoYEMnRXpXXY?y&Pz6vD~{TpWNrS zSEJyx*w@Uuej~y;=;g>A4;W2zYu>#vJk5szQWqSLozr6N9%Te%m!!J{qtuWNYVNxy zJ3H*=_q|TKmx7-F_f%TaQZ;(ueb*#aTdqGMnOvRs%C_H2M7szpYvSyDaMG=grrY>r zbAz03gX+nT?pZuKY_Elvj$YssqPoL2xBl7uzEsbwH2F%+Wsy)?rK3z$8*fX}1IP8z zwK*4@bEco|TasTYCtcFlybr3Lcyn4(y|zZTvJhSSyO!WTg3!{M+S5G8`FH0z7F=9XFH91M)+NgXO5@FTMZJ4-UKoVtt2JfX&%#IT3JYsQQY> z&a>&eR4bl)wlo&E8sc^_&aWOOJ@fUA0+6|-og_J7xV(d&ftC^h^#As zi$kITk@!6s2U}YkBvubV*%^TmLSTq2SV#a2u>kOLK<0x)<$Xmj{!@_ZEqVFJ!JV;`Trt9h`~P${xBZo z0FmPXR0)Ot-$G)1b};9IG7*kTmYH) z;b6EBKL|)*1U~}I&kr<$-~|it0knbu15n@^nIoU`0tuLA1c3U;HiZO*fc$G2GCvy( zuK);{Bbh)B3Yicv-m|*=z`(z@0nh^do?Y|u0fG=bOZ);tXSt97z%NjafWd(Qe(jeJ z=$9W32u285j}Ocb(2zNh&h&!p?rQ?^14BHkiInfZ7XX7_n}C1qogc{+!1;GG_^z(+ zIiNDocSg>*;uC-Y6~5{bX$L>CiX79KF#!gJto2VTNT#4r(Al$Z4F~~~*!P4qgs(P$ zw1c11nXCc3K-$C)>C7IGHt{{3*~&MYK-$BP3CY*lJH6z)B+WkwX%|SXB6E{}COhQ(|9!HP z1p}fx162bv{TZqoIoY52+%t^yw+Rx^;=gtN(+LBk|H%n!Uh+hTVZ8|knSgm$-3sk^ ztCe;?aii`q*;k~GrVC-8N!;-c5KL=-$t$(rXy=Wzx~Q^%rUyl zi&6WPCyUtd2*Qz%?@CP?AKGp7 zw~Vli&6!31?%v0VXzDXhHi9#vr^M`A9{NJo2Sy`RR>+1F`qDEW<-l}02uGRTeHI_D z7rUAV)sg?*GsA!M%>N}Y;En%LUaH907mSOHeMQc{Iu+BrQ*)Jyib2Xsd_zsR3y<{0 zQVoT@Z}a87Qq7#;&0L~14KGm{172p~$P%*;!_c0b>4E#ZO?o@z3slsUe_-oJ-Txn>f%UFKm$j06Q&vqr$ zkmV7EL{>!wI}r{8&J!Gqu4VIv+Bl)7XlOW0nD-HwwnljR7f}jcQRV%CslQzr{IW3M zt>gP9rs%9chzU28&%gmsMHbm{sSNJ&?;Qj4#NWMVzy$Q8*@z~$5&}ym>A*5`v;*wu zV9$#D&vT*T4&&0oYR%XP+{Y8!-D6<_k~^}H8XmV$5Pbp@<;nZ8zFT^^ebr^Mn~{8q z`2w2jY6(hfsIL$4{KQ&(XafGg-jBa*`A_qK{=^<{Jya7=7K$aW7caUOqjD69G!G*V z?@bAwY;*%~%<%X)!4Mn)(}MExq1UO5CKXvilOzn4uSp7Nd=@fxKvCBrg8YFS1>pN5D&9IF5>KcA$wAno77S#+;^u3dc)|jJU<~?H9c>QLz%J( zr=X`XZY?Xg7hiy{9MdOBx%pGiH$y3NRNmozk)o;aY zC@lqvu+bX_LgmPambE7&lIxA{?VC!y$7{M+*DdL&xK__9GVK$f&f`ixN`PXMMwL{z98YmdIM|zJY7P zvx1!>wlc41FG}#{OnkJ~s-=GARoqK%T#y;Afs7FxxZG{k0ztolkfgjr|I9VN&*z*E zye|Dg@W@{y{f*!OSNx-u^4{mu0L%cYJ2dEnZS&_`j+i@4V)h1{QbHcz6AMUX!E11o>+ zC;9k)5{;}zYmzh!_l+nzIhOfnuD}9ch0M7QQ6)L|DEiCdX_tv|Vmp;_Yct%IK-Reg_Sls*FQeADo3@%8gA5)i5W zon0Yg;2#Z3R7=?vh%)9Lvo}kWzE$-ss94yUziM=SMCT3&f_fj8k&RoyEWVi*%c~Od z&eLQ`bHCl$(QcJ~hX+x1Y~X3u5w&Mgd3#mzjyxS9$9)^TY;-VYH zn(#(VN_kSCx3OV`Oi2=Z z1R3-lSr^OrHk^CTNc_y6nE;oldu$uJ@At@mW`VR}^-#vXScdj#GNCA2UEfMRY0d3V zcSmcf3-DP2-5OFdMn=jbNuOQ`D8K*Misq?r!N(7u&helW{UO`{)?{_8*vP>_* zOZKA9s{IFTS-r*su5n6xMKt2sr2_ELs_{Eb^0yHQa{Qc(YhdZf0KN5*9nCrGo;O$M zH-xpxV%)h-rP1p$e1_OF&|D;6PBX5%k%;f4(8=Y=TykhsgePh;38u!QVc()0JQesjFnfIjK!<({qL+KI2 zKfEQ^_n!1-IT!fL*ojR_r$*XK@GN#Eb8X}DXu5;-?!!g~{K zr-FL>GkS%WaO59G`nMq-FR)1baisj3tsZ?)+&BE_E30_I6M(9GJ|@~k$V@}bp~{UE+HOZ zqI-JOkKd-wBWZdO{?;beTHdj?df0}2zx5q9MI`BJtU5_t)LIyxEyErws%7B&8B2YA z$Q?q;=ZZ0yT4hIEB;?ko|Cf(O>cpeI_5PdYxnczu9rWfc(m2SxmCbPETzGk;Wss75QHG_d3r zMaNkl3)&WVJX|{b#f2%qL<%8Al#!=0m=Pm9ly;Y0okmRAGNvtN#nl?Y%pnU&uVFqK z2kiyNt&$wgKg+6@=F7Gh3c=y!xI;zrL>ZCi&OzLA^F`OeiHBGQ6>zM zahG%4kN39ds~eCmd4|R>it5yeQne6pB497?O&PjH9QT^9lv?`qd2|08)R=g!!#{}U zZ(~qkpYccO$R?g0F3BKQN+B$^g>0GiSE;~CFWl8z0-J@2;Mxlo(J^QRxLncNdlY6ko}OF*rFUc{9D6m?5Ligx62z=xiQ&9}@@W0SA5OD?fb#ErEezP;{7E9$9w6;S z#!FWco2UkBv$lW2_q2liM5)0Ig*&F2S^EwhD?{0fDf>gHKVG3KvmVzp%l-6=o>XL# zO0XP1HCSbFQ&Dcr(>KlKcba`{GoLsN5A|I=u-E5J7wzWXWPPi;eq1QhP_b@wvl=HO zSfWwzhN1>6xY{zTIRDC3hJ1=wVwCc^?2I0kuAr?WN)+3N`pu7CqtBlc@c)C_|2jkj z_7Hy@B|qS=fl4J1y0D)`wis$Pff_rRA7(lVM)jhZ zUxr#l=Orrm0%3aUMFOU-Jmmycg$z-}Pn;|7N;3JpLs&SyP69nENF0j?u}HxN52i6o zIN52Lh(+TR>O9*8^kNsT>8JJEF|$w-%(K+0Hgvvzk=YQLn1Qmi88S;6#|>Nn6lNvUH=hD&=;th)#Mi1EI+6SsN)&eXXVaYY0J|Nd>8O019y7mH1_X*wSP1SiX%?bXPzy^P1QINiPGt?8*Y2}Jf9G(oC!)# zPwQvK`GT9hz(c)e61v=Iq@Yz&nkl^YAaHO z3oRbAq`T4w{sR}mwmDhxAQQbfvJlC=oAUR9j|_ASz!TflitdR?8mXaI!e$k4h3d&# z%OM{OGjg!Kw-Xq4ZgmnVj^taK2!FsA^X4k=K)fEVPFDJ$!0Dti>szeoFv5^tl`ZQg z;1(>}EOEdybk%eeEUEpR%^_&~&J8QJJW87=U+O$uzJBVM5n_=TUZRY{V73r5WtJAU zRB^luMOV#eixcxP+|GGT6yA?VCEuH|sHQfLn(WJ44wm2`4lw_4I(6*-4%04c90yS2lqN2eR0p>q+@(qLh34~<9a1X-L5 z`vbZSi@UiLUA|pKFNb`n$b<+;6(%WkRLf1;XxlNbP}`nADB9rJ! zHd!ioC9k0sS85g~QTK&x3^U%a&}DSQ{b;N_H|0eWH(B$^_LhYAB$cK%V~1ngI&3>NBj)#pyll=J3I$FUM&WKz;@<&JpQ z!!PtFlq#1$f6E>pr-V0@Hy%=7`lPML}$NV-o$R}|9p*p3X zBcJGF5{Hk-}?g=ax=GdUm*%i_#8;3aCz~^0(h!s&`#vb-cnb;YkXA zTF=){fkiA!E8;I(=Dt+_-1iwD(NkL-wXAbhYT}7_DCP{NhM`+kGD6Ce$)$yEZT(01 z5-8dYzl#R`W32Tjn8A{7=XG zD;@$++K)I0;2Tl@;vo0{t^RM|AkGH%|8Nlh4+rrbEASP_@LxCxz|#Kj;vj%ie2650r;@-~A?tv>gpzpJZ5pXB~oaGE9p$OnXzFw(<&kkz=Iz0g0f-D08XgR>O zozea1ZGb8M$Fehz1AOc5YoQGQ1p0P$hBG@m0|A;k^Lc0c=Rgm3V89B_+He5yKVPq0 z0VK?~E2JCb`#OgMZVIsAGiH!(l#6tRNyb9 zKW>{y!oGe}P!gc|?p^=+Dk8)RK&l}7fJnl>lL4Oi8#wV#WW3+Wklz^mBO&sEnAi`D z|FDiY0rLe~bM10yjL#DBH9r36}fOP5x;b)JJ+K z<`k6SO>jGN3}r0uVebQl&+hGBW4dbyBe<%n(j`(&P15v&DLGSFfgH-Q77hJEdJ}K! zBau#t9ey`Fl?1B3LTJw0=toeL?x1yTMxF={qc&6Y*YDn{_X9^X4deLvZh3a41X=ihv2i1OVVIoD&LXg+f``L69>#0BFnRV&U!X?B?z4Ue2jV%apaeN>E3;691 zfKu`U6PqpgpEBedkNg9>ek+5+_>r^F&t*^n-~i>>#xo55TfZ==GwdvIRpZ~emg_(Bl2!T?zB zuVws5=kyC564Lu?nIO_J|4hdR`@Jq75~KPnod5vl`=yK@1{`hwTE>fv9FTwi%ATJe z@ms(ALcmV_FLVev-*0pXUZLOgf#Cn0KZF3^Z|w=dkTJ&3ya>RN5zsGXe82N7Aczb) zex?(IA%2syAg{o$I)cLx!0P&+y561^4$d~7XCF*fbhQNn44?uKNkFvR+>mqR*|oMq dfDLjN;LOiiczT~rJ8(ePLLg>lS#>$k{{X3SmXrVh literal 0 HcmV?d00001 diff --git a/doc/pub/BM/html/figures/figN2.png b/doc/pub/BM/html/figures/figN2.png new file mode 100644 index 0000000000000000000000000000000000000000..7fd4704a1343bc6b4cffd64e3ce7a31d75d7419a GIT binary patch literal 20041 zcmdS9RaBh8);8Eka18_rPJ#q?_dp09+}+*X6Es+GXe_t|2<|S81PCsTySwYWC3U2^f5xL`rK3T-1oKuy`)t7+?pR9N#F4YC0?vW?Yj9w>+nMkJ>*uZ!I~G-a5dL zhkY~c9P&0&YU+RvL=zI={()bnjm%2wI=yg#Tb^ph29@=AH!Y9emLp-uT0U$xG_*L~ ze)#GF`gveF{YP3sb2WEJ7Bl9#%UoUKx6TUNvc~F;zvqF>NISz_-1`2ouHPFJ5b=jQ zB8=ydXXoSZDm-5=W{di53KX1Fywp<2<5KI!kx7$>E^yyV_EL-cDW5ziK9J-~niW=v z+c^zVi`uxz@}2u*aI&#*!0Vfz_U+5WP<(p!A2pZIZ{6?$%{qg~@7ugq6G1LVup)P3 z>I$}v6$w5dzM~U%_|#{wqC|9bqyX3*5d>OT)t|^Nf!Hi?Nga4RAp8{=A5pAX&<#1l zdsvu$*hT|HT9AYR84ps42%`l^?&q6ygp3t|2H5g%&=n>EgqT1q*{{hW@JoTOkiP!L zBFqtG`|&ytV^`!87Ex6QF*$cX5z1GwYztcC!2qZ1j4#+KD5|2aUvMr^T`-zp7lRP8 zL;vz{z&)b-24SbbPZ++^fSV4TV?lHa_FcVn;7CW=?zlbC1AZ=4p^N0?%?+0T8tQ8D z2{9}dWtT_-wr)_A12*P+f^aeTeziQT50ZPaYI#JvACD-+Oz>_4x_(;FQcL5ThC6+_ z&Aus8nN(9@EkSXGoeJEOSo|23D_%5e!KOu8_N`9j{BPing$rI2NH^L}ygs)gJ9I35 z$DIHzB+$_C56egjxh%06djWwn!9+-7Bw>%N!Pak$UubHG*1?w@A%+u`uV$qWqzAc!nXu_92nDK-<7;Vbh3EEeF*gj zNZw8Gq>PB0O1g|2juWKxmUB~!W-|W(Nv2QWXT1JQnlzJiowUs0FWaW?CoihlDMcZX zt+ZOw^^5Z3m>R2mrR-_WTI6fF0j&>e#j?9`Q0`jt+bQ@U2RR#G*uAaAQYsdPH$>N82jh+2%IaNbg(P(|lB(%;## zv0~QYII~j?-d}jrDb$)1pV(Ea_0HSyK1Gzhsc_ZUFX#GYn?sVTTjnc?IL$g0KE?Zs zy#b|l&_{K=*O7SGei4+Pco{&Ghm0#r8=?lBN8WU66B2yGZe2<79u6QU@MM zDNYRrPo+dv?b;}~vEGO9&M%Dkq}_U+W8xLy?cx|KmVgP#nn|e=Jq|eH+(&Pqos=dH_mJ=bl91(LAxRO z5ZjyPUhEzQvE*CU$HZ=B$u|%S2%cnYeE5V};n0YMx%&3wNUp#u(yW;toW_(!8^>K& zs>ZtpwMHT52{)Vu*XHgcrX$1!$aG1CoECH!M!gcD2qh~F*Hwb#AX z<+;kaN-31=ANI`F#g)TgL_JPd#oX2QJqM}PNz<#9#=n9kC@1q(3oD^k1y4jPpCv2k0_)mrroA_J0&{1Z$#U# z+7v@kY|)qJtL0p^pXvq<<)I1(g0A^EYSp+X@Fx^HWOXYUH#8 zi?)r;SBR6CmBMHTnyrOSquX6!UD^@|5}Ohcy$@^J!f&56PS?kmidx~Q3aBKn46mpT ziRSF9pT&#CE#-H6V6NdvO#f7MRn;SMleTEu>F9ZfInb{E_}x{OQ>s+vq`zsTDm&#+ ztLv&AV1fiNp@ly)|GD*@TR&O6d&!&gs8(+EFE9&ku+?19rn3G_SNrse3R`o7J2+9 z*0vh8541Dt~X*-%#qNTYvFUx zE03?jr)M!W!gw_2S@BuP3LoS+<%s06vK-SJ?~N8C29iH8 z@^S9+d{${~d|X(J;YVA%vz!V5oXCwoE6IUCUeq8^KoAIY2RsVc0fAguK%hMX5Qr}Y z1j2VnZ&DNhULY9BNQ#4=U;ci!7RCclkR3j2Is^Yd|NDL#uoeP=P&~X;G+dMn-N_xC z?9D7~Ovzn59Zbppp)!r&jAYw5lVWs<=V}z@Km}w*MELI~PuwdX)!Y^87uk*Z>BIhb zVFvvuKYFv9b+k+7v-d^n)>mynSm<)Up15(oHhR8x1`HwQQXBTal2XdEoLCs@f9L)@ zk?Mbn?B+}Q=&)Vm^+fiOEanMvE7OE#JKlDCb_l;0{>`*(UicY5L!Wt1g-Y!9{`_)S zepku#`PAq7cob*#A?S49KT8;IXLK~?*>GojJJz3Ta^O#LDX0G-OSSRN_2{#S-`(Y( zW#QVZGYkgmu6wDwv>+si>S|OZxYjrHQ~i>Xy)31){o5g(zT?~n_ks!k`6qk1=et|z z#XkStLlTt`iEux-GBzB=e)h@pV#d*;c2n)`k4N2baU#q=o6VB>VfV{Zsbh+6{7%B7)rq`CFK9-g)Rm@_K0hoc>y9ijQXNw_@bjhPfPi zJaoOuuYnb@W`hnuZNqN%V zb8@0w#-DHtkU!sYF@LfJ^mmk25)kb=_7Ni6oPOi*f8xne8II|Hr4V!w{)?fhGt;~> z-c$n5pNB!yF8A+B`mYc4DPP=X+*Ts{RfGr*HRm*oS*iXudHRtvaZROs*+*!PrBP*R z-r!Y({ll%mW3qOngKBUt9I-Ve0Fw)&2j|hp8Wkn$#Z_w)zf8V_)Zm$Cb6~`vk-yPf ziI>SpUdc0_V}pmKVPb(D5e)w2;kkt=w)nRCwyla*Fw4r15Yn#j!6Qt?w?R)_- zmA;t6aX!N-EssCwJm4*?3S#B=#V_mqj{2TB`O*k{k@V^Mj&odg=MOvD8nhcdE6s>L z-qNXhgOpl!`3nQ9HDA>8uM|rz#C8*!w2DU>e%!vh7!;X7Swn(h!ArSRM$!|nly>rG zD{i7_g`va@qd}>KMe3`MxL&C(>CTCbutwY5sKN=|v9fNg@>WDUnIDbEhe}Er#Hv-Y zbF~9L(M1HTdCb`+53_P3aAzJ=4;qt`fAKjfYvwt1yk3L95zz{z6S|}Q^O{749?`h| zbEAiQ_&}{5Q$W*<^p7Zu`k<4v?aJI9-OG>pJ(*pl9Rh)TpW&+BsbZ~c{IzKhD_9;Z z(JGzh`&1U6MQ#H=T{x=?b~+AER7} zyX^GSu$7_Tc*@?fSJ?Js^Kft~er>mlw+@rAzZ36n*&XAOvF9MSmA9YRwpJ68 zV&keB?u2XV{xhEWR#HMU$8oLhl}7p0RfdTQk*&US^hSNHvURT3TOL~$*=E{t(}@Vm z&iCOD33#$UHLV=!2WN}so@mcvF|%IfxN`EE0!dzn8pyqGR_^x|i}|#ff;C~9As^CW zq3^{PeS!YKeY!-jt(Ky59n#mMW!<=SB+#w+IGBGpqvDUs)!aeq-{) zip*AdO_$jR|5bT3r7g6RjT&K4%U-R2x z9pTDW+Z;^V{}~vDC>b9xVF|(Vx=WI=1Nl^DUm(3tvC`uaL!1kPzeE(fpIICqcy01= zyS#_G)w#Ii;I)phS4?*Bo*sxQ_xjzefE&xZy?B z?ndjm&&xCsboh1=VO;E7b6PgUM`V8J@0mISLb3?&g-U!*u6!R7kzE$s`_@m0rtlMC z--eI4xTB{sPngnz;l5h#gRC0-Dp*YZrt^|oAqHiOg&fW4;@1#kG%F8cB`9(=7xT7F~E5eLE#Yh!-y{L*pd*7qBa%)#6Iw7^$+cw&$-kb5lSwASG&gNNoH;r8Xg!4d(~*bqi(dlA z*Mc}THhn%6#kZkJ2B{*KKj6diNSNSi9{-lbwDA&J?C3mZ5J~!@Bi~p+w%sI!XNN4l z6yQW`{0_FatJHA9SffqqHb)A*@vVmzkP-(#{|uznmGOpVquV+*Fp|tZk#mRhy!8_i zf6R0Ynz_&p`XZ57fHSeXq;%HwG^I@8=NXKCR;UHWiKx2VvJ4`bo`Wy-$Sfu0@rV#{btq5bwrD#jKsi( z=Xh@13!U9QB;@USlG;E&RcZts-~V-OJ>ojM6+r6_D6n)qmWLPv3h+JER8vkSx>3h& zkere37C{s=dTyYprXlUB%PreK2C1H7;E?;a?a3mC6Jki7H5%#&FAAFApqGY-Vga|^ zb%!AXtvC#ADBFCLwPfPrs7z_YA>5_)N~a^;_o=y+#eu>pUP6BNf&90!t$c z3#0&s#exY-3ort-jy^MfbX$=D&k#&KU7=4_Zmx+J5)$%-0WaI~Eipv2QUguDS%bmN zrwg8s*0Sxo)lpJJywX{hJQ3`rY#YL^xkNd7-j$|*NxyXc*fazdTxLPraN;##BP0lc z0RbWkG3?+NW}zs7N9wQCw}tlYO_gH9!=MDiG7O5h=XXx+dFlW#+{5`UUCwYx^J(5flRA7np70@?o*`QkK-%aJ+ z*S-zqsDLC(LR@t*OMMuJjHV7eNt6YL&gH}p88YPGB}xa^e^ODk_4gMZ-{rWwyX)c_ zrq!K^9Wd##Rj;oti5C3@&X{DS0=AonAVg7-6Sco&lpG1vZ#83<9G9?{V3!M3gnk#2 zy!Xb1Gb{81PUw|K4aJ{c656#Ni!JmAw}OLWb(#>tTgc$hO9mmyT5)J21A{ml}&Sc znw2ClEG*2d6cEDGS$M-oa>^Vo2;f*$sM20p>12!c0A`hAb=yLGQ)m@ZPB;K4E&`uA zTp3Xa%#15N-@+tV`p>m5Zba|N88)=8P^Umiu$Q1?w|na&BmwYJYb#B5JP~1WqubrF z9%r{cy&Z7T>ngvFc`)wo@4LsmzV}ihAR-pFM88!LWHg;@0_XQ*{jk-iu4+d^k&Xt|sBc&5%}$UJqJtf{izB zDT$+G;ROZmH?0y>DPAJ*hsm=Ryw+~r?FR93WF6o4o3Opvmng)Xa3%Tn?c0d}Ca^uX zK0*x0-kJDE$ELx%w?GtuBANB$;+Prhk8Jaml5H zongeFC8%wE$1>x(dhf@>Tl-y9aU!%haAAspK3I%G2U2(VoeYj{fXR#`o+HcuWf#H9 zGZ*$yE_bNPxn{`-pPMS?yP>Ks43RO@C^OpPYrd(e>ajzWXUzT$%~Emg)=cS;F$m)j z3eB&h1YT~ODQ~X%k7ykUq~_mC_K?1k6U#QtF;{ZZH?J=VQd+_7m73gNH9`yM0!^#u zz!Nsb<3)G!%4pq4S?aX7T_4(sA;r-0xjR2ALz|MOdOat#!*ak0NvD;XY8uv-3*s+E z;V>yu$w3^u=w6MUv#_vyaTc5+K>wL@)E29o=JoO;dj=)8)q370Ov$B1cPb&Fp{w^o zFJ@xdwntU(BeZ;WVHjcNqF}8T5f7D2zX> zsWV@x=#3s-qFAxYS4*&At_v`gh72-+N4E0CHl`z8%A8j&JAXwPwCjj{eSIgq^ysK6 zCenv)3%-#DYzcem+uT_d5<ld=R9bwVh_E3WI>kXCnp4K zRPfh6pTH=p_j<{(&hqux@^Z}B*cjMR@H{dGoro)^u#m>*#q{F`CLdv9O~OkuUr{q3 zsL~YUB{B$(eyr5^b&E=qArEP#7oRMUb*{2?F6T(2WU!)!E1J<>vbp=MCDp(uKV??@ zx{wZ=KN?5xb@ePi=tH;+cyA`<_}&%z0lm9uXG_Pr)#9B8U1Hblz-fbfRM>ovP7(_~ z5L^(Ub%63nAWw40pehRwu<04AWCzfiv98gp3VE_)lGaUaeT zTpt0zB43MY+)K4LLNkXCR+ZaMa(`Pjd1sGlcFOhU^7l1^qB%GvC5%5Px|SfuH%A(5 z^+&>$)wZTTOTMssL^#QmB11lOzjpi*+-e@qr=XyK8BI*s9Y+;TN9GdAZY~c8N@CV0 zb*FWvc5F3Q#blA2pnLnP)Z3YJUrMbq{y

    GM1W}+F_$V_GBmxZ#s@j#@8YM8kxtn!Ta7-qeb{>M!%!5KS-iinHbmK(=Wdyl@x z9rv8!js28#mSB;v7UIi5zqPM#{~N1K`-04a@(AYrtDF&pU?&3M_0zqWz+m zi2z*yEA~Y2xM{QX*vW2qM*#r7sWkIk+HJ1aLFL)a;{J%FQmkdC26$q7_+KW(RUxxedqhU^ZO#q|hr0zE4 zHCp2!+S2^z8X1In%unnN_{7N&J|@emcQ5n({|x~jAem|QGK`$QB;fynZae*Xw_7vi z5&ZDLT0ZJJB4Tac3jwj>D>(=>TZgVYvtPc?g8}N`kV`3;JKGmH?II{&9O(gMJrQ3N zSFp|l_-4?GCn*8~f>pUX10G|(FRmzkm_(^) zZzls(J9$XXo8Nr_=3z=7hTk1lL5T30^fk6?rVC$`EH!)C+pfINDpEK%u>!0u2F^LziaM`#IBE8ka!-|IA*LiUv!J?s| z$x4NYV*RYBz|2=dF0kx!)yG+@WMz<7Yy9U9+tnRQDY+Faqm0MHA7 zBeagNaQ_+6L}J69CIH%An{rq-m7{jEbOAPcRlM)DXwBQ5@v9`0gr9fBnln7iXnbU< zI(#1qS1v(u3Q>KchLG$8(8weJ1X#?!Ruydik@+u_XDq7&G;SCj$jAd4$>2 zZ+YKRCN*@T zZ<*3q%d%ee`Y*+3m(Ra-w-D!SERXOJiix{3RhTTWLO1z4@{P#FM)GHwugwHtY`eR& zdC2{9@&Ja@nyQqFe?fB=O)vAA@4^E^$AA15Fly>3+#HenTGReobWm8>yaa@L_+#kCacEt6C60Kn~L@d+#K#%^pYR|T>^O(z)j_FKiEexF)+GVBEP4V?}uBC z!Ur%~^Ad8tjJf1B`Zgp`EF09#vT`p3piaOkV9vO%FKwfws{<}ki_?C_(h7_)RAr!v z2bwG8%8>s!QPXhU*abwK2FsbghVGwSY7BU*o}?hn38ji~zy>`s5#=VpT*g?6Y{S}8 z>_}g~E9IB(|7Jk#HDW#TE$`UV1ee)xX`lLEw5>HU0j5O^kye zZ7PqDrN+4y!Wk|G(8c%rl1V~SFm(_XkRoy><0yKk)aO|eAOv9nE+%uk^V%24hI>?H z1fK^1u>Mn`!;$!=4HM#c)BVE&JMFsCT(x2@ugQqLVk&R$+?OnKn*!-8VLeo`)NW{# zt*MF6&B7w5XK$y{9A?hj%|A6?4xFLG6slAlq{F0MEm!@f&Nf$tE?b3;B5Txm7|iK- zjEAkN`8RL3iRmkOn0Hp8!Fb+WSadBhqh-lB4K^bws@8M7e4!#;q5;@!;hv<)?VeAd z<=A1t@C(uY@i8XiKH+1Oy^ zCSGff`)pX&0nt&>`8_nM(S<&FSySp8at>uor$$q;)&KovM${a(*#Kn=hx2cA`l;3@r!uO=N@0l z!>+mw;*e zA_wmi5P55jmiS>kGebK&ML|_TyZ1s)jXJ}6HC8{ZsS>597VU(x0P(zGmwk!dg`Qb& ze`zpu)!9ZyM;n=#z~|=X_HUoN&RqMF{Uo?@N;*shHf=VYMvn8e+*A`30N4h# z3}?Dtt~J*buCyqY0q_}gnmXF$!61gfkTW}fz(skZ%qUT4Xr(e-`WpwBASXMs6p;L3 zi6MDpsw+nAdFpAb{kmSTNc^=Yf!a@}$=Rww9iMu3V-vc)PW7TAe_HA+ODsDRY`BQz zr?n&2c~)!F0VR4v1##g+2N)JGH6RTSIM7g)neV{BdBqE^QZz)d*#xAk6JJi!4_FCz5(na{288ulGQW^f-~1eoYaxIi91&~YvR zkP09}RqNc6xduTX0n%yC$Ors$iM?K&nadB@sWdgJKY7E2#V2t%fOj=pf^tm#_S$=$ zkhk_f^NvB9l=qke)R40)kMJLb7{y|l36Tfj@g&-Rxx#lm6aQ^71O$pAY=BA7a|09) zX!{Ak`?V1^c%KVlmXoi)f%>;Mm)(*^;e35XDhUAaS<7nq$)Nzp3Sa-3Z>6c@`627f zkB<>&H&mG6c-fuIfS>~+%ye9u=jCTI1pY43IHGgbGSGw3KVzSO)qcrHxZmF|My8lE z{oX1mzJbx-^PLY_$|i&Hq~Z5oBOcj2z*pKfZ+TgrxfnP(U8UZ2;hVrJF<&0J!-;|H zJtJD4TaRI4gi!L7I>;VlW%cG5ejB}VILnhYsqh#4ZHjtwo;9?MhB4S40+T}RFd@el zE_ebuDR-BP>O41ztF>+PE8B#F>ljSzI9)%hbkTYuA&jk1J}lKw994!|U%eGajiMSp zKWc!xIk9bRc)DwNm+gj=F-DU%Je|G@x%LHMSCBF5HIRT&TFd?F04&66lXy7?qFUCW8+|WFZ8VBxEWlf#<++?2w9e%o@W!v-*MI5WsB?LD$g)eLSF>8%@+rTP4Er7z6}nkbLPJc$0?TfOpa^fRGb`5>UM zlt5n8vK+ZTXk~G!cKSFl|Mioj+%~6xf?f5PUyYYQ&V3^R7{*M(B|y#X{5%Uk$9gWP z;dZMbH{UI3z$C~L%t^+L_O7ME5=@JW;##jvmuO{Y&J-0rsQvEUyQpMSw)p)8&qAyK ztYqD3FEn*ssx0MW*Ao5p_GkCTC)XJ{Cd43(MyE4g>uID)uUf3ls+M-6V4Xshxa-i6 zjYsd}#Ou(|U>3PsbtH-P1I;l1OuS~&X-(xtv^8)pe>z}5~T3}Luf!Ldgr1O%v zY13!DE;S@LA$k*oCqh4rHAbg0lbD|*`$?$WZZK2`o|sRidA>GNUElWUZtPIkukpcq z&h)H;ZrE;)?XppnJlTJpC{eXdK}hrVntwU=-fb(A;gxxHm6rA}>#Aq`;cqKgnm6%$ zA=c9%W8-)#e`wOw`eqeFDnMU%T(@IhTo3V_d1|7lKw0H}eo0x-!|I0)_N$L^&h#r~ zc}0ndwtdPbufIO1K09^Ne0uDm2|#83wGiW@XE#x`AygVy<^?1KXRfHhgT|h#!Dq+b z%cu<6A-_KzIcLQ~KP1znq2RSPR9}f%xN`e@t%TilPB24Am})6hyWqc6!&`z2`o+J~ zOr=D1pI&($&7KrxPIq```)ih}&QB_%bAPT5y#v>fH-8R5N29GZeM95QXkABFWR z@Zd#f>8D?)7yytN5Fh~L;Os5Lb+M_{CcdPpBmpmr%Uh=Jt5o&i7|uG7@Y@!&x3^PT zZjEsJ@J7Rpo__%_hbA5|q_d*~^p!lo7K%<|ChT!S4%86Tt7*!ecK%Y>%>Jf0TD0Hk zkA2thg)@w5wt7@3!WLTBOY%wi&|#$1c%j~&41-jFz+J;RM5-DEl9j(pmpruU3I=i_ zDmiFfcU(S@rx6kYgy3He^iT1W(CISmN@GZL$iBeE-;aQurL8YLhLHL{I3CPZ(;TsY z@5YncRI1_B$7{NIE214C(Im4?uGR~N!>Rw0&7PKH6S`nEclPHM<{yoQ<_i=!oQePN z(;7x@HV}W3THDj!pf}mPHyAV+s|0@ST>%}P?r59H& z>*4r)YZ%{(ia*)rNR2U3g}W7jU(K(l*ME$x?%bQBGp2!@4%1+)C6(eSkO*Ka#sL{3 z_U_^C)(bRJ!HG9R|3{D1p2Wq|iGpySz23MnYz8kR5RX~*qIDNpa4W4ZsR@%JzLB}a zwG_nyz$UO;LseisC-xa2o!$04@%5LsU-E+3GZi{nOJKnRAQcGw9K+vBhEjWk^(SQo z0nA!=c=Fa1*f0U%&-T^l%hPM5aTXM^=G@EAe!N#}b+(R@a_^&}qfz80f=o46Z)&W3 zpt98O0EBl#lx%_4RIc7*L=eJ)_Hnk35yWtVWd#Z<-&bx!-6Lt=Vda)nwg6KPo(uT> z{GL7SnQr%RJ^)k@l6E{f;tiWolHYS(1%dOt-P>fs%ptEC=!LyV8N(-*)3yL^W z`AWrWf;O5-zZqnZD3+n!+mf$GH=O{CGP@Qj@6IJ4zyJyZVtijkT(qg?8~q|j638cr zeUozIaK6|2YXzF8;4c{*xx#cP!D0l~W#{dd69sj0Kt@$|7QzjB-Y1J$oj;^s!w5#| z*r5FgeEzzUl<((!p$*E#Dz<8AC?txAZZ3T)v1-}*EY(yVS_vs0JC#{ev6~1N#u(eP zb{)xmmK=R{Z#&PJT}x-AmdoJ+Ohsn~9U243haiU;8K@TtWnn@$6D-)|;yytX7-g#R zGW>fTiFu-NqEQki&xMr%$}Pd6Bp0%?Lo1!Hl#;FjD<_Rin>^H6#o&>NpX3LWjDTDz zoV8@^bi!5eG@RXXo9xH&6aWzc{B|Pv<7Et4u)z>|uQp`DW{CW8#uX^!OmK9QCQF4Y_o^S?6vwsl9Z7{ZPTj0y$)${CR@aU^gE_6& z1U;&;-E;TEQjY9|t_K}YzZC*`r41!T(1Ahri?PWOgivtEq4N6|Vc_qb#>U3ZuBo(A ztNt3(031v*VF>H7pkgUZQA2Sw*JZSRqYfTmvG-M2B*`kHJ3F=or=v7LvBpR-K3^R| z5Y>M7a6DICXv+`lNYKhhy5vqUW2cYrz9UXCB~Ngu0@J_8>zbjrU@J-2r`4ScpkQ(& zBSJpL`92icr15}0NX1m~!q(UC@1#jeQ~QL;COuUqPm7~j-}+y87FgC@vUmZ~dBKFk zF>dl;n~7h>{;4mVp&}LaFdt7LK(MJAkN+~BDOmbfUI{Y%pHi+F$-m6o3g~_CA8tcD zfd#nHsZ6y6SWK00sp{yZ%%lHwlA2_oX*T6E}YqLx&gfW@4>o{Z$nJ_55C| zj=K=k$cAKY8yXn=-0}i6ZB=9ta39g@e>@o!S~^~eqa_JNab^Jh-jF4oZ&l9C$EU^G zX9Td+1gic_zr6rzw1kSz61M8Q+Ax_$sTE#Py33TSu0|5)wxi9VHsN(PKrsbL>5QX& z1-&Gqj|K0pO0DJrR?JkBFC*Gp9UvQGf!7>pZ~#LEP|&s$@FrahcoDRT(wy<7U?lt^ z{bA^zr!(Xvp71c*graDXDdz!+Z5!q4BDh3k=eDiJeZVuxzIY}IMX3o#;yk4f0LOK4 zd8zuo=N}MXX<{OVAj{-fJvR2iKmm44E2qTV`Ts-UZ=~3@N+Q)-VZ(guyeKL6&;X_j zw5L}836Vvks1uFHf*%`Of~v%e%kn>$td?9IqQE$`-=(TyqEP_iSn)r__PzAHRB5@^ z#&Vsc;b|RGpGR%<Zzpm8)Z@f_P04ASi$ss7ps}bS^PBU-2tMRDE|3q+N)B=B-Eb2aNhTy(w=OI{F+I zM*AU!W53~2Z6znlyz*R~mkvltq%tP3Ex=a?FrTh6JGrxXkbVjuIRH&zG{$WVte2&U z1v_7;ZmIvCMX)g%mX`k{CocXv-V6sYrci)`jD8`Wr9srg)NZzWz2_CbU#M?cu-xMY zO-JGJLdf!c>p$kZ0TBCX>L`dgvbh=HkyO$5X|CFl1)9&v#>2I3thfh}`1l&Y2p0SH?-Nr!PIS?Yl!i>7|c+p!07iVAiY!+Z`)Ux1-8q>qUfA*EaVkwWq z%uhkNr}N)6rWgFz_+DO4Ot&QxcVvVWePji0%H>*T^@U)T5tnYZkTvSlgV#Zs`0geE zO|{I^(8+Om{SW{C2a2)A~Ya5w_b^(#AUh>=QOQ<(@1ckBz2|tVf8<-)b*7uj5RHBBDk9 z*jD_{jBjp2k2AZn@s&NE%ZA2$y?BKlm(^BQC6f(Dk2V`_){t8~emcS|J8tO9#rreL zJdqG5_{%9r2LlBK)CehOu#0Y;zFu1{+bJQ!$(Hr2y6ZxR32k){Xz7kxWWuD@CT6Fz zM|0&bGmXP;1#xJbA=T3^zmukraAStKK5yWtUzO8jHm2_eu3j&)P1_wqQcDxcH-8UU z=*+L)=Vx;3needxZ5?8--6qZhAbyT-kK0{&K`Z^?jsjQlxJzeSgUd|({)Yt9)@vu8a3y>II6n|Iir1 z%c+yWn##P!05mw3$jyceqm^^a)_@3ZogmUf+;Z8Nb-85y00=3*$M6=@!gKIcthwuY z*$irA7R;5-JB67OmXk8Fw(r7s-==Q*H4RL}?d~sbc+-ViIgS?JQ)<%HN6I~w@ccCWD+ z`DDSYtZ15I=~1!%@~g~bUZj{pjG z6I7;f^APiyrOjAMyap?U13!=qS1;oEF<;I>*Yi2?k^b~d=7Yd(I&qH03SHn7zASYs z)E?2T)sM2}`PdG-e>?E_`NXEzEka_2pm>^k`M9omLQID+K&5oK>-O}H$qU7FaGcx1i;!?ePa}q0c&;Z|D*boY+kpGkO3PgdVo9rn${^rCn z&{tR+!;vePOxP98nwtsosDIEkcS?0rh8elKu$w=7o5n0kq#v5ytgs^dv|XEm1v~m+ zM3_|e7LvNYIO}@JtK+{9e{}!ACcs$*FYI~te#v3Z>7hMK2UZu6YvG2SxXqvYX%UhI zIrejS)ppAms4vtEYu~(!8~x<;>Znzn+G^8w2K7bd{F>Zhul)ntp5o|lE0TzH)TYo%nRO6XJ}k{UI_Jz_oQF#WZ1~|7qF5sm|OfEXWppe|+4fo~83Oxli7Bw7Jh= zU6p7IIuQjqZYIVg)%Z%kW>f?C0xQMu!)VJxj zUuwY3&8;fYn%H=HxUz?HU!;=B1LpkDnbCnL{KZp_9(zp8cwDC^B-AwVAk@#EFu?St zLn-jg1LfEsfpruGOvibebFP`KEQ^=4@tU zEfio-Gu&b8bzVITx5}*Wh)`F*4uOI1WP2Fq?sK@y`37=Zd#R1yC{UlIwI_FVj*%YR zMxo#QC1a2&)nH*41u{=8(J0$Vc%9;yN_x8z7Fv^Jv@$j7c+P( ztLx4-Cod^98k-nqfgp~!^JCGbL@?9jlyBBlWO%yQj%0_cTLIf|CZCKoT} zaEHFr`U4Oe`CN}3zw_oqIc)7@)UqCbwQ!zyaWg}L>)qB5kNvSn3Yn3O8y?+OuAs@V zKj7Fr^f13mXo3^-A*3b{7^oh`1bo}srG9Vo(?@)_LsH0dzxz4_5ff%Mz+8c}kN@d1 z*kh{^S>Un9$>%c7oYgM}bwppm;6mS|)Nwbe%F)6Pb>D4x$}Gcl;T_=c)3 zOf*5Z3QO&KbQEgk;7EzX`??{Msq^L#4Add8j=#{bX`--eh+BE4+(p3QC4!bhz7hc` zX^5#N!14nn7J%-bSi)fRi{g9mAZ1|z{nAAw_S1BulP(P09D4P~pEHo&jS0Qf$hW=f zZu;7k)3ydK&P#|~c@4n+J~v0XJKAYo&HlNl{57?!wWa0We(&?MWLN%n;>r^zQ!PZL zjT$m&d_Ar1xYReynmo@8Db~xHaSs2?1z1`0@tWo3wTqp&L4K$}j4;DXMVivMniRIF6|-l@PXRZ$zRj{6?|g-emuoU9q!2e^3KjHopb zbvuEtzMl1c6j`LjW@?HLj+)bumn%;Pl`l&6;-eWv{dPjZ1_*tE?`b-~>=bZ_> zO+TmsC&$+%lZ_syCp;{lophQW88V?W!RS!=^}p3amDblhnI1Jn*4&w$^oYs9(B_kz zYMwJE{JZHaiQ9Z8oxkoF6-miPiv{dF4_$0FZ5PBC0`<7|o@pIZ?MYqz3L~rh%X)A7 ziZh?VGnu{Y3ht|erG2Nf!tU#}6~f_#WXGL@Q?NAA`F~qw4IX4sw3gw&p1n>fco20s zAUNS~hdWq6eptf~sC!~)AsbLDM)i7iryt)nmfHbFr8$|+t$Ms4bgJnx`y{+ z*7(Zo|LSu-ywFV`?pFh+C1^IqPo1P=!bA%I#jgt^M3W_NukHH60*YiVU-;V;%G=^) z^1+74V=HxF@3~E=F++w4#_x*V;M%V7;TC^%RWgR5*Zv{op>kpwGcp=BjcefQLN09s z6ZaAhpe8+^7mpF1cC$>zuYFfaN(0&-WNy26t@YD%p1;!vw0hI{?*MBtbiS-LKlLX) z>S8h`R`6K#*yi6nw`+j$I1ydRIEHU~_FWM?hG{!{vl7j}5!0-H`(}k``G>I2QKR|U zjgE6v5D3G)edc7)r~JTh*Uhw@Dk9a$^Rqro=>%CK6QV3O@9>x6|vdmp7Zdxsep6 z%er1X9&A7MjEUTCs_(bj@IKD)DN~ZNo8Sn(`D4CuNA_5>FoGyVPR5+LaX9*{W4Aq1 z5f7d7DE{?&`BxTp5>U=1fISUKlJ_W0y*`^9*u`0-82ugiip~DHdw#;8u z%Qc(n4Hl1+bPq?Si7~}=H&DBHh{%4aPTRqQ(`9}0-qR6xCUyYUK6ezVNZ68)Cz_h| z*;5nxncwkz_4e@)gxfBu@>=hGR(pYhkl!iQ*=a5Q(B(|IuzIDBO#4IgdR4X2r4>cF z*TmfS?XqRs=aPcJCD-*gP|aGP&`_rAIIxMI9^-v&r#3ngL5>8RjAY5r1Inww=~q@h zJ`~*<&G9@5#;oN0ATf%g|EF?ujfZMqU5?A1EyU0?in22?F6EMt%ji5~?_Qkq`Ro_x)q1hk zv(|sD|Fiy==lA_N4(Cm_b*DEdap!PscAckyF&pcg`>TTB$ALZ0628t{@T?BT^jewD zgZ94d9KclK}B$=#nQ&wDhM%>!`{QUmw$ByPWZVrqXc-g$3{g>t$61}#f^_%Z@ z#!?@dSWj*h{-m0%c?h&VqS$bD`9bK=Vu-~ou25@!s3v^f%E}~JarxcjXJ+qYu9+S6 z)~jZBm5HrremQZE9-V*sb<=OTe?O-#aqJjIId>&o(2a@zw#3#dXF3g^=Nw)mmQ5z8OT=s)L&6hdH#SJ&l*rpT#>z> zpL&{OTQsgHF?vP&hEp!kR4~~jw5%9jOG;AM33#d1nJ*hIAYh6`bp!*1b<>?*RPm}j zbh{L}Z{@$x?6BP&hPbhdU{6%{0YnTL2QnyD9oQq-hGvt#@9Y1>M^W1B8)4x0bBO0} zYjF`DWYG;%johZ-k*<=1y>)P)aN zs+uz}6Bx=R&5pH<SyFZRr`SCc@^5gK8Vh36Q66S=TBh{?&kq!WL8J_#E|7;j z`L^B&g0v>Kh_>5wN8fvmdV`m86g%(J<7`mh%(W%TS{9Z&;bB=TzP(7V)w&_S`_bf+ zr3voE{P0_DMNZ4gm+|TfQ!lHI&@Y>~Tqx(=LM{9es5vS&868N#=W(UQvZTxu0pN?@ zoh1Cyr*1t`WMGqP&y`lvAaeb6afz4i4_9dFKwq?hq9QT@1U{GC@-ALvf{Yt*K<8ck zoDYnI?p{tW&=TMg_;CL;yGU!1p_N6&3CH3HlV%7-vB8VTkM(F!2P&wFlV&$oVtjw? z%vdGlR;7SC8s=}s1sbvd=lKY07@Yzf>ok+{a=Myjtu}D1xS*Wrpmn_eZM3*&YN`&Q z0Sl}yy2TdXSFSzVa>>c7y1-Ux1Ld0J6a8iTIO`Aqp9G{vl1R3Q%8}CA0M<(SMcf?@ z4XB#kx@`(JvhbhjU|qN7k~hGBR6#BELGzS~Q#PLi{!Y=RIhy5sGZS%mCi3uP1Nw`Y zhjlF(U(6O|W_PB=N@6D^s5r_?)7q7gJM#tj!Dv#q-dH)ckn&ZF7#=)3q?k97$0#{0 zJJ3Jo?5g)K-Ri+M|B@YvTjJmEMKnYE&Wy_8MP4Tfn*b&;O+*(w}xM47pmOi@k{Nt8O|Q=mWE?oTftI0YTC?TREd7I zLLB2yPhT8giodOISG}ohy6T%C!dDoWYIL>3t*UVbu*XInwI)nJa_79HPGw(?6BL9s z(mLiPDLVJK0q`Aa$9I7-q-p=L+?>kD+gRKk{KE2-p`ls2r6Jyw_lZQ%?tt=lXivy5 zs~Q5w45jG_poR$&B#su5wLHS=qdxnu_gDGF1b<~^Rn^-``eB6?)jaym4J zR2_`q#;JM51qIZ(f@HHNXYN&IQ551^g@YzYsbd7Fa=#LMvQcs88rWGRHoCjxijYsg zidY1Ril>?#*gXQfq2H4Y!L@>Q1>XMza3<%>#n41T`is!a!J#0^a3Pa2$hz}ZcdWdM z-HN#McrvqVH<9_ZjA&XxO(gC)_mQvUa2r@j4o&K~pp}isdc~#~S`t%^h>aG{@0VKx z-$M3>wbt^XX?(87u0zjFSs_u!vpz4p52Bn=VNx<9Sba|Y>FgOQWEaRxkaLKD(QWRV z#b^h=Eo!IYv7-)xr>HsM1iZeO{yzV|JQED;c@+}cz?&_IaLzj<_a z-DvH2;&!2f-yzpEWN%NCjV@7riT=$vrV7C<0G|~sQ#n51(sRjQ8K#EUCyg7VO#1i& z_@rRKA$x;J3r44`Rx9D`DT@adzQQI_wbKK*V-vUPw7sverkCUNM(3^qAr!{aV+cN= zlM1&$a}??PlRX5!3tbs|=iHE_CnN1U_(RV=oT7!Ci2R+h+Hgg{sb0TTqjVS|Q{eT1 zul-%=B#=C$Du=R)O5taQA%B+5Yb4Wk`R#Yh@!QrAKdBTAZ+w=`vIu$71jIgov^u?% zzB%TBYjb!_@~%P);A54}Q42vA1mXyoKNt$8nCcRV#PL=S7~85}Z}lWTM#{;ZICZM7 zG&{J@xO?Xh+)E8uIQ2B@g2t|NGHJ4cW8xY>uOm*O_}V zPl?(ZIp%^z!0BR+&sq%&`dBZj9Bpge=C!7%ogl3eTR}{%V zii``4Aj1J+fYCS6!x-!7n|ff3aYp-b`up`TSR4iu6qqCVuK-#&iAst2^8h7FJ!=R+ zNuO|vM8P>y-N}(Q;q)*PH7v*`oPod^W6&lHL#z?ncrRuzW?xai0bIZB`}LmWh)8O9 d7~-(4y`DZAtD)+EW59U?-p6UI$>d*qhp+P~AP(T_%QbIx`rIGFs5fDWYq(Mqh z1O@IssPAjt@Adxf?|JSY_a2{dX3w5jv)0_ zwQ?kslmvmbE$xsVAQ+IN3j*u=x*y9)?`ek25IySrNJAyFVRpdSS#pnaqd3ItZY4D>1YEtdZl ztAflyU_E)##&a6NNW^G(EjPC=Gb#ZY;c>rAi_2mIlkS@TJ z*w+S>-CaH1Kw&-uNG}I#q^>g1HSnbGZt3FTX6cS}vG)Bw@~`EAj=-8o8wX1{ zS0A8VbZ0`s5J6E8TnHg3jP6GpT@Ce}Yr04eS5J3qfQ@KA{li8yXMf-!nxyYED>*m= zYyyFmoB=*6Agx_(0Jdu&UF=ZyAShZoy6B1?D0iggWkR3y40H7=K@rc4Yc2P}ZDwkw zZ?j>$$x&m4+#=@29qSUUG{@YPlU<@)*kkWx8Q!si;+3pTl=u|d`U(zr6u<@ zEtpld#=$_myv55&=<|~|%rbnnjtS=qhii+nx1~Lnnuzcn_7_aaEE88w7%G0ymR7_1 zVYTv#ng_XWH1r^wHsMv;y`9usr~QVeovdjte_`-k)CO5sOoZCdKCm-TV_Z@pDH*?c zZG!iq*pdaj&%u+2b0H<&D1hY>9-q2{ZcsDdjoCaYKVXYU1ApxLLv6nM&ne4!^Ox6*#IXo66NJ-Z|Czn3| zw8b^ra&I?i{Yx6ge9=cZc612@?t_xF&L~6M?W>PextMRHKc}iD^4h)=_N+3&y~cTM zQq-2Vn!3{L^kV9lx7%DA;=PhW=Vj!@Pnq4HPkLawR&9l;qj;4eJP3>ZMUZk^AX~+w zQ>2ajHdXQXH6JD$EY>L6{7uCm2}>-+7B&K$*YBb}rd_Vu<{2>6xY*V_$(70yrat=8 zaf`Abn1dl!sz$EYOY=gVwbS~UC*d{_S>aE00TOS)HhyIG%&^H@rOl7JXzDh{aQ38Iw^#XgiMhaS0xY;!H*_XD{K>r4)Pv zvS@s+kI*9GR4Ujf&@I&4siz!9sXWX#F>Rdb>Z9kY8NjF1Vg@I~UT_s++4(}g2-OG; zw8=4JD+;QNNNGQp^Uyy}rKJH!qKY*;)ba_oJ58hGsHwoLU&uk5o)A3S`Y z>JWH+`K-#>UB4tPI;H@yBU{JIF5WCEGv^3O_j!FqX`bhzGW#`#Gda9jl1hb;eSH1v zItJ2LmKm1b-H2ix<;)f{b1T-yN_vFPoyw0iIopZlSWC!VqZ&e6KaR`5wq|~g?Yz!} z3$YKI5hhM;Ptfd*sKZEub2-%J=iQ@l#oOw6v(=&m9LmoYsNL_bz4uU4;P$nY^LXQ+ zgb?Z8B0{c3Vgy&NiL(uvZ6{05OR3Tt1_t5KsJ^!L^!CTS^>OrAKE|dJF6XGvzi%1n zCrquWQRAY29sE3$DJQf9Wwm4epqPp`Ua~^<{_KpqzHySCfjHDR|N2&8&MD2tfMWa* z4rN7yjgo-t3x{5>M(|$rTV*%4ZNVFu!+yjP;aY%lhbH0tG3mh`h6 zDE49+7O8PDbKULgwc<167K}S@&#={$w0>KZ%^;srZBP#0x0_8ib0z}ic!gg9+(sYF z{9x3fj)rkn`HW|k3fB{ek(4FKLG-g9OsUA2l=uK6x!N7m_?fQ*5tj;1oK|cb&cy(l zh)K0+?Se`Q)Po1)tF5Vy(rodurTw|uc+y7uo#an9>LCVFFWw2}Fy@H)$XL9C?IPqh3JM7OlFX1%fj9<#9{A?rT9c5EqS~g1XY}xW- zMl9C8V&1b69%#c=#*LWbQ#Y-nEFcfbZ(T1H!ydb_3T<1I3sR7Kq`PZbT_;S#kYidO zYeCc;yA=|1cU&jgQKQ!)G@zGzEZ&Xx9oepWlFlHBY0>cTD1{+CPWZVSHx2Qw&E}Ov zIr0jE3kKu_d<>9OJxKQ96q`aEmRoTg;-_=UY#xZ{(?7bG7e6=^w^tGsFX4eu8j0j% zzDlkv+DFBr+eWw_tLwvRIGj#Q=FRg+wHn@M^;(#bP^9ocLiEgbcEETgME7A$SWB)K zV_L=O_DM)D>Vbk9kDBz9TKDJ|!F*9czILLV(4|LH+N9Sw3}9cF&GQo}QRC8P++pdp zRgrfx>+V4;W92My!dTewcN7~A4OA-p(;k3q>Ld#Yi}EM)jb$)Xl-K3b*d-fYGJF&< zWK(n^z0AiLKLh{7n^#%ve8ZE1+Lc=OVVA>pGym?X#TEjd(Bg)Blb(e$y!o+|IDH8m za^Ozwd@8(X?VJtDT{*A23&E0hPZc1E#5=-Z+IF3!h#RIs6lBmxi7O1y53?>2SH0A#Bc;;;F0CBi^Qoms(V!;wrCB zQQ%Wbf2s*~dw5~Qb&9!3Cn@}H$Cde%`41n?3eXn3iDBM|(cIKH)zjJxrDA%0xhdxG z%=KGa!w)dqvai`)ey5C#QzB_4Dkp!N^g)+$rv1hir}aqMl!B>m=j;Kmdg(hW<%G^# z&Dru?`{EY&X?kro#m_!{{=t<2&ve}I_48$}hRCAs52eC$6m9$mntNAt-h%hi;ty19 z3Z9o>IvY^imxShcyi<}w2oIkD346L8WCW+}xK3bjkrgX$IEUvSlY^7H-QrA}s-^{H zhwA%VGpGlInrE$c_zwm6cZTHJ^$cHl)ljutJ@OT4WIIFW>1vqf|L~!3#!dTLoD4jS z5{C-`JG89?j60swI59Y{wij<%u;eg28}_J(;1F3_+V!~eq;+s`oa=hfWQ-!myP*nj|Ao@5ofeG zT1Xaiy%Mh#eSe|lsd)R$D}j5fO^e1m*P6qldxslWEwhe9Uq8-%5XzLXN7AQ*s9XyO zOwTO2VnR!wm**&563uSU%qER!vd@j7vQS^EoucEg$#C?!(hY|Y(wU`K34bKNraj)9 ze=Ct#RNgs|(MYF0x0cD8_J*wdOi+r`AkDJybM1%GTX*giGfsAe3{s~#?M5(%xzlVf zTWe14KKNWn^|__%$ovySZs`?W+Eu^M*_!T=%NEXe$P?;4p+fhR3|`dUs%x9edDs~F zq{yiPFX@%i&c$%<`V#4ncnPchapnb}wyda>>Qj>h4~%N72V^b0EY2Kv_L!Dmc z4!K6irI0r|N8IQz^e!IA5Z!V=RJv>5W@rb!9%zbdrU&l|r0D7JEPwPN?!D@a30{4P zST3hTrv+p@>PejHJu%E$b)fBlo*I z;Z{)3$k(T3^AIWIcRz==UQizwIb6pQ7JojD@}^F`IPkd(SKr?t;Sdo)EJ*oIP!l$p z%Xw#tk*4=TVJ-Xm<*4vu-ieaN`Zf&JfQ^l0yusBjkE=80}JGnGLes01=Cuu z#wWkq$QyZ>wzqLac;r8{u{N>ZZ z3!{Ohma#j>@5i$)QE^PtOqb9vx>#tWlpXCJL>|;Gl(?o`3iu!{7y`BLFzrz}z0&3; zgmAlIftbu?e5w&9`Dj$kR^A`N?MZ>kJLR!8)1dR(X6#vZMy_x*SMF%n5J~ZQ;}S+h zl=9_QIo4yzA1(CRF)7}KQZH^QX(s4s867Y^%8{Z+fD(fr-kU4_EK8nnTKV$Z@~=fz zz*I`GY2V`8S0@Q15SCxWoaOyt53nQLf@ZTm5lO5N$RK#KDO;Y7WD?(9QEH{F$P}+* z{b*BobipE_@m$9&{Ps*bLZou~X+ zgwm>R)NgaOIp%gQrLBLK*~e+wzPBHylq|Xd*St1sPr#BAc18_+kw6LO8h(Q5!_&1| zM;A@sJ5uT)Uu{Cmu8zsLRuNK`wZ}S_wI8`*!SsEiQDblEa5esFKT36#N z9h8w2d2MiU)2ABdRq@m(RF98RBx48n;e{WDGhc#R@DN@^Ei6)OBmnyc-(1~| zQarMHSY+Vj7iI|~&TUFF(sQ0jiqMLEjLSxxaZG8qEqss^Bx*a}0T)v{6GsCuUh!3p z=!+AB0+BRt)xy2YFQqE>JIh?hOg8BhV8+F0gWY`+Y)g-^0>Pk8HsK6rR&wir z4XCJudxNY}IfU~LB@`+&M%hOLg3S4=_-~=;vKa>W%{oZfo2*ljfq5Mm1SYmG*jFH} zD#@x8YA;D!6Oi%!R-eg>D+gg#qn1In{8fW#C^`^yfM1@A*)PhLp5s&AYm9mmTU9wP zI!o~Z{ukWL(@!&z=@p3+SXB0@3sL|{tfsd6dMl7~G07$@$;{Kxd}Q)a(jr#1Ln`tH z@H*Vo_62ffD@I!&feCT=7XBShnxWAZf3SdOrw?@1l6kc>esDb%*#k5OL8hVY zpnQ6YQs}SCzieTf9KgnHpJ2E5Pt8#UYVTgOEu;d5){#QrpeRlp^d1=&4{!iKFg54$ zq$T!4FnlZqMc2A9Kzzu~j2rX?CQlBGRKF#vLS^&|bNeMuA5LLbi_ZhtR2)Ni*~hRB zL)CCuv2&@G)p;G++S!BU7vDj6*|_0@cTjXxw2AZ|HN=VgN&1O9IGLx3DEf&Bxz_RW z36Z!(^5K@`@~N^rc^ziH@~JbQ^FCy6jX8>VU_Kj@l#1ggF^Sm*|^hB1J3yW#IW@tYS-0$48Ni^PVYh z0xF0%p5%zp>ksVLz{Pss4D}wr!(J0O`Y3d3qipX|)3l6f)8lI^>)~@TGPw8{>6CY# zYa#c)Ja^pRe-mRk1DalZ?r}&A&HmUFCOhUebw}0v^s~sVh`1M*J?i^lhKZx*);)JU zI^PC+GkZr&2mEpIn zXXFY#XLO_>hzU0JU0O6mpp1}=^LvdCA}fRVdF?jt6B`MY%vVTzJn?YU!M(AFomvX9kl-XAeuwAk+Lsb+A9a zeh-^-{qnrswb4A2V(osB)21D<(OC;$d^$^ImO;9TFtsfO7N&?ZbPTii8{ZIi%h>mo zhBONaZ&$@581T4OVgWK1?qsRk&GaYGrQ(+^{6jakGEVed$(f=H^r zjJ(yFr@B_X>{d22gKK(mPVt_>h8@2V^Gx!)TJ>{C-;UV{)US?)G9*x5u>%RWec^w`K)$cA8wbNlG|hQ>2|Z?ZYhG7?As#ZUXn}Y41OwyQO4*=JMw0 znB@b7)`*7ZG}o08_-&h9!31-n*;VhF?~G!q@Q{hQZ*o#Q6ReMe#PO z-fmdPigTlbT?;>*Mi;d$TYhfk$xBKFk49HiML;jU*{cL8VM)8AwPc=%^4g{u&U;%I zx9DQp0{uHv)6VfI2(0w6$m3Appvp?^yHDx!V}@u+_Rg@WioLO;ZKVp%^ah=IE1hmA z-1@<8b*(MD`4!m`-c2)tdDw`MRK+q0E?e$?a}c-b#fJjY5NrD=A7VC@tJ$xz%dq_{ zSSL{jE^W%E=vbDimoL}Jm1~ebCYobSCGj0K)OjRJDviHmapytsZI=ym5f=0JTX$F# zdM+@g-B7MStw=?!5v@Agb~YU$=vS{1ST5>(svR{^_^O}#m8^5<8P2u7aDs)Pl%vpV zR&wvcPS<-|5e#3a9)Gonw?ypKs?>TUL(YLsEvS;;ZhmyT&qSyvXYzBio9&I3(Bn2% z|9i&v&Qx<1pLeXMGiA%F_A*5fL_5}7yev30^XKb)Kaw-D*^El>@UF{RbAz(5-VWhO z+7-sKliwqKxq2u1`rV_muc@4P*Dp55Z59z_O+pnlm#L!;*o!9c-_NX6P%K0esu09w z`nmapwvrVvY#u^!V}m~#dt@PIpE~r65}Y~Ddgr3(vbb*`t6_PBG@p;?KIXG$M|0_H zTDNUwrL^*vnkU+8wqd->$`#diZ=WEKJM-2sY9Y!CS#AZlhV2#3lYa5#maAh+<}TrK zz~OeK91kpr6$BHN-r)(Dp>*sCEh5-UJ=JY>b6L^&jlU*q&I&Uou9nptcCf299|pvSV39!#nFE`lr)`qLl>--!PK%Kw{dmqQk2g0H7M@O zfG;INF}P+a+akchNU3*ei)nS!(ED7C`lA@qwx)c3#o4A@q%X2r@%EPy)wj2@9z3oL zRm-%Y$Tzp~B+bdjm29b!MVvYQxcnH8@pO`bspW3qL#H%PqtZGE?nJy4y0qFrEgEeYUf5)piC!v_$c4DG4uq`&Oe+wtv$#}Gs z&SNWG)~@)}K|HDypCe(yixo;X()&4!$kVL#V|2m>e)p|`g@NE1l4}{tp-v|Ir5T>F zgb>s_L%LyV;m=pJW%{Hevn$(M>y`|honT*-Cji@`Wsn3-cw#EFgQI6sqzH9imA zpImWa8LAhmayegNlEbi~N#}EV0!&cw-M{$-z(ATxLGclhmEF7f59yeQ+x(Ugqo%l+SJoUe+ooo$a_{851~TpatJh z^_O(1nJmt$vVLpND7u05%%M?5;H_m;Md!8sN2!L`CLUCpvC5~!#Y^^QL7U@DF_4x{ zWgejc`Tbo}0cxk^El~}agnlN{4wE4v=;+)ywWa`#(@MZOi_fJjX1?^@c5D>NvY!g` z{2FtWBhP%qqtL)HrreF1yf!3lSZ=LO8vi89CCga*J@L&RLtMETe93YY!R}#r!%f&e6DOT*wylpP!sXR!!gsQWEbN{xiQO zBNb>v?-@gSt4B#-#IDMhJsqe81s}dZiKg+{%tupsdHH5;J34Gh1fV^I)hm2heC{?a zrs9c|brxX%`y%g5-dW9;`4DaF8pfkWu%4|D--@BgErgM%?gZjS362*Dyy(!GX*SXm zqMEfCn^PN{TUV5qNQKDfAL^XTU7oROuDnh1)Wx7nK&onshnf3|90-0Z*(QMHv!UqN z2o^k0nVEg<$~*zX%?$m-@g5WLhS{VmUJhVSvC`NjYiVrDOAjb9^MaAN{wb}sHe>}i zMXj1Yh$Y6bEcBIGckqKRG(S1d33{?L=9dz-k;n9b`w}KCPw1qm;JqYsg)jV448WXl zU3@h-`#y+}x1Vl=sNu{ZV`XAdXtUZ&IrZ(=%Aq zAvi|{1=t%19z15jIArsTi48?SL;&{y&IFbsbuDi~h@gaPOgxmPw33N+X#qy$SbL@_ zY$KJYS{Y)sUyYmPKJbXBuJgGdSB5!$v&5>P){|UUonpj^>zrngDHLdAD`!I%deq*Xm9;n+Ya^M0@mybhES<8oA6I+GXu3^2YrG z?~YMHHtC8V4dp!M9e=vQ)Vh^V`TPj{iF9-O$cb#TmIaITCa8w{<+OnP@|#CBE%&o< zkK|+?_-M&!ez@_Nah_UVSeS^jk(r4pOwXi(d zSZxdMXVomq$q@J7xN01oc{MGJ=y2p%>QMhZ?LkgIc-P!=Z(fh@(sKCf`pW&Q%_F27 zemSgZmG`q&x8!6{7qw)lr*1qhDa#C4EhIYBRsbqp_B$xj72K6~MeQnyp?0sG7Zd6$ ztBp@lZ0Y6zhx?T)Yh%`EiJCUQxxXeHH?%|y0>@Fqnrvn}$ik>uy)T<^b?ia<-iv6; z%*`dnI^Q+;=pF65vcBM%hLXV3484&ea`U7A4N|+=KBs`7w;u z&qmril~1%U^%qW+V&wMAH*tP88=gmip|0gN2bpgo-cR5Xs=aZANeR@+5nS8}>6lxy zDEDeB^sfn*o%1+j5zTD)Y4(63okiNutA+MK%&&lcB_gr0Tb{P<;Gmy}kF@m15=1T%I^ z&kFltdwO{Rwe**;wtW9SKZ|gqJz zz=98>+dRc>iV}hejyJd0*)Q>YEOKx8Or+XunfhB;&Uel#$8`B@ zm#a9M{}L#Y6jsm^bRt5=J^Tgk?|Hj`PZvmRJ|2!zzB466{C3)PK?<7@&kd0*?^d^>LjL z;sFC3hDrM1+h-@{?nZw+&?4RS#bPAmxQ>y6$LT(&sGKR+-A&r_65{fOM11q5-JL-C z57W-mgO>2tv&MRbuCx!^M6>l>26sR$B8%bzZpE$=!>dFD&WbJUvYn#*OiU& z$p?%J?%TG^miMY7^$i|Uo%a`OJl+))j?2MC4(X0v48rFAFkakQwc^xtj3Jq>`sjDw zK+#yU@3<^EOAq7;oDIN=Ia(Te>H!cj0WB+Mhbx|FT%5h7I~vjE0fL?&(g?wdE&y~3 z4Ol}%%oH4KZINg=82|(`0||?Pz=|L-5fIoCfOG+oCmJ~R6=3&IH8j#p$-&(Ng>D+4 z1B~um?kns}015$U6M!S2AgHh~1O$Z$i=vxA7g0cZSi3v8pd*`ImxO3WYNJ|*2T^liS9?!(&vOw5EPC89`#Vj%ZBLuul)aog<#`<2>f9@=mDa~ z1Bem=`R@pX|J8QjfE5q|&J5fR`S4g(>8&PC7x0RzFr z5QIPg2_u9-2n5iGFdQTz1e6s82!H@-bc}uv2Ldq7hyeA`ZHkGC0rA&7bbK-xxCkLS zMiYS^6gmh);U{$wz`(z@0h9&$JxRlb07i(O1cZp#Nh~G;&%mW?qQ`WiOn^e6YyHy-nkWc_@Z{aM2E>3#?0Y~f!dD$Y>%q_A zgx7#xpmpMhaH0=ro%kM3bmf~)p!MO$fTru@8E6Bo8{Y$Z|M$sG5d^UA z1osQf^e4b$^kjcxb5CHr-zG>vivQO6k46|6{ZB?%N5UNkioX^}Xbyrex@OtI%hWdB zD3vGJT4E9qEJ;m9bf(aaGB4b8HOa80zshy1>5~ahdK0Zqu)d-(+xm>juwz)82fN;k zJE!!(AlEaio>RqBxb7}jA$KK%O-l0&91)6@9u7srSJ5c==@}aXI4rNBFq_3qx z^p$_N%&;FV^M7I(@WubgFHN-X3nE7QzLKYAj-_>q%dasqo{_VbSvfD>Mne7SZ8@F2 zcb(~WzRsP%)l`ajZ4W60V>pL+Xx^oR0Z7N4$=;gHTEh+6X+|dcKalmK?EfQKLO+oO zM~E0ew9Bh8CE#Ul#BBZ?oiS0s=31>I% zx#XfE9txr}M9+vU+vY8*Dl(V*@@uU z=LA!5j6vZ8BO_$}Fkz@JL`VpyQe!AO#|jdC&O|*{R!sX)?1BS^mOcge57eOl@Z)c& zfkS@QW~q3I_znoMynn`Z+TN&73G+KPTuFOJ>(a13!ywg23~sPzYGr{W$bpmK`04Dd zmBfPgWaiLd>mvm)h!flnH`TJ zmy!cJNkxPT@jRo|+fTn9w>Kdia`QZvXjXpVjgX|htG_x%dz)5=dU2mk>t6VZ+S>q0 zZng{l5G5LldA+CSVyZ5bzP~8X;xUjNg1opzlOXbBO7@<00_} z#-smW=5H7enBpI`6ka2w4X6P`Zv-JEtPiOaGK7f0wN;?=rwD_M>n}hmm>>g#6rC_3 z5qYbXG5A!-1Y^}e6H|zbIAm7iBj6fWKe(H} z)p_6S>}a>hwjn57@X6TyQbX9J`_nY53N5BSaBsD~agO2ZT>0%A+?1pk0yDulvyYjM zw`mL8o|oTn@9?)dWuZ3aa;{tcDZPq`EVU6U*n=B6D65Fm$rwD3 zOHN1LDSVL~9-0(i%sjdsmA7TqW*%v&ZDuxw?PEJa_F_6NnjF6Vl>Ukdxs#04O0Ev< zkq$knrd`1HYnblNQ=r+=-q-!|c2DHEEY2>f-dS>EpUs?-C4%l=DNELJ-n+wN7`KHZ z(T|DMd^jB+{ltHUKkCBNI}NU_rw;d-C`%u+SRIDQIIG{i(8Cw5NLARxjeqsIEajB| zNv)YXf-3ii1QEMqa`QrwkY3+dG5_rE+C zs$j|_<@9&0zMVKYSQtt@=H*vd^Te8Y%scb*!52(HjGRBX8^C((?*s~b@;?bwQ>XE= z0`NJ!p}H`GvF7BPNu+E3(N(?=Keio|1LnEULG!U&op=_{jP&t*oPqo`78&R}IKNci z{FuyQeV;znW9s6&ceH7(W)JPLn`ltrGkpDcZ*C)rJ%;yYa>-IGIE5@8xTZ)id2eMft-CfDBpHtC}6*!jUBScjjBkB?Q&-+B#g%~ zwRpb5sB`N&U*ZG->fS_W!{AKG(nl{5dGPz0!Umr};x+a1_T_wk{(y&atb@B665gup z&07VOEv&nBVaN1D~_Tic#n!d&85{ z`tW^yQ*!5pUYAI`Nf|P(Q=v!3~`$n~pUvwiR(b zns7J%V|kp)M9+SnMC{A5ZzkE7U1=oAU*T3Qx>wa}xO`~YFeA!*E^bzF8SflHvd6Rq zE=Feu!Ap$JS1+e&c(_>&8g8n#m%hKiC81q2QjBU638KfN6SL@uFC}?y?~2hOO}cwe zsp}(kb|F9LX!wO~+cdOKrJQRK_q@~<@vb;GRnhKK=Wdq1rE1t_o_q_YeHf9pMc^rp z)NE?ceEGIVE&+d2_sXfQGmI|g%!-ekwU!8~Wq9M^R(UL)PYL42;)ltpQgAEFD6r)J zkr>5Tb&3uUvHV`Tobr}3Za^lgL2a|8luffL=zhgHTyHMyqDMHCPjSGP)fgn}+^X>&3!~4o^ zA1-}RuGmN)>TI=HAPKeoz_Sr>ms!XBwav9rYnKV9Yt$c$%c~#cFN97Fo5QoQH;S0n z4{?e-#6$lu(!X{}fkpO@BSq*mUg?4muOV<~w=zB@3ElQ6>z`^leNKYfk^%c%ag&W& z{e{(dgUa@K3vqDU(u;K}HMNQ)k>sZ_uHqXkW#G~8H3*OpuE)c1zfv`F zpDB&1_SN+*WQ2{fV@25jlIMNn2Ld_(p&p?{6?tPRn8fzXHW!wa|Hl_rMn>Rba{39? za6Db?QD+wlh>&9CY6#<`Q4c~!OZi1xpKQy`wok&MsbE##OE@`Xw`q?^p1kO-h?hT0 z^*-=XYXrM<5P6Q)Iy)Kg7ukw?8>za|oCY^HF_hSt0*j4cO|)?@9K(i=>CIp865<+| zux2T_afYah?%3oKKGS{ltmX5oxg9)fYrmjB2;y&@R2Z-w{G%YGv>VCgfh`Lu9HRN* zfOU~41Nj3-m(Ql>$qCC*B&KQfC5DUl$3NiFVwP673U3NuaIq2Q;8g@ClymHj5N-!X zE>i8i9#5{47fP`gyGjJ-EoNkXrY@Z3#!K0d{i*|gaQZR|%P^d!!VEN$XYsrS6Lug!{gQR^1325G(i6+y8hOK{&ndsTn)@x5FrR&yi(Y?!$6jR zq8xvNKzJF4UNI}znSxi(?RO!*Bw3mqhWwM9H3ubj%me#Od#@ZBCghV5gcy6!og*O(%{e-QiMdSt&E1-P&t zdg*qGqM#t`r|p)n8$kSn!Tjt*vB^w=@>4o-Tq*PIoa)va@{lBz>f@5tdS8V8M|su~WM zG>qrUL$`S?X~sh|GZ96hw$(A#HXS~D?HZ9c!a69l?*=9j6sO#N_KJYLKv~mky}i;* zc#};bcOTB5*w({+7Bl?3^;yEv@*whHvF7rd&5J~0G_4WHj5L1Y%k~_Nela&%N7AeE zExVs*KHoXQQ#A9>{eu+#)`R;+2ypZV7t{glgchp*2OpX`4Jc@3p6;S^on zcoUX*iBP`U@Mc;JO_IYtlFdLML9k_A^>QTR`@KHGNXmF4j+gnwaoYQmH}qTV2AUpE zq?3cxlL!+M;(IuVj)+sH1(}wton&sadrSFQ7(|6qU=C1G<~o^(H+G~&e~~P}lzB_; z{?HisTTE(|(+yW%Qg6!d`1Ue@Z0{uBBs4dSq`4})ovmCNxM!?y40^iGr0NzGtsNKS z75q|#SgeYsu@HPdx&5_Fua=1d`R8;C2WYDbBFwPNPTO@McOQ7V2RNAEcBFhf^f( z26A7$q|VvE9VbI_I_KIYmfWaxTvtqwr&%>Ov1qqnSe7wa+!*ajn-4t4OX+8EaQys} z@6i`5rkWuAKZxdU9WEGfc=My03FwH$W69zVEK&Q(c|6;!7>L_DULowq3<;A5i3?oO z?$#np=9_-sYfwGAnM&8@-InvF-;jEq`klulo>(7cJI8PAK!_S8-uf4bx(Q(w*PY{En96EAMKUCP8>bY$==ist%4|`^$Sl<_tuk7AnEe%CUr-J zvTL&Gs^qM47Qtm>v2#mwa-*5LuC5HQXIL7rI}(4spg#57gL0TbNP?zBQY>HDD1D%> zi--^JuxY4OQJPKMG1C zC+}M5NOJf3>7HBp>V*?qJU1iNNP1H(Nb>VWneF2ItVHQvu~ah-zpYp~Aa9H}3>;Zh zYT&THuKZ5V;Z^b#%1$f<&MMF9h1@2x`m7?ONul`H)s^v1q6Kf7mx|VopBYSFy7Wdc z>JLNuTjvG_+{W{xP9U@gP6L*u(T#pAM10?(zG0pf zb!l-8(i*{*iZeu|o*eJ`(?FSgn>neK6;CJMTDUQAUJRlUu~>G9We-pFCgs!;61<|NH3%d zA^0oG0_@@dQ~>sA!SdfgC}1U!2+)NJaMKcSTn~UFfWvtZ*bpQJ1%a*HEv=o9C}*TC z0M~#Met&j90c?CjuK*{-U?gyg3?M=5fCzw&H~^3q08QZn5C^1$Kwtoe0)SP(C=e7l zy$1UNC#C%}@z|B+(q3P60mrZhn(=kGu{ zLjbUZ&Lac>kbs3dDfgqb0jBsL^S)Xf0C;nfi3G4+-%=-*=;T-nXzIl1og9e(J)GD< zKoL%A*a4`nuPFxr-}E)*0uuT7GYniWh8USvd)IghV05o=z_63yfYwE5dRO#r9% zl{NnsJ$d(ijQ_x5K>|Qg;2#o0V(YaN*5JH?$8f6Xd7j*nWJp>BwNpIAY4f_ZZURcLGEmYbpv-P1T5>Z z3x-@VYr2Rt_hCz`2RKlw9wAL%Q#<{;r+LDW|CbqnR*vuEmv{99ki&rHsyo=A5!NTS zfa?B>s{VSD>A%_oV4Q!b3|ys&a&UD~Kwn(StsoAAKt&7mM42`P&H4pZCo~Q`0^!m9R41z#g)t}pg3ZpH-FL`2!-}(iX*T3=^ z3Pa=ff2j)-0uWli=Ar$oU-Qu3?yq^GXycCl{i6p^I1E_t{h9~=jek(M5P)$0wVdeh zb%kKScK9#lgb=`=E&ZA&g0`aQ-@mjV&_|lT=D~lHFZARK`n8Irt%g$p}3`d7v!a UQ76+A3@QczE{x_-)KVh+AJSgy%K!iX literal 0 HcmV?d00001 diff --git a/doc/pub/BM/html/figures/figN6.png b/doc/pub/BM/html/figures/figN6.png new file mode 100644 index 0000000000000000000000000000000000000000..21302b27f10469aa4c07a689f422b0a40147df7b GIT binary patch literal 20344 zcmdqHWmp_h6Yn{=1Sd!!c#zXM$bPo6q-|s~ zJk!@mu;e#MI3m4dY8%PducCwD30*^&;YV7X2vkIM9M|4t-;fEnzoh$2Iy}4VuDVR# zIl@s!hM4^x^)XRx?Oqy=dqezs3~`MiCO4($?9vrsZLSR$RN3d#x;A-Ng^K&#>T##J zx!w8heXuL&$FbSpFs^$YU?V`cM*Wa#mNpvG_*JyIFgH}l6nc4w2(CH*!Bj^yMQPkI=wfNL9Qq8 zBKK2TN_H*PNxmSylhY4~=`Y?tlQA(-1L5~YkQv}LexSJq;j$y7bQAJ`h}PkJMRDdq zx0J~5;NgbgTZ~W`K$1ojJg8+NES4aJ9|T#*+3Nz$@Kqs8>#W4c@j*E9!KosMt3gPp z!M||cy#n;c=*YM3QjzLt`CrvELC#i6IVY7UH~@KK!m(NO~ZCLM3KOco*37!;*ncmdGs1 zS?VtD_NV%+raDI%x*Pml(4pjtcx-{h&q+&8UGmD1Mv;p@LHm}jgsmX`ID3hvg6h18 zslx0KGlc$7h(_NKcD(%I`XWT&j#APHKl#O6XA zRb(r+1D*rU1F8eebosxOO9>&v2iqOnbmto9YUe&28EHAhO3Y%mnc5B75%b^Xnb33c z6*rW*m7N)Ol0GEyC+Rq~H;40R3CR1Fdp6wio~fMiqGg2Ug+@r^OVmiz^(jPIF{YG> zy8Vz+z9=31sa1*(4lS15n!c^iu18+vT6{+)ZZYL`$>WXtWN{DDeZH4gD2GResA* zN#?0;l=W0li%)5CDAvfI1VlTa$UlAzj6`>2Y$!Y-ezK)=#Y8s#s?T+|$I z#fN6}juBst>9Tu`g4TC{KNZ_YpEQ$;Pb#n}Rja|0J2U)$J^MciJxwChA)_G!w!$n8 zXASoXUkMK&wIV%>EsH&=rmJ?>>#%c!96|>82EGqS$55pGkzZDD&%8|Y<>dN6kk*(n zkye@>mNAzWTYF%uj}>fi1qdn#0AI<{WoVbK-)uagqEe8puzb8h`n3s%BSXyZ9Jxvq{Ad%-n`tV!QT5O#T z+-O_wn>Aa6TxQ(yo88)aPgqY-mIwcqRV(N&9l!~AlA{RZc-9K22?kvCczStj_z2!` zUrt<|T;$!^URqsy?b+;o-h7x>nqlj$9?adO*j(EdrZA#VCZWWb#a_nqyiIKo&4w&q)8ycfBYk5Q2K8nb2o*vNS8mgS;n zx9r<4{djg;6dD-XD4}O!x!Lm0TGdz8q1MFN{oLrQqO9mw_BrY+qq0xyjF$2t(|=3c zie^%U`=2q!WAxu_rT1*R_ zS}5s)L*0u+sf3l{K_A==0-4!ZZBK0z%182c9eX_kpGZfBt#7}2D)Y-#E1eB@Of=-@ z92@lAKF9iw=VaXV;zmTo&68VIYW!p=d-4Qd?}zY}w8nYI-k*O-`;tDG^YQlMm@Id}6_;?o;;V}Lg$#FHA7hl+no{xEH#6^#(GN)k_MEFqmTKFcv^osqbG3vF z{Y1~jh9!IYAEV_`omt2A6;tp^^zRD10e zo`9Ln-ImYCpR=1396T;FFPmP>&p$r|9*7K*)$!|ljKkcqtKPKpn0|2G$$9EHy79Vk zz^*1Ee|tvi#ZJP>=!y*{;85T`2;{~N0v#HG zKzwN+5Rqe6tBL?{0@*}PS_1U)@8?HHNg{9r&GD0t3-JHbe}1KgZG=D|bWd+}ZC6!e z4@yU82XiZ1GfG!4M>ER*s50%a>{Po1(^5>ympXK|ASE;w6vVHm&mY&tH9eG?Rz7^; zXNny2#t!{fbwY5EdvZYGd#EgP=chR=EOd3$^tNTOA#Sk&k}>#}TXQ_%T1K_dYG!$K zh{oe(COzN`&E1du$#JjV`sPOcThZo2dC9Ex}t6=GJoDEvZ{ZOC@0PAhiz zaB(%Rc%W+ba^`yjo5bIE3_Uvv$Q35spPY<;G2Y+X`yRkOJ3N+J&J_SSyZ^B{Ad8Hzfnu@M`S(cl*$)wHVWAWtppKr*T~04sR(D4q#ee0ETH zxxZVwJmSB9OraGb6CTQ_`5uMtF#qgz330M)*wG~U_M{&rL5e-L(a6E~=jp8%KC3;nCzwy72tNEnzDqrDxp*olEyaW)cBn&yU&{zvwS;vEF-0tNBTy2%s(Yg+v8; z;PnO?ceDRu|K-`t{_BYUT?p3%iAAgfhTtv_Gas>$LFZ%V!xoo(@BH@8)BCM&83XK6 z+o~ubH<dNM)gGZ#V_1(>`n~a6bep|ZdQ=gjPacC?aleJK?+tzgdw4E(lq_T8M-GR`#J`$V zB=5hVBrRb1fv`L{7E*xygEMfbyS#>&^uTH04Vvv)2sq%GCtrO$eh5h^^f;=5x%GFB zMNOiaBw+v#vyOeipY^PY_e^QZ?hyC&m?3o`VyMo7PALcNpH?q_N>=W}}h;n($*Em3T}yPxM|HT?j|-&I!O3z!@NT4sgWA%6oG+*e0$~uA(-371|z~Fl!fW z_ty~Suu#_UOy}DY;%J*%;zoyMRCs#rVvDW(=?_nARIbl#U6NmYNXA}6^iLFJ6Zutj zEb=?T0%ZlO(bwu)%NzvZ@Ln~PtW^pqYCNZ6jfg@SqOj=M_;}Xby%F3gcg;H3N^-YY z{Fc^G%n??^e8#{N0J;byfmcVV`MRRKHRNRIg z#n(qR-C%@T*S=_Zc%xoftLR##>{9F?sa3ahqWRk$&E<#)1bq_~j-4>=8V(mjR!z3# z)cd1^ce0Sj!I217svuQ_>Zqlwm^gYZoX)ZOfKZrPhc%Ncs-!marlk4n1p;X=!g7Y$ zA04lgMXQY^AuLAamX_&A;;+5a+p}DfTi{LhKE4u8>isV7&Y@^cx}PKdWqP!%oLQ_+ zJ+I(%pcG?t;HKw-UFtZ;N92z=$8{rLC@GbFPb=Ga&YW&G5pP9wBN&D58OHF)=+zv9onbyD^gj z$@{#mKnd#=g&$_+nNgJaLe>}Nb32-YKPbXJbdYa4s~;L~x%{U6(Xyh}|A9x3iA#-_ zQm@lmyY#mTwNVwf5!+V>l@1Fv6p@xhAu)CCH5Jm%BF&-(YtmI%Y>8_lWn73Z?_aCb5`to?dTq?m`_K7x;k`DJ0z3m56+;R?JN;^e|nLQg# zAsJ5Y+VS5At-WK@IV94OI{8kUjY!&6bJy9X>ZEptE^*sU4Qe(8x+FYy?DB04(`GZ# z)W6?FJth&#|Io2^Vj7wMx$w+z{vA6HDNIa~4W)=EK$%gm+6PCOmi)LdqKYu;aLBlX z4C>;>`1l=JRZ1N5eZ0%Tv`v6uUd0OSH}3qx)*v#x2qT4eZE8b)V)0TtX*e@x*^1%q zmWJMZahI5nAJ0~aS3MP;+!7Z3s<8_B>QsfLIOA^Q@Ev#Df1DO2*N^no4tj}U8#58r zJpCju6G6`l5jDmVVP{MwQ7%EyCEdm~)(;7cU9FM*vM^aZvP(Q7nr$7Vg=Cd%7Q*U< z6O*To$CxvK7_2s#)*11elMZ=<%1dX^WI;EE_zQl6T}<@&Kw#hF5Ra#iYq8TZ@+`RO zw)p{rfdYvQ0q)0K4;pI)D(M8z<7g&Q?4~wt{#^6+kM;OEBv%>n&-ZezZU+z%?9Xo` z{WU_@l71-f1Xo0p5SwF#vHYwXJ343OGHLAFq%w29{2{HpaNJI`hI?DO8EwO~>A%l0 z!JVhMGm>)jBPbF@Ix%p@Y7oc!K1I$RiTW}E?Nk`y;70O;e3E)>u`oDK4jBPlG zGjL@J+DN&YFy*M>DIqp>ED>Eaza{ZdO|fD%#4pY5b38dcfd6uOETSL5MQ^gInQNPxs zUrw;=;DEy>Mc(skNT#a=y2SPy)H5O)1-(&rCo-sv`X;})h4h=V-_qQdMS?<7CzEpt z^3Byyk=~Q9169|5DlYn1_;GW}--BN^p~|L6fwXp1^cUY(7`q?M^^vpK;q1@t>Fc!U zXN;~n?mp!G-2qbB%yXe-^YsEd=J2AMMFhE23Tz9rHYnuuSlnh5?S~T&i?V+bIJh`L zlhcq7U4KkHYAsH(|{nMYy_)gw8miu~7*`(3|m}s|_gM04M!t}#~0u{Ie}dM!{RZvC2;ipk#wE{zRmHdgXR+LguW!*$cMa}Z7V~9Ytj@yv@Xb-UUVz>b+r+6(l=WiQ zX|Hwb@*~0e*k~(MQ47+G+O~b5c?^1QUF!69M>d97ePfJ7G-XaR;W*+e8)Lx z2yP{3AS*46)j5b*GR?okH}t{)BKexCWy>BtI14u(3a{man`e_6;b-Kr=MnIDt);dy zBdGR)e?^1q$*arYDI>k=B*#HRYp%&3`|5!p6bBOr7k4TcT@3ZRJC1~egqNfGA{=mN z2Pu&b7gt40J_9(2#$knri%Z*AC?g?}g!BK$6J*)$pb*E$v%#$_mkAku3~FllUw#kn zdOpyG=Ew8V87#A@aB8bYT^63WM0a8d3B+!vw-x$r+)(xuQ-%aOP+D=iJ1TJ4$kvv* zNhZx);C<9+yb;Tc$rdJ{ol5bY*=L2moyGQ53Y)xY5cQ zOc|MS(2BLxxFl4d$2xhpe7iTb&i+3x|EXE-l28nL_+>|W&; ze6MWI`n5E1IJ-h!rR&>psY zNEHou;EdwmZu+xFSNj2wqpLinv~C>Q@wO9Let2rx>^-=!p5+MZ4|*f#p}uSJ79>86DrEgd2lEdXO+n3FzHidoSU)U4-^`lSDP*CTp&lvRYNgw-_o#}dzM8Dhii4jPywcK>E#dRMjF z!t9eSWUZGyMz)gC$=yf4{gOCfwb3pnKEA-9y7$IUUO|EE?%EJ8oX$$5T$4;a&!xW_4r0oR@B_hD-sCF;5tx%`Z6Imur)!LsuP=4g$&2uy`5W?YpV zf@sOt=wbUD6dD@*LPBvSHIX$nH9r+TC_>zL;_&_4Ek2hY68DuY=!mOL`)w$rE!thmSM#R;r%*WcGnkcywSdvuw#=wB=F{%f5<*xiDn++$>5j45T&i<$}DJLWMoQs zhecFmB%aCiJL*GaRs{W%Ka?anhdR{xg=HGNr%Hnu`f{qQse2f%Rmi#@F%WY08r+%> zxRbSU5Pw3Brl2&9c9Sg%MJ``C0%szw#LW}UiNmK*pFq3EuGWL_rbN~*QR1I*ajz2h zyryoeWE*tDs*a|Psm^w)jv5tccUxIyi&VihG93mHdtNRlEApKE#AGdFnv?gP6bf2f z4?+Wjg9V7X)esILA@Y^kqgwK?NKQtfQ{F(!c|IF|#m*4*?r{?z*Y!2=FA{F8VypiyUeH>}zY>DG+q|PVVe-Mv3j- zFA!&`kwpokQd16u)1a$>N)@`rBL=zJ_m}~PS!LvFmPeJAzRx>qWmix`rIIL^!U64O zVYjK|Zt*ByD!(8EI=+=wXC%NCh90%DGLPcLN{JJPQ`s&FwLjy8|Jy)xB(XpoP$Le< z?NIa!zYh>YowDUd0s()PHqnEN=j$vcv$RQL2hMHR;YMv`J{TBMAi@Px1{xDj!8&0D z$_z#(CMFY}Q4$hzMd-5Fg7pHqC5x0AXcY$4lW$TcYMGm>mNem^k1_8y)nfW5qgRTt zF9fhfaU>Ge%86fd`K6Hz>5tdqRhu{3@q5Srn}#zYuB6%xtt-c9Mns)S|DqU?@Y!-A}>AmTe;Q%luSqiyFKS?{OZ ziNl54J=bTfq3zN|{jIsq>}s`P_UCppovF~dUd}ypbAEU@PUEt|k^y0LEt#?}%6oI) z@z#t9GjIbVKdJBxxa3N|os z&wa1ea@K@aCCi#AW|o`jsF0R)yj_X?z*#zzoh3Ep+BKhPr&aDAZ2&Rqf4*?=wdIPf*hxA1xI^0FRqZn2sA zMzj5CxKBD(WI_K?Bbz5uZnsf0Ky+(G?ZxSYH3>rG@{|K{;BV{$zJ8g+!~9O4GavOO zUJ&ccmoL%YhBXXJ+`Qa@>a3|ic5~H_fIa-2vsR&}^d%BO^d!h;7PKnO682E!iyd0pK?Xw?xAe}VPRD2DrnmJ%8HbB!|-jM*zv zw-ED8PC*fI)CxEYC{VTteg-^)1ksX7^+aB1$fJ*>k;h*}M z5;1KoECmwx@UTo;G1_P?Xoy>0VFo9hYGj8K_*ggSE@mtjdMiP_p=x6yf8apKf%JpsGu5D)cFLIb#rxRp$mXlt3wz&VQhyk6#x97TuMTM zH+dn!yg@e>Z*zX07zBt!AP@ri9mx3%UITK;uNE4;z)%euz|fiZE?AHgjNd zcyvC0fQS8I%58Pny6Vy$e3ya?~#x}fSVlu6R+ z#P&3R?Od=NHRwTz`~!$tpDwm?Ym#=tIns7Sf7S0;FEP7$O;5tER@mzG4sOl~Dkwv7 ztWc(Il9^xs(7afMDs2E7X9g?E(jaFqPq#8Zuj)9;xlEh6$LVcaF)z%;>vsa33cN9K zzCkq&Dm*CAZt2gLB2I`KE+QO8Z=jiu%PP!^ygiP|rBMrS2is>7k9Pgv9NjGKSB_wn_8TXmGLTKYq`(!|;t zJ;h9Prhfm9XQ|P6i5mnQq*k|6l1O3#>`(i{O83$S2T?_%BA^*1&pML>gN8Badb8CY z!l*|*A_SOO64-5tiHYnA_<(k&9q?4mj^Fb_4W?vt8%Q2CdyS6t?-MOquUlEKl^V$h zIB2*rD#WeeI^SLkma})I$DO<4gI(C+og@xS+w*p{7f=VFNInGG6_leZK0u=$NfZLx z0F{a|RHR_4dwyPfv0Pq)4A|FuN1kv)_&@<6)}X7eKE2G@NP;cc$q$_?$cmyLlPbVX zW_Ht$073#qC+^CH0C5DYK`lFAGgqY=qQfPZ^d%jN3N{>kCE{NEY1)6*YINTE*b*w^ zbH!i*STqTwf%fM=8;~b1M78wwN<#igrK>&>V9wX}$-!DIZcAMwfQqG6u6>r2MAX(u;lO^{+%a~KWC=c;O%4;Z}?bHyuCt+~}( zl6nPGk$kBChEX$21j@50?;7{aY}>pzB3bg16d*MD9{NaA&VUP46k7!&BO}lAt7T@` zW03YRw*72?(w)e!GG`9VV(q~E8XZ{gaAOgO8}KB(?Ajf&5)v&OzaGFE5!vh*Dy*p- z92`?_wFcBA-`_4`Dz+}ZF6a*$K z?;HsC%@4b!Qvo*U6@C6|eK!Kfvwjeaq@bNxTL#djc!1!Crgy?Fo@Mi?JXYMSq($m^ z5aco{HJLID|J?ZPw@$oI0J51DfcA4peg!6;;^C^0n|Y@s`g`r4t^7AYCFxFI`ca7D z6rOiA*^w*QG(=f;)6oTf$E7Tq!K`F2r5$%}!GO5U7z9v{e24i;({+!s%01_p1tV=(-I|x-XF}79 z(>)3Csi}90GXv$=kY-DJF9N%sKHvFN7nUB?RnzmSB9!J}27`bt1Oh?aICy!Z&d!{$ z3Z#>09DGQX89)?ImaR@{yS8{>3|^id6rpd|fAfqveifFQ3rvAA$IXFda3mJYf4fje z4@KRb2Cv8Q2Y$Af;m0E(M3Bj7t&$esM-l}o%qW+t^Td#=gttHLaH3t22s}jyBNmqg zvH7b!KRv-l6IN3@liar@LO%Q|w923ZQ-R zERA+;l{A1ZRQb^7OJh(9DV2Fu>$<@s$Pz$DhIb!Mx;=T*+4SJNh9XOQ*M?(NO*VB- zeCwByQJkk)ZftG+6`@hh{H>FP+|TzVBySeEAF3*&!{d!Vjj@34VUQF}td^o21(dv&`_txe{M1w*W@ZQ~u;5 z7!~(PUX=pGdEe?|*otVnvKB{hUWYiykfB8>Xm6n7bdw)ME<@Xa?7DQP?tQvHuT>4FLJ7IP7EM~g@wAA|l z(o+9tC@(Y8hlIf%fXev4j0(TZ=3wLIW;Bdw!o5|eY{~xULW*`IFNjV-VFt*sdXU(v z_BlZt2_5sL>#SmA4YKl-(g{)P0H6e9)Pb)JEtAdEv21MK#xvTI!V;CPC-zgys5+9; zKLfO9rDn*U*FP+<7Lo(q&KvXqC<7_Vz3}@r!i?mSrE~l0N!i%1d&35$pv2vqF6_tQ zToA3L8!Q+>Yopqfi%8ja_$Co!Zp%GvUUglV-5Gs* zS)_iCu2;uPD9av(i{CL5y zo4jT)wgQo0>Q2@g!C$_H*7n^_z6b;?B^VDSQn4KWC!+4mH_Ym<$z=d62{m}%e+-~i zCN;z>kiVm12&PPDziXqy`+={!7Y4k*cGfcJmfkPm`gx_9`%JeVbL!M;tv=Z=JZPf@ z({%Mmoae!4$8T>l08)12=b3||FhlrOv@hGG)1IE8fgK~*9=9|Yl>Ng!JY+7Fu zQs+01vL>RP4v;3s3YZdi_Y*ax3TWtvWG8UjB|v}ZupK;f>ctm237v&L-ktVZN#~HI z=StpR1ECA=0g(gPK;$6?uR|69(xwYVLznhQ zaP}Pl(+*1iW@!n~9oI$E{#ksfypClr?}!%=J-(!i-||0EHDsf)7> z^$7l8IyA_8Bdih z&$n~gBzR!tW6=eIk*P5&cw)3T6IbTp9yuO#euc9E>8IFcJqwPd;#!s1$BRo67& ziXYw`gmA9;Mm7$pw=B~Y6o$)YK074KT#&)O#nYU9k7tHKDO#~yWAMe%{;_hFy0()H z=y0pvJqHcuAVgmPsFm~0ijPJE0%*;8l@&1?3mWL5IDn#w2|Kk^D@yT3aymZk&>uXoWQE7&AJsU(r~i4diY0t}P^ zoR)hN+4G%#{3F)>KUs4;H_-rMAY&E5SFPXXZrmG=X%$iPW02ejqa@p17pSEeu-CjS zVwRd51}C!lXRvG*egTz`=;ub;-)}6kIAAs>Cs~9vA>_WNzw5<6ZC@`rDDj6%oS$ZU z40sN_u`n7}4`A2-m^QH;F9(U6ZHAV%< zFf&Ac1nmxZndb%<@psm0dq<<8Zzg{ng2K!_$(R3hC35zDgU1`oMvMB^*Z>ZAHs6T( zbXu4Nz(j*;gzn(AfT}1e9K&%44l0hT0&O7(US zH>0@WNnTC>Ss$`w!h!K$vPlR6C@8U|EeOCD`4>qLn}`0U7OM<^kxf}QAlE*%`o%}b zcIC>?#F#_{!~+ug;U~Zt8=;mudp7|Syn7hF6xyufsQ_)H0#Rwk0w{@4g*sNMz>cDk zSx!5JHuT%W_3UgCj##8@5w)wn5{sh3OunfQ7Z)ukQnQ?(zd9k#OvjFXw!U%8cWKh; zEugX?Zbu(yz-@G0{e8ly*)kveRa5|ksNLrGlKIhL!fyI)>!FT5N`}IEl{5v4qwL}a z4$BZbhGj;$YAK>VWW$Rb&}hDEDC|C;BaDfGYOn!NQZZ%hb4nVqC4b?*X?QSRgU6em zf+;P_$rXH-Sh=6UedX129^@~A&rfS^5dG4AuR?$=!E-u2Z{HCWks+&?gd2BXwEuki zGy@JbCVt96YP18uzZBG2Rv_Jc*mp^ZWwf@!@8#=JD*(Sq2pG!OUQ{vdUaqd=5CMjcOh}adq4OS2gONZx1qtXCO#>$V<0kllV*q#}SN%!f&9TKB>>Dw7VkE7T|_k+i)I0m+0kI!rXPFZ$GBpETj1o0-V0-$F8A4NKi z#8ZKWo+%OpR^+JN92Bj%7n&S%3uM3^Pv^JX^fD!OgoW2OZ>c>T+S$g06=PJN|z z4YRioqJHsT7lpEc;p4j|!Q*<^1t05W@IfcOmh`Nvjqib>Bg18L5_kVJTEYKIFA~uo zX8-L=OO9|pt6CHmpk9?SNBIay-79ZFYvOop483Z|l2os@b;oVo-<*0a;9BQ*b+ig= zH$c;{U=2{;K(~W2Isc&qAZ=j$-zH?jAHw;&y4au^v;YemvIWXi3~=W7J`AXxnyaFL z0uz8@e;Zm2s+(5iGiPCj`NosvM&2Yt^M0d1EEG*w$mlo`KhoRvGJxDIeC^*1e%Zow z0G+hGP`V1%Wdcx(^WDV)_5YC3(*dvqfG!KBc=A=)CstZfp7_wQ38`8NBG%sw$SmD%2Ec(6PpPd{;s6T2mR%pBH-5_Of-|`DiQmj7iYe(> z1gblbj+KD4%~qA_d&h*KXInM{hTx1tWckh3$10X>#&$PjZy*$ zKW-prb;cny<6|Z|zqu{HpJ7OiSK@y?!b3Nn)^+=8{X_0j1XHzMvO9wWSR@4r(Ix;> z;mdU4U5z*Q0DGX^%f3tsXH?`*Cc1H~|3$@dRggKX+yO660?<1ubk&Cvio<}C2IdDy zdvBo(*z{!O`L%%)dc>c?FyKOt|L)tVK)W}ev)xgX5acuTe`SP6Mz$IQQ-o(rdH;rP zoK>Am0S9#S7fzaIkP8>L5(l^l7%s%8NjQuhF~C{>R7yx)*;<+AE==*DSH*xDZ34(Pfs0)=0Wg<|3n z*aEy3aa9(zHw(kqtyq7zeEzrzz)m1+ljV`xgIFIz^l#4%E8sz|;{?;(KS?Ax9i!GG z0@MCg*aB}!a*9GB9O=>oF9w=}*`YX^FB+?$YqQA7)D*VF2*lfc@wy&JtwlE0uF*|N zFqzh)gBY^$C+nS0h@bi`42RaT8GQMtCz_wVgS-WyKN(n^3U0kMt` zDh<6*L=@Wk>d2LFpBm1XDX$A|eu z_k!n1rJhWZeyfsKw!AhDi&VuXCSpS82|QZ`-NeX7OWjC+X8;i+@g^B)==_)2@+^O{ zT;g`sM*#sZO03=ddmZ5-7w|sfrd%{ij<@ z1o~qM22jTg7H*fz*%Z%hG`{0uqeakqQlr_w@4s9ZZ1A6*SVD(2EWX;)j{?ENr2)%z zd)B`jmQ4t>VoLkJGK%6*qQVnL6Q4E!t>bwA`xfDF!^btZW@yURF7w0gUemhAnqUYD z`M0J5jF4pK{%4jO&&QKC3m$^yZ#wTE*shd-c11Ef1^~g(fc~J4=H?BT$xIUPGD08h za|Pf>+_vAyKJl%1qzZswEe^st{*OF*9%PPl8VF0SrSly(pvIWWfs+mtPT{#}-=eQC zPDjX~ZVKPDx&@r6im?%^-3S}=v3{WO?qn*kCqs@$@zIY<{wp_($vo+I>SHlf;|uzw0phow)VZQ z6Y0PbxUEKUI$EQ^R5;3Uy|8rMCjQs>duXH@g~w{&egh%}-jBGC5jUAO0Xj-LzzQzO z0v%LkhS$J?!VO)^YV7bux{vNSo;50{#E(U@VxHw-3!-d^@#6H@7-YJ;j?ya)C(z5m zFT&=?$av|98r~)40CTci3SUt9SbQ>Aq0WfafbQ7+@UCveWcD_bB zc?h?R70*%AIL?%jS);c*?X;Vhk1V}HViON&@P{r^F@-mr*3-_1+xXmtz&-P8co;T^|3N6wZz(($XX__jJ+&%e_Z}5Oy`J0nCA(e603~8^|wn_II;ixr7s9i(WC(kDru=raHVob=P#UEWe7GVOgei1%E%m~>4%d$R1%=6b zFHUYwDJD96-0z&B{BkQz_suXCYcHwmj;c`~C@I;f(yJTz#ZUE~y`zShwtgNjH{(dx z9i6bxR>wYPFxKim91x{MlXL-<;X03W<1)C7P3yj+9dbY3v5Tm}VmYS#e3DxdE0S%7NNy$U&=r^03cW z+teCh`X)vAeuxbODsj+5c=>+y=P!BS^Yh1%9Wt%xXPP6G=cUSDvR}^gtCr>pi>b=^ z^^132Zch`N_s8%TQ(jicZnnhBb#HrLY6$||2CdiwF0?FtpYs-OBBP;ow ziO&TqjEivFcl6;G%uOGd>Av=JKtKy=HIL5qJ?rxsx2G^4ZV}$;$B&m5Ye&mXgdGBt z|E6uoOt`@_k>zvdLItf5L}fDO`piPLdcLZx&UhYdceVJ&X@Cg}RUvUS6sM$mB$ zhB=MJq%kw|^yGh`H*G5YBkT6EWM_=j@b>6QyF+TdpMGN9Vw{KyBdFTn23wElb^~D& z4&?ukKj-*%ou%su{;TAA`-M3Z{)ic`c6%!-7W|#wWAL8~b2hQL^I@6wDe0*T!UU9; zE6SV|K@aZqBjFbEYgpy)`zO+~!~@){rxeZl3Ii1TR;ZVrkzlOL>gC%q!=)(aH_bbQ zKeUsCM|VzM>L8Mjz9VKvw(^}0HQ9nyw|`VTSG-VaEPAWY_FfK@XZ3cxoR{PkN3KdW zgQ8cN**W0!Toi@= zz`)Oy*Rf|#ciL#1WIoc$az2 zyV26h0xGJFR=!}3W~?LD*zmTv>N;MZiv~p1={6YDh^Gixez8G}c$d{_!vTs9tGkgX zYE5m|dw$+jyKYMUq5fg4Y%Zejkn4lV+@bhp+Lr12quCGJKAopw3u$kqBS{^SU<>VD z_RB>HdhQqF=P)1gyN26^naSvbgBAr!3p;)W5S3JXs<6`E8#IwftsFu8jsd6btwMzl zDhKtSfz*#D1_@)wScq{xFGcA`D}fRO;z0Bv9qL(qgF;kI{R4?R?rKZOIGg)l%;c** zJD3jwTF=L7PY{y7RwyWQy*hljpDo#bp%6YoGF}o(dN~ebDE@*q==!|0-P*Mw_U5M1 zW!q)~d>v`77D>T%6Zxv5B-9M;tR^0<#@|(Hzw13@Yo}U(vTLn(ZP2M4JF13=;`Lzc zTiw4`a&k?68;9jP*G!#mQu0bo4i{ndy3V<33W;KAUtuR|6>TeP^-+xpi6YUbZ-R;{-}vpR4(wovL}< z=csC~G$th{--P27kec_oLB=~vv_w^)K23kIzaCgotvV= zzvwmg*#bOF$qR_L_tBp>6^O4gFxQ_jN(4AFy z4RshTYZ?i77*0qm=5z8p{=%v0cCNZKRBNvEYiF{d?|$lyO6gcv zVsyZsQ9b!ymt8fx0@dn+!!}>eQ}N1w9&xmKo~T2rzu-FdVj`7(?t(P|<`qHPv?FK0 z`5-*!6EEy5>bT-o$H7e^wj9uflgW*CR^>6g=YJWk-*e^9T3llBkn}i3G5k{U98mZE zrb(*c25G{%j#+2L_vH!tAW=Y-%a**Y)ceh;Ypb@yMQCd`bLYN&Gj?()x{!zhuf&%3 z9mW^;mDy*UmkDOMNbJYo7gL=jZ#`~%H0_q~8;@<;ZgsghX)+Y<-q7bV8$N_kL^k^E z-Mm2adWl-=h$QKAy_IcOTeY&Rj&bR;cxW19MqygQ;++GY+3nw+51}aeH@>%D*dnq0 zEw4?q?Fs-k7yoibb@rW2h%xjV5&l1uHJ$L_bZF|E<~o)=dR@)sSE5-$a)|o)p(TA> z5c1argA|ra;fK`6SuSbd4-Q52O4+8~kmVP4%%XM6161|O7JeDaU&!*Q^`DH$ArN^$E2U7krO$3u zs?ZPk)PDcCP}pdFmABtF?{#7cUQ}{Eli25}qjlEMjcKami+ikUJCN7Trv2%eFXt`9 z|BU!%bk|8ILC~yRZ*VG>^BG@pLD)jD^9j1}u8cJfFlX1ZP5PMso`-p2M(dE;(et>& zNa160_0Y{vU999W`@2KY8uLq|&vqkoTBF)8TU^eWxzDOCdbG8koBNI9Agab3H^&L@ zn3r(9CN@8r&b9jv1eK^RP5hpxdlwT^s3=G6^^aSR+*N`~8{h0XZ@u&96_xh}P*VC% z3Hwn<)=lug_#ApA-4(NHvvi(cT-R8Levo(=_fz2!(PV45i!~ z%*bs*qAQUo5{Y3emzi-NUYC+n((FRgn5Jf2+U2g~McS}P()oW@a zky9+s`0XL_Rp*lJU6<+dE>#MYqfCu8o$W5fb~F-^!B|1tW5CRKL~Pw@e{{59aIRkI zIP537{WpJSq`m*7q^LAOt4La&U^!j8FW?D-yU(YiW7@K~Ihru}+6}L1vdB&t3Pk)>&sbxG}@mrUfXpN~&%+>(|c+ENq$@sxT0ytyZRawz-fR$lt^dto0L zx63X?JRkppSJ8c*xkyj4b65lkCvYef&NZ!8OHpr5?Iqz3)Q~ zF3C1Sx6EvYB3$oG-^P5zT|Z}g{9B$oOE+duzcKbmOi_!->ib+nE8cUdqrkeJCqUx5 zy7W3SXBK8Uc#HyPXO>hoPs)p%`asm@r}?$rNhw5cK>zexbM9wr)p5%S^Ytck}6Pr51&X15rqt!$#jhH5;UI-A?JPizzj z%y?5B-gh~d$6P(l zt339m+WP?Cd)xFI-}GPVICi#1sZSL75B~}~BCOqCbr#P)epw#+1HGq2Uc#5 z!wBURhyMsWEQu%QXwX&rrq^#G^ZYR;yeZYDID@L3JSxf_?h zpMx9ySgQ5fi=77<4@Slt0{fY7clb!Z-7})mNbgLa@QveKao8c>0Y`BkC_2&YXuuH1BZeKzGO^TesCnen|X-*a~|Rynq>R%c8DM^y+rLtDv*N^s0h1q{ySQ!Vy5if!_H$g7v$uK6&)^;4uZWpK^)25>@hLGh$)`l`|W&IBIp8jD1h)wXUxcMlfky^#zP3cYQ zk^tK1~RuIS(GSovE1R>EF@m;+Doq=@TV)p zy_WSM)Y4tFUPh^vkRY!+Oz}KHeN$s~_x$)csdq9enwI^HtfB{xgP0KUCUZJG%43X2 zT-ai^jzH44q9@UZKiPW~}B$zy`<2 zMAMQmGIw7&DZVENWOG`u-O%PcL})amKd~2Yu&^m|xAvxlmNbJX$6|&&t2%#*+6hfh zkQptw3A>5-kz3kAD{g!)ji&f}Y_sfKuTg0v_J}j;)~={CfkzZ?i~BY>Hw+XL4MtSi<4vmxRsP=`1emMhuXxt`tIOz`sQOF|5fiI1&{)Nn ze6O~y*8CWnN-cMAdqnIBRtX3bFmjs7ZFLRg2MSs8gVYuz(1U$^w5Ay8H)rR0n2d#a zeaO#1MfQ8a6uqa_vi(#7L^+-*>kGt=_*O>>g(F7YcK^}W*Q*(13LO~aB^}oN;i%35 z8#9XqfWW<)W^&+3myX#xL1REfY@NJgLd7io(Un@$EG$@60AgeT$OtkY8+EZ|vKa^z zuvmbpklP%2#x*#3E*Gm6FF3!EwGIW!8~CeAXWZ7qyumF*| zEn49ALzwO)zibKIbD@n#otI*h18NKI6fdTvlUsM2R0@4eE?yb&0YA5#!6#&`_|S6V zL$7f<$Q?kWglg%tRJmn6k#1T$6?$&?iWhL01B>e!dH`pb$(;bE3y=vY`mr;Q84O-> zvJ{{$Bukzwu&Qy-eER zPvV0LgSk?NH0-~{>L`DYlL$N=PmP<{zSZVW7Vg^pzqV&CBskq&@LK!ws6IR#MI+ME zJ<`uR(%;ZG+#e1oEM~7h8l#KeYv6*>HPq2F+^dJi;0!UCK;Nu|e;Ys!^&2LnJ5v2=mY=iXVv*XdW7k!s%jm>qqay>Fn0sh1rGKli!Dh^Vj|J kd{_VQ2vR5oh^}L?f^SVZqv7t5tEe94EAYK4TMHhiPRpmJ4c6%i8e1Uhpp|F2j@MC zp*kw$mISHw8@n+*>M^y~@Zq_~^jOBm^czSHM;B|<6Jr_3lL+65v|h+mLBU0VSxjRv ztueJ$V{}&{VLLyc7jpa457jR_2Em~$tmHm6e@gb3YveTX*l5#tap?eGyy!4F;L@N#6!aK3w}(wTd9qJJEoZl&MQ zro$(F9JIY0^Tp$pvtGen(yoR|^|J3H+w9ed?HZisJ%@AkNEFi(+Rs}_Zw8l98Eg^T zjh(hfE2OnprM7!82?B?rOa^T(4<&u7lQgOiJ}lpg@ZGMFY%_P}h!l%ApuT?4=(w<* zoT0dcka;58nO5d@#jG&I1)Eru#!z8kHkT{K%i50=H7C9^229eQXB_<0eN7s&{~ z*JU2D3jRUFqa%v9tXetBEgM@6ZN1Ftyw%%zohD2CUFdsQdE*6{Hxb*?JU1K+E#C31 zy;~byJD?GOj6a(oxMdFYz)7`*dIoa8ma@wmf*6OGC>XB3+Bd+El)0x8vWU<|puwFHD++#ouRpg0jm$&1jTPYN$U&Jy5> z=%M$mopDvXr#ruPx__3wj&RgoLdGY!erEV~q|Nx)gD8$1)){jm*Q$0)pPc8h z^th?Kh#9n}d@c85+>czD8OQI({SGW*FXRQVH{Qk^yc|1ZyJU`Jjo`O;Ok>Z8xnuLi zoZlPrFmms^%%cPIO)8&o_u{176>F?pmhtz(rYAf|m)5oX8Lh*z+sZ!0FI#VmglP2& z4l*gxh&J=7E7%^)O*1we)Rfbbqi)DFY`%d`>_l2Jjee|7#l1mxqs4h$1-SWSnMR_F z2myRdx~UiSzX@n24i+exVjK@R^31Z#9xF-yZlY`PO6Qe{FizM)xA;`SsnW``vjzR9 z3$N(yT$FhtKa|X=IhAxJ>1={e()_Elotd8lqZQhcCJdtTFB-ENzc)K%+-W5p6z8cSA(WNeC zeCx67*~jXOHN#@hRJ>|y!~AahRc#($eL2VWF54+VT2wkzriLR%+S}n?A6~s<=KgoA zLS%hHeR5#jScxvNl6cYavO~(%r2*%63-9`SufNypJ=Xi6S0VRZQzJVoJu7caMW}n4 zdosyl1Uqseqv++St>CrOi;N2Sp4*;y1%3r;FV`9W_m}1l&R?D@?SK3(ASZHuXvTn; zkz1|#^`3Q)P?w2yzFfYZ78keIS)O;*eer$Sg$)`FnhiaTy+|VVVtLcYShaXmOK*>4 z%e7mdbX#0`YZK!W8y#--3Fot&6uYk~74tmHH)Y7_ksIe_QJFE<4_809**P+~nO?l! z`#D6ojW{aZ<$237*|F?y;t=K2q08?s`*gjUTAK1-@tW$I;t>@UV-jOMN3>L`CRSXu zV6{ARZmc@bV#D%_cj>i!=U*(E#!ZmHu^K~8)0djY#;(1io76dh0%O232M zG|u$Rq3U6Jp_XAXp_5@J4yw{)XmlA%4$U%n(PSPTX4itzF`hf@9doI1qWppR%gNF9 zo+ClB_c*wZ)?>Ea!wtga@&cg>w%g(Fn3uHEwP|^s^06!;SG|Z~4zFiE%%}=532GTV zN*g>p371m5FQ3A#sO;^wuvHUWbJ+dI#N&xVk^#*F8Z)K`Oe%Z_`BV74PjV=lWMo$peN<_a=@P;{WI>Nin0lZ+-9n2{EVPsM!Z^hu+2)z)hLV z*6!1n=MUals6TaP?mp4l5P5}bntvev#m$kfm8mg?=)e)%Bj==4E^1v`>gIf-_eLS@@$Jl{fb+$uH!N9Qm^Z?mj%-8G(VAS(vz|M7 z+-UnWVa0?ye%ds(h=y@U%v5AnqF;MMW#jM$&lSEma;7=fNKuEoj#3`8Z>~h3A{x4y zuee{9>5|QQX|t9uzw1%eqd9fI+a>*qf?Ho*_eSo@578Xa&hyJx5(7*u`d&mt(ap0R z5lwHHc_oqCaEk3AOH7QL*qUXX7fn5`yuhpEZ9jv4L&`OM@jJt*DXE7^t z*S@^ppZCO{+Fg4$t~8TIF(b%h@A1{tvC@943*K}-J_}Jp9#wxlesHyHckJ*xhi8?w zH^y%VxNL>34RDn!mhT`ZZDvu|bo^%8z6v*tD{%(a%o4T|4`_GAyDDg@o-qw0zIinJ z_{)Gdk^kK@9-g4OlCg)ogFN=TgQGSdjNR4eTceY2C&w5b$~va>Zb0bjf_!G+rtaJC z_itPg$b25?Ju&p&v%gE2;rMk57dADT4V`BQsj?_V&1M>~$K_z<}yQ{l4Yaeq6VwYdX zL)o1jEIA-<-UqPyp;sbhAwtQAeWacvqqhp~H3r$W?HGucLKfBzAOlq(h~?NPRI(nO zr`A)rnkKg)bDyRAZ{m`sJ{K=*Mx0=6?r3VNJ_t#eqgY3Cb1sZ4$*o@!q+dG(x%E6+ zG56#}pI(>rfiu0Z~=1RDA0 zTRfeV(dGumNDa5^myj~z(&Cb66YmzR({E^++07H%a3Qw_+Nt4QQ zI{jw8rWza$A4;`jrlRt>)PCvyi=Mk#7nI&gy?yKQuqHP&Iv9;?JCs-TV)0~r&!X>J zeaqADlPvBW*=UvLxnO$t$Pv;9LT8Oz;Ony8waNY(9W0b7A()a6NkQ||mz61vWh}v# zid9Q#`(GlzONf#sSP8!Gnm;eczNQg^Jo678RsQ*Qf8RL=@)`5%gZ}r5r6HE0_{)KX zCEfp84iz%~m(~4!yWZIUS+jsMuV4wjihrTG5|j5}j58+I1mzVPn;2->TT`r^hMu;x-d`kvVC&CmXz|&dL zWf!LH|0Agcg48iHeRBJ?Gz?dxR=4pfk|({ae)GY`x7QLeJ3V{bJq1qYRn8VquYdXD za$EG_u1ue?A^*LdA6-7<0cRftZVxo->B=>qa`mQCaDuCtRJE>bI8A z>;a@UHZGD7x_{t8P{8YBXUeQcKiv5GvJGG7(3cl?gvOKZ*dFh%Fa(@#{RqJGY!r#{ zEh>0fo`G6UX`cqM=ft@GTX=~E z@el%g!u5ktz2P|8lrUghcV%(-Syz?INHn6aWy_wRk{z_`yOP?lUC2IfaPwiY z#%1V9VL=!_f2;$N?LUr70Ax04`J>q-FrrKlVOQ5>p(#4tO3jEq4VdPmmF5V#uJv_1sUS`hf@Tt$wV8ka*ogqWp7<+Vk)6|q zObn8rlI})Bpp3KD1l4sTm5qYueydkKzWi|q{T?}o?EmPbcC$14>6ISYE8X{m+i&=C z=a);B#q%dc_MkVXU;Y?#8uOjboEr5R7qc^=%;8e>_Uvzhhr#ch|Bob^2FOA0XJ`NW zXza1GwVt1%=AL`bnqvAZ+uPeY!jJfibuND(O?Jc(!UWr5ba>vY5gBJ~N)|OD%W#yW z`m@%)h4TLE*GL67P!JPE*vQi>Mjh~^`!9%zdMxpV6tnBIr5yvmQ=^+v;Hi?g-a}nQ z5*3M4YK={B2h$O1lZbBXinbUwpQ_4p5cH*37KvCw_RAYTWPgkY1$m4I{J68W^t=Qj;JGULU0VZ1#$1PPrP$2Lds|sA!xjSZ5+OMd7@72(xd{8~o@%stOA^#*j2fsls^h(UMbTRHll5ZX~m zSMnlC&k-ag8n&-@z+H8YXegoQZZ2J@xLfH_rW)T z1ha%39avJme}{<5JAlof)2`~+(hSHX_UKYv>VHz4|JDk7WKd;ZOXJswXj;NF7*{ate|pWR!L{jcU&zvC!TO2Uw?n>AneKPJSM7x=3) zZxwbVCdaJE)WrX8UKjhU#cHz?B3609k>;~#}5rxCor>D*~HHac?0}nqPk5$s$8IOsJ$JKxj!`Nr9jTG<$mcq zwnm(8DmvJM^9q6on&R(vW~gp|V(bdq83;OTG&hhRuVNXysj5reGVb2N^>jEN$N6p~ zwE?wzPbHw1-Kvd1i@ro}21!5cI?lI0F=>D*$pJ>~(tF}Afsypf zoFabNMuTGZ#n<{5_`GPpm$C8jCyER+k9n^x457D;g*Yf2ek_e@K~_0x=J(f9At8Lg zVhJtBKOEJJKAq<*exd#Yr;2}#?`&bJ9?HS8RCRBArOTs7`D|Wr{}7$g(G6<}TI@@K zxy}8cY{L$68pC8U!@4Z9tp>Gte}hhw-{h?mdLnqGh}591k#W}i-9wHSaJwIMMdnxY zAVjCU(%7NhxnW4VR#D6IrD4UW!RN$g{`@6_F?p)1)ghdAYo z8x~VKSdylcE55u=J#kdlS*YrASNb_?#dkq0!T0kfe?wx)V*viOE)`7IWJh)8)<``j zA_C4lGB=5ZcCV@r4{UvHxUc{mY_8IJI)DiA185wLV2~aP?QFdPSi=kN#bZ^DZ|}bv zTYK6d0@DRb>zOkGiC2>yuayhCl(G) zj6nb0*_*jr z&S(_U;Ebo=gV9rv^ys*r{Y0p-7;X&VxRaRG7MKLpI=TkT&x(q4$TBx=B}Q?Z_1XAau$o7026^IKDYI;4&3x6(}2~BwWX7U2Txid^SsfVC~m|nGd2&yO*U>4brWnd zz)>oG+U=x%iM5vZ*giWtpekuggUixo77g9w|;vN9F^t(YX3pDfV$CPQ2Sa|6%U0F*9!vegSRc~s*aX0`G(u2nhHk{ zIQIEs-e3?m?CRU7!bfZ|P-6+9+C%$m8g_5s?JX*yD>XX>H`)l4=x>50HU$Cvt4_yA zH-97;-dqp6cDqtSX`;tjA;2I4JoK1?%5MT)zB+i}lvXYEmM#mNPUR-DbrN=By<~0` zhkY>KF+cyqV0HKI(oJaZ$*2v#{Y!A*1?faKZjP{MmYLpekB*s;u&x{Q|K6wFHT$+G zV`aL>ht|R@#$mcUvx_%g$<(hJcFvoV}1d|g9i(~~#dK(ryE;E`)cWX1xvj=ir-9PTfFN|Gh@Io8 zO|T^%&&vQoJRkko@|j-$?|056-Yiv` zJ3`eeOmz^{=oy9jsVSuw@vXT^K3`w{ptanO$$XMTumTqi;^Qy6i|TN_hZ~!F zxZMQtn|-%pS;_z@X65}Oa)zfi@zdEdYb#20R>RiP3o5bw4#vcO6?le*vcIbm<=p#k zm8nJ<8%rH^anL@D>C2_s`BKxbk;Ob$_gdFgy@K&s9sh5uc{y!Falb`~1{rN_R+efn zTL3T-<7(_r$!b_pvTFK|qO|2CaH%xy9+%YL3uG+xEfScGL8l(~n*~4t{XY_^Ty6DV zubvD8c}p6(!9?>$Fr?NH5nrNvYQ*0)CYyi12^PUah;iP6MygVPW7297J-wuPzvhy5;~b0FXK!~LzHD_F><>9U|5r>I@j-1KC;s%v+) z>V<`+7hnI%)Gap|3T2c{)GCY5Lee9fNWA;_WMwb_WV(ZWR_uY$f&fj(=0uD{y400! z=^~?S@xa;sLWX8WEk5{OyqLw40*gAldj+V(*rc~leK@c`T4xS}ZCNLJDl?@W2yvx= zk)~T1RYlS8J~f!~;FC+QIr>LA0p` znlC)3tcoZv>p&=k&~RZJMDQ}3gtc?&`RD!TU-~^GYiSu`=c=TEMj#%zwYJzbIB5RV zz9SAznRrMm0O_9>e3B;Plg;||eVY~I0}jxv=9PLg9;*=iDQZpadsHFspkTItKk|^t zUgRlBrP&L*6fMaItL)mMCuX@>p9X9%b1}PC{JPr^YbImJ9KXQ;e~1MW@>$H8T%Y1o7Ni^Iy<$1N$KUz;QO1GfShB>OCTUeI(n`L z`)BOCIf|@FeLgTVRUHXYHx%V+TCR-Pn|oLmje_zYpXo2u?*ddX9qc_tK@G?!R=Nx$ zm5yd%eF}jp@Lq?Mtv}GT@Lc@>gy#rSU*4GjHxnbqzhe5@W(kV8PI1gdR^i@)~X7Ej!+-bE60jsNoL#s(79ml+Wb8 z(0UUM;N(&Hh5!xyL5gS`FEo++W+!vg^xV6Lh2roMJCgO@ zGOpvTLKf>c7_KV*LLlccFw*QTLY4Z%S* zVR~EZl?g^2l5+!hTp6gnj1Z zoV9LFLAc4hV2Leo3GMGqrV`-n5D{~0q=FQ$>i8nBy2VN5#(jrI&jyz9i9@3L3{)9q z4Blld4R)hm{~|!iWzyb}$B52%j>_bykfW^^i!nOd?@2}JZ<%I_oS?d4Rw0xC9fmu! z!B+6(s{!a8go}M5c`2|1CKX~Ua^{iHtX*APC1#|9aUAoBRsWZU%pxEF#3ho~S3XHp zNXc+V(Vu1z=DU?>sAb7cXVv6zI2+op6o)z_W?5PMR(*_+LlfDgs`txHn9@!F-&wgi z%$U^p5kBL}V}vud>*YVHq9XR0S&h=1`^7%w zNx*&uLd%`JM9aiYyqmT z=1zA1T58gf9X5ozF+LT04ptDLqz*@59eGq$RIK}3`&F`tn7lxy4GrKW8b;}hMT*0_Tf~o8}PLqfrU+eNi?McPdx0lLk<8W;JDoGGgTw5^Due+6j5P-pgcxg zIZ>7!W*Pw1d9Nd%XZ#Yw*bCr*R;eC}+-aO4yNNa>nu8$wtsufg1V)OqRMVx-w>BQ` zFcTrLNDa3$9}e#$A5#f}T!-G|P{%9YyfucI;{G+9o|2Q{4exq@Uugm%uZBCa-8o5=ezh1)+&dS}Z24Z8-_Ozuti-lX!S$Z{~!0XBYk3a+f z6P?1K!{@4&!NMZ^`y0uUQGZjv-PjsQnN#TOuk^7p1t73fd&KwRIvaj};CG zj;;Bq3U^}hi-2uVW$4Z0P1}mua4tw3x=z^kv4a1HBfrFT4mG%)AEfUz@a?C9y`w5w zRip}_GRT7S3CO{18^deCod8cb8tZBY^aF7_IJX+1ZH@Glhj(1~7g9{GO2$ndjUQO3 zhJnokkVB~Q>_JrXP837{U*z2cZsirKL)~I8a<;TGVB467%eC&8y@7}ZiYI#3j}IAQ zfk-O`XkuHLP$c9DmUT)JXwhZ4hVU|b-7P$@*rW@O>O)(S!LM%fnq2!kV@J7YDxWOh=ZbWmQQO3c@=8HD6}}~0~DEs!HWD@SDx%mEXif!OZ~p!P9wKMFJ9b! zLCIJCk@Egqm0dv1GuKD}-8k4ElRf2is)WR>jW2_veL*6BA^lhMFV@icyN|S#N@uV!bySK$kGDsUpvxKlpqb#8k?9U5fN$`+-0>dn3ko0JybO&cz`|BP=7^s zF$HdpZ4`9t9FAV7jyHOVw0_|}*?#%$eS@|nffVo*GRmkO@v7GN1@UI6!QliNdu1{u zV8>}rzEjx+#c6eAd=cDJlqwGqL8g(0^Z44R!T^69$~#mi568S3P_zyC#apn@08n-i z1YDG?xWHbjZvP~tjhSWu3Xuy--xsm0T|7k-MAvBU(2*j zeUP0S{H=g#dj+A6JlP6rm0^Nn^t@uRwOD|7Wgsi4XGK(zZS86NW{niUA$L~f7wd|` zz;>vbmLfrN*x^f9BM|^U0SM>sv$b9;&sU!IxCI!?SR|40_AmA36uDUt z2MGWhk;Ic(Stc?sgaB>Uq0yhmE3Jm$RkQ{wh?}`cX*>AhgcGk!EZAvWb-?z=TQAT3 z+Il!i#$P+}-P2Geb~`ezM8Tp9wE)5~AI1Ud5RuOd@4F=WcygBgOi{Fnlu%|fAh4~} z(HY`D*(C%4PDffjY_SxW#UQL)qsbu}%S_=3RCh8tH#co1=O_QwcV{!fnZ*>zc~YWw zgUG_GWoMD0F%Bg+qB53m@QdTH^8p_Wljd**ZMf!Qo3|^1_SBV&+wLYl`B4j)WUOn=8bY59~8Ge@Yghv$7y|6iLY)&TZNX z4+6}tgc+C4K5Oe`K<;aw9999>PiuUk{e_2wSO8fyol1Hf#B619Be^_L-vt^9pqaw< zI9MN=O!ll=9C&)h->L@!#T*2FL1LjSeAae%?}h#>Gpm>Nrr;nw=P zCydu1)&XT51}cSCE?TuB7ysc-K+GF}D}7|7IAQ#=nIDrg_;9XQ2e`{!CTwS}mN6OP zbhq+=k<aekd5mg$MPdJwSC%YK6=wL1rXO;bp|0p zT^I_=3v{ppIV2r<$hs7o2|?0$xDFT(sPRNbTXm>fSS75_qi>>N6w<_UpNu6IG4V98jT)tOUk4UeA{(wfI#rK*5avE;k>gFrtZJ?I+!b4suT5mp=#HYX0) z%8np5Ge;Dr1XBh%!T38U$P~ps16&5hl&TYts;ungt#vH%S3?Al?RvJZ^gJ*G71n3t}AGCi)zVBHeV}HF{s|5xy zT48|d>8{QhQa$|BiL->zl$o{UBwlm}Y+rl9=DN?%-Ucv^78?Mazsipc=|d${kw^qZ zfNYlRCTi?5AxvE1Mc1T*G>{Xkpo#<7FvF?3RPwV8rpTI=VKq+2kA}=>uSw2Q>q_dX5fU z@uV7%2H%tL;oNf&lLLVMI&$$C&EtWh5a544?B})ouM=0f1ya8#jLcRYtqa22$RfC@ z1R-4wP#91epT*447!XwgfuIFS(1gD48b8}ZvJOy=n?4I4fTCgpqXDdd+g;EOKwVLU z5-==d{uy`~xwV10dSDTx=s_@fqaST<+Pr)}gyECgCxM?&*oK?T;}2X^RR_`3h@%WF z0q4+ZhE4yosIHB$eR$?^?U(0&;x~F?*Cm&R4}HkU?_zXeGFtb9@}oe5-4y6N$(Nz$ ztuCWJ?WLFTzFgcINQ3E@o0y;%VERe%3|)zmHJBSAPo0dL2F3Jv{@V&~AuHSFT7N8p zv|yXqw7nJcQ4oqmp7 zGCq6$&9Ynd-x(xHQL;%d6?`xR5$Z+B)D;}hemhEnc@o5ULD)M`nMC6QKxzbmbEfyA z3B%*C*+UFe6u1nxhMlV1_QM@==;@;DKue{)9*^~5JP*{cHGA{9W!)_6#c#fKSssNd zJaZj&jP8Po-UJm3c`&So3>nW z#0;ui5v|?(*&Xe(OgO>Wg9svQm(b--8>A{FhbE zAIwMQb4ILVVaRZc*y*{MT1z4Hm!wtFW|aSqMb;H%<%n)f>1(2 zWG|1N$&Pt6{;%ak`GDmj3cg+;yLVbRR-pCd&AT<1zkT46QS&smQY5_c5}16H-kAd! z^<)Wi__78LUZ3S)^h9^p0?N$vk-du)rG5Jm?+qQ2dL`dz3_-@a#r+TWOu|vq2WZ@N zQFKbd^6@;{9@2?eds(R9SMBXj8+G@`IyaTn>7(J{#_x9I5Sr_jgbA-z515uSZ+jIS z!Qw6xPa=LKm#`i9$N$dpnCk#UqthMlNg-ni12oE*Ab1_k?An?QNZOT$Gc;5=j`ZOYcM~9XH0l>yJ8{K`LakQIt7ktUZeCxwuH-gY zo*N~p`@k+vJI@7fjDCbX_(h&Fmqnpt<;Scj>~xgM+?uml6;C`Ndby>ZH0+lxxA%>6 znqdSRkmHx}?ZlcyyUO!_xO_+iU<#E9-*=qJX#N!0_*u$P7EHBll36(5&w|aN4&lV( zprSwv;p1MN(7H&Kyoi%at2!P-arM(VrY|j|o93Y~^;7g6sP^#{&QRSrBkK?m6T*yH#INe0pnvXRBQn4I$OA z);#&@YtUamxLe@QuP(eAx?C6MGVNs^&|T59oF)|QsIWEV^p70X=n25hATp_pj5r{1 zrlko$L!z8Bp(*c41*346CO(;VapZzz-$gVBMkuS8iu;Ob=Bwg9wFA=Ybd|!KZY1bZ}LYJ5>PNk3LN%QQ|#1vii1 ze+qXG&q?+u3VbvEWCx$6&w%9S-9$RuDqU9Pr6i59XD6=ktBkbQqtEjziy}1cc`VRv z(VkfA`PZ4hgOp{gT3id@2$T$bjeNV(-da7mwJLuqTi=(vCuyJ}2-BKPP`l?=;tpO@(q=($C&qdp z7*i{gr#bcg1h{R55^+B_R&sa%gl?+6^CHVl=13?^Ey@(TpgWHT0#H`evE;tW1Z@~# zW6wf_I5J-eshH5L`!y>kNBw|)TDHG}g;az&u!SG;cOkIDR55wMa3#qsjp9%EM&7m0 zgVEDz6PUhBx-sf7j`l*8jzJ%DzT>ogGNFvqJb6x(=x+W&8zTM0`51nF=0un{!A}0` zM~_PkZG>9H;eu03Rj>fhwtWGPD&SIAhX0M9h;=PI<$G1G`xU9Sd^dW z>Y3$&IlkiuzV>rNVX6c+Ok85^ngJz`Qpb@{1lR;{sX9lffV@#7!n{?`M ze_dJv(_&c9#oIzu0*{|vX^wWl`Oz_V;SM66CyE>3=TeXyno1MT#21l`q)=lW99>qW zT5Klpo~+~p)jgQHjymIJFh)gX?0n6&TzvmaIoa-e@@6=cX*EHo z$*fwS19x?hubVHC z!f3-v_73$p>iko?Dr@DX^TLUKcRk{YAHYiS;auG^-HNd*qO&&Js{SP1T>Xp2{VnKn zhN;BqGk@!Wk}wbm2W?qC93%(t3O4waP;y|u89em(^LQga+F4r!Y+3Z=am-wmB&H2b zTbL=+htfU$s=$-m`*nO81w~v)0TmjG)=T&nL6YnFWWjFKO4F*0i!wKDScb81veg~z z8GNx#R1+=nAfEZ-Ictx6hkbYr%;R1*K^xQk@Kv{bT1$p|(hKc>CPx}@U<^JLwm}w) zLQ>#1>aQ+JC(=CYYzbFw3%Z-Yl$#(FfeT&s5h(E#*ZWTGD1ymThtt_<@zIZkgw+Jk zG^S1Hs_HrU@;h=P^Ovru|pu~Jipa^jgP@84wl4!FAj=G(#g2I*}h&0g#+U3KlI0}SGQeGDe z`aU?w(Ed3rVRXV2jwIGL0T5{DjtOrV_$Gr1_;EPgvT!J!Xz&rX~3o6C_&T56C3-zoS4&;>|F_Ho}JQVe@uJ7L-O-mJWzNYChr@e z{Xvg>s~qK7rRmG)hlDEh;vvkTY?2!1gd6Ts8;j+%tXBQGzslQQEkG4qI8Xo3B@=Xd z8cCY2@Y}_d;UUB@5Q9taN!!^3rq%=DyaM;UJdOJ!34ChU;s{>BA}P)s`MSC|anu)) zK2>}xv{iv#EyP{VrHG&`{JqP`Y`TM#(GtA9cP=?DYFgt7z(sxad*v5Y;;+T^u#ZkFTkb8{fK-(_#z-y&;vHgS}MlQv-bC)unO8A~(! ze9w(>41oUh{C5_CLdU~~(d^9ciLoZwv`K8&$HmwIFkqGeTC@%1u|qdq3XKBtP&;Ng zj`w{V+@HP3T&2H)E^2J!o*O=V5d34=F6GMZXjgKy)mPVVtUh2XeP`>(3Gm?t@jmaU zS}B7vpUKJe|EcGG(hIR`N)@*{mc?>7_WNNp8-W8A05g)- zMeFGpp^i~Z%A0oB;qKrrN-8^hhndsCV%hA~Dq-#BD6|a=hhYI*R;@wFN|;V6rbPt# z##MA`Qyd+dloEO64dFwcEPR@~ZP5OnQ!MI?_6Jg$P7bUz?#p@Ywjcj1vxdnXbgs!6 zYch7t9Hj^kWqEi>-Lj}?Sx!rjy-eQ+BMVcDBueDzUUf$~YT?nWc!m^9!GMvwDW!74 z{&+MF9$<<~J0(#YXp6g@Fn6a?V!bGiev&!yP!GQ5jVeWJ9m`TX@0z5y?tzC+RPLpi zsXA)Z9fBhI`EsmBf0NL{;=jV`KcDln(Sx28mwJVxx9dNW7)e0p8Au8|AZF+)Ite`9 z7HfvD>h-HmzSZGKYvEIdc+lrj3j zasdRjElHZ*6JJC%nlqcQp=QDuRU8a66;8)?K4%G;I{zT1JlXZGr)0O_KklkUmXI7` zpWYyMY;4A8_{8Y@z|QA^NRRonN2CTcuL-6>)a}Zyj5lJ)z*Q{=^{1tMw#!dwrsSoR ziPcp#QN`@Cx=PwV;dwZOdh)z75*r;fot7g2N7$?vSfW{(#BhH6(2qFBl*OMWMkLAe zK>tjMze$HAcN6oPil!K|DkZ|R$-Pby_GWnUS$Z`g^@CYLDeTEZEGg~0!Ii$YXhwTc1_51>maTGfAy zSfA_n%`2gHHV2g8KpfbEydjwUm=RtMq(CvWz%}%HLC~ zZ-F*Q}gf(@bYf#4b0@EL4J7MLXOMd^XWjzUFylHl#zJ>rw|PiIr3k{srH} zES%E(k@EyDs?b15J>j21ei0PM#BO)5GIF*7121`plWNt|*%DPSlni;HbK++~u-BrJ z&!Z2UzNG|eBfd4>HvX#4FlUJ}JAw=6VT02*`N)<_-J`<1*wV_)3cCb^X^}v4s`3z( zVCrj=@Zjr)pX7rasN)m#a9k6%u`6Lh$hVB~|&mJBA>26&zE1L#He?UUuJEO|FO-FI^?Nf4V zLz>X2gXoBa;#y*CvxjpbH{3B92W``#$~Tw0yqVLH_?KhkSOLcvmdRSW1-=i-6>Mxh zj37jbsKZULl`&-u6ydsJWihBZ@Wbx&HpN?zvtazTjZ=i{OF##$b>)nWERS}YdPk0ud^bs1aadX^gYcHw0gmlk5m;x`M4{Mzm@OUXIScwSY%bln}Ue#TMSh3elv+252j$3(#o;@2}rt5R>6;p?Jo-mO64FR?rNqdd_+ zxD~x0$@_hGAf0S48B!k*@DCST>pJuG?UBO100z5bCRp}q#)+PzZgnQHE1+|%%%4Z! z@Io(hFzI z@=PHfXfCool(nhm;lk5HnXwp!Z^6FHj>j@*=$WI{!yE+Y)gH3eq{V(6e)f%*VW0+Nq#shxmg{W z`y=d6o!#a>CEPMeK+}>gRDW)QZ$|{Qw+yoym{%L;?`|KDDs(vBr?!q$r=K@+bMCd4byh2ubB+wMn0^9E(Sf*dxwTa~uL1^cgj9edf9mIJVRKJ#j3 zL*=5%cpwypjWTTEmAvk61q$Zpc#@qqPSf~{+(<)E&{MbZdbI3R&8nQ!p8EthSyf5E zZjqxZD0WnXyzs~9#*4Caq42#9xGz&g7+mAFhA5a1VB}@EX^C~bVIXZVH!HcJ3DILK znAK83iWYJsvVNd?FMc~2Zju*w|3P2avc@o5`y|W+UzHdo-)^tM%gXIXb5avIeF$+~ z_<>>S%jz+|hxRy#GWck57qgL)2`*=r`-NFR=|7*dkYpzpyY`$PX#X^28ZQOfA09&P za6Mp09Ks#s{{oCRP-=A!bSDm&_O~%zO7>GjoUi3|T|b#8Y*!~f|53`^jXp$JKU3i* zUbDKaUiIm z#Z~tX;QvX~H_6?^YqwI*{nhSHh9jR*AqbwL0ySP+>jhU3t5}88){$4x89GNztKe1w z$I*CrAYQCK3$o%HCEL8hshMGR0fwMX1i&psmz~X_g}=JHyjvM?g{j_Sy4$Xxm=xIm zevtLgC3SG+@wfY4|MvM6T4_Q)9_`;x6I*LWE=;&ho~lq^-r68H5P^FE7>WRBP$~Ig z=hUBen!!-?pCc6OX6Dq+xD0nBZ#IXgN{hu>3LzrkH{tPM8io+*d1$K;bRX{YS#FVs zfVn9MLcFoMo?x^barcQQoi-7I=8wEBBQPDW^i8J&3>z}hJRC^$n&p?Yg34<%g=mcm z(|<~|o4ZGMPmuHfqwBqcnvA;b;Si&Sq9mY#(nD1dh;#@YDbnm9y{Mo_2kBLiP^5@b z1VUA?0HO$l-i!1iNbkLb9w3zOMBn?~-+VK7#(#9i$@84&oW0jxd#`mGPoHJBI?IaD z3!jp#Z$m2nm*`;>0QF_9sIl$8l+hXhF{d)8Ry-3kR8o2&K%69C&TGILfZSE%+1UltGJ&6Z*^P=K zK)~vu1xY(M5&c%ltNqMit&^AR!CGw_S!Rfdeh!aPhF@0P@!&E{`A_`*?}Xdml{?8p zZrOpJ_P#t^3M=RqDv=dS8=y@k$y;*z_9@Ow0ln9F+8AS=jdqMB@Y@X)=JejJ-gGSt zzQ>f#PnK!ld3jPt`P9-Mwnx1KEJM{7p+@DcbkF6xl1PK z%XHDJPoM&ViR>^Ap4*MV`d8ePIgYe@5izTrsu9lN>MlIPy~h*68nb-Nb)zI1{yVe(|NlO3K?53h*6M7YQAn)jXnzr_a!({Z29 zgO;W`B@-yv>0aGZa(Gg5`9lw503DPrfG*{SD`Awb{$&={mS?E!=$)?d)3a={hT*vy z9hgl`^s_v!8JZ4~Ay+3YRFYv$`IrnDl|OKt;(mwhSWm!;w4eCD&X){Nlod0|F`MEo z9}wf~D6`-=Wx{|c?Xv#DYW*?tjX7XX<>5CqKeGV^a76Teke;%~neAXJ5)MmUmq}KH zPRXcq*3$b=qx^m78QIwE$T*yTAv}pwxgffx7{tIQ$47@Z4!GW*F{9huT_!da`}_;q zRl{V0GC9rq-}&O;0iy$&_DFw(%`**ZcV(9DCN~xwr69d2jsL52*Qll@)vvXXQJx}#2A^lGt(xLo6VPzB)ii<27{r{#ru5rpM zBc2+8V^X&f;4&)WrYp#^)efyp)l({l_ixvs2tDr$Z`> zASSi=p|q?He3ymXj#3H-m3J}VgrA9d65bRF09J30An_8)| zvyu-YS)}{QK%2O}XC*?hlX1b+IeSX4KIUlte}H6_DR5-Ew7l-H^q1i=rJ!UV72?na zfzum<(gC6K`b)&Qu-%uym0`ryB#T}T6@ukKInJjAvtFq|;Q9uY?{_*p1rsRL#dAJe z*>u^F?%mZ9Y;vPzK$ufs8;F``yd0T%IpWZl7t>w!f4-;?#WPs`_+3G`&U=OcM6`nP z%}ggc+~DPdGn@ffD@v5N&}i*uLA)xn$IgJIbm6n#S4yv3TXQ3=`^xoARONev9AVDf z&s@s<1FH;XxUaEPcc0g+?&kc@Sgp`5IgR>;5C20Xj1Z`%$EQd-PN@JZT&8Ci`f|e< zC=mO<_3l>F-##~i6}WjBaX1~L6WbI@@#ns{0yGGMvV3NY%vbqko<58SP9KQTp$tpPeaMZ?)5(Zqz8qhF=A{Ug>tWZ2_SXs>|3zu`q z7>mJ%yCQMw!k94b3R!su|r?3;wt?NI>Y>peJeFQ zb?86py<(u+@65p*{7({5O3fuHNwos#mru Mtn zO&ro!jWWZBYC+gd%6ny`w20B@E#fs6SCuGr_tt)gA7Y$Q`x56-T?l>KQeNd~*?I%h z4agwg$x>`{SAHK2~t6xx>vpSpTanmN50x}ao<#VFKYFC z#w#5ZYjr#&SqhM&IKagLj_ZRVLF-fPvTLFJ6gCYu0=_G-Cy19}62atY&OzhkOK!O0!Y#~Kc_c}J$7 z80ttuRPuze%1XwVCt#;iQ>0HI#F2_20U(^0$h$>6P9MbDPVLTOnk3WUBK(}oFaE*! zAu^2j4jhsDw|PVq8p6uFS;dE)e0j*W@d#_h{d`OU1Il~TP4;K*%?26yc zdF2o|^5pNs{zp&xne;E?P~Y@>%c%dDP0}$l(;vx#%+L}8VmB5Juo|2cB}3(FHPmfv z$GeS_{W%7@&mgx?&-8tq7r^Fe1e9CM} z0wd-H-g3Y=VjZwTc3@a8XUvF8C>)l3dh6HKJm=?R-=@`((y-gK*eQ$(Rpx&SaB3aU z^POm{y7mtt#3J36a*E#nbGkSt`icX2o-NLMXCNY)N@XRrAlz;sw^W)_(Jx@0c@F*T z5U5#^Gb%Y>z)05r^_Z`cs>Y6Xie5qS=x4_Wioh;05-@nW`t$Z0S-W)^tr@TN`r?K^ z+eBMd#t86q>PJ(39CPV6s-KTg{lytAzY8c&r$BZ55-9BiDc8JPws};d_ngNIM)d^M zJA8~Hx|Rpp8l}(s&A$Ae@-NQi@W^L<6Rg-xF9>Puxg;HpEC}?SO|X2QZOjy91I zF;eA8pBh1=eYj6ZXN_zgq~P1Zq4U3QPC;R-{8v6pR#(nLU%{2pR-rRTQ*uWZNdL8S zKC=uhz;1oah*1Fwmjyqv0_O5|nN*yaf>JH|N-kBOs56R{Pk|%$Vyyof)CBdLFoU(bE=tMf&qyh&=ur^8S%IR!8?`T@+yx?$iOL)3>tNGxxfaK zrr0Iwth*}(;lx7gxfR*&p+GII9g@*(g`JBKbLFfHCj!;X*zhX{x1)~C6o$_r%_$2HUT<3dSsoTSwVvjkV}Wnsr&}# zL)$OANV#l6a)_g z&GIq1u_}Azyf~@16a}1&SlB~T*Z>0qSIU&`J`{KwnbFXA2!>5PR3X&~Y6f*|>Cg6} zgZ4cb(>s6yJFF|H^RXFRLWR24TP)=>M1<*C#eWdTEu{Rs8PsdB_tPX^{X?xT)#NJR zPtMiXGo>5*Z%%a5v46j2L|B8%IhT6^gR;c+_4Qft$~~>dgZn;K!2@UFs^k&l((ecl zX}b1+`&NB%ANzjhI9W1r%G^`cfs;97;-U+XGhDn{bnjp4e=DzVPF6#oQSrAd@kXzg z0}19O+nakqbUOJr|9;Lzhp33?g76I0NHO1|&6F^204tK)U7f3@nAzzf^jXef|GM=E z(L(`K`xM)y$015w0|pg#JF6oG;e>~AK^)_Z*Jy|XUZ0_z&$IVf|L)!>`5S(T-N3f& zhos}Q+IptD4X6=0_t&VoMXA;_13@fVh9UgOY0qEn_$CZ7Ql0I-{l_Bm{8Vv~+S0ek z;vqq>gQ{GgFCKL|#H4Qm)fu23faJ0`ZNcliMXv&SjTV46AMCLF-``_&h6ZL9u(e#@XGkyYXU^~1w!3<3pAI5A z#&~PwwZaECmN*CtK8BTYKSef1F{91~l&iLu*6kU6jrdWQ4bXa?hi zy@w&U82}!);j{-F;_Po-6B}dqnRllstblrJ0GPz(gK|Dj{nFe?)L-7)KLH`XMP{nd zRB}B}g&JIdO9hVzX;Z&3Th5S_|9~u#VDWe93pgVRfs<;B89AX^g;5uxO{x=UZVU0N zFeGC>zVRww(;sZ2Q*4)tm#<&!-y{3ZRe2$?Yi*)YCMb2^YHUDiUq_l!a43>ec_k(u zjPJgVN)b+h@dapK;!>dkkX_wV8So!;oGc(|f&Z)tzqNGQ>|zGQ*~mAXU~n=up>w&z znkD1Nex*_WJeW&zf%T~uXbV)p#hYtyydzw&d&YdQp_W!YN?u^0Z3(QrTiol30v)_YMC&gnTI zbq8|qw^H(%0k>iMxuvUDV@= z^1^8YDg9C^hZBR5Kd(TxUoOJG#+Xzj9GE&+PM9+*A>Kcq(3M{_TOKVhy|-f79eBI* z8ki{T$!4G`LuyAjY&M;gWANdxBorfeJbi9JE|X zkU&+p);&Z`GN$HQ#r=WtgE;8p@~c?SDlbefwAQ*FdlN@=Sv!b@d+uOC?dc_47Vl*> z*8rTa@c@!Q!RT~?oS{uX#B4MzF#yDf(VdFi-W~Nt1{CeeQTK)*>a60~XoZJ}&<&|{ z4=iL-1u=zQnjpA!HzIt@m7(j5I$^3j_oQDe^AWV97;z@Of1y{aI7P0}Ck1(jUm#VL zS{UYGA>zfmzcuqzsDo*eYRUTK8gFrOA>_--=y?0xfX87+Xh*=YK(HaLdr!5 zFz}>PQ>Rb0v%v28NW{-Ko)}FeZr0Af;nuKK7ma`$REp;~@;4E7(f= zZ53YosrcK-fC?tILnBl?&&*6_qEm*aOza_YE&2Jxp(|jVuF>#bGH9I7y3V$pFMUz> zSk?&uOJPf~H-x5oJ`*aq@Y6=D{9Ti(#h7>0mlw=3Urp>yjt6MnWSBg|P-}P(?=~BT zb=j~sr^)A=n96*>#X@6-1+%l{GQ$q4EawA0>5c5!{hd)y!n-wRjClI-8tw6|@JsRn zbXL;-KpXZpm~M& z5N9$CKMrvb`-hntOU874S&5qm*cKU?l$LS{;;LI+C|2}q%q1MUE{(q|me}wo_&}m& z&qT2IBqtMO%AtH(T9lNs{Ycvn9*E8|WmQmzV;(}^#+oL#GAstIq=ZwG5|+-Tw293L zb5>PT^)1USQ4NJtrv$k*EUhEM7j2p>MiXtGBz}*8-*h)3qS3!B?y;3lWTZ#~NQf6^ z4iU9SPeR!^yMq&D!oy~O84{Q&zivo#x8O`iy;vn5`>AzVfj+%MS%I4DgT@G!t2pY% zFm|2?fu0S!0e>AJg~njTJ2MRzDX^&GK=~=Ww#;_#-TL_561CJ7nA%IOjbJtQ)id+q zjml>`1h4uP<;D@_4=sV_f{Q5>(5a7FkQ+chjU+k2a|NB%e{{uP?(J5H^#5#(z~^}-uH$$>>y1lO zHH{}(rs8C@|L|i}tHo->)(x%p(y@rTrA^}!&l&Tk8zL)@E`JVPtN^|$5x152`dAcy zSvuuM%%O);D&O(?90Ut?XRq$vdEH^X^|$*lJ?NMBdg)X@){u8G(JI?sW(pV&Fo=Jh z`>bzeID~VX44DPfdxAeH2X(q+m|dXQG`!%rR};Q^&mW@9HLOG{bFZ%Jw&lzbF<5?e zdySFk9P`k((A2{^r_yrQO!PLdTRT%7EzNDzZOraI>$^ivDQ;r#2=p(#bb>JVL)|2T zBYSP+#x(2XwJsjUp_<6N%zNB5uQuZeV%-vZNO9`T6MZpaD@f}#>=l>L-V?N#&OL`f zaG)#tm;4m!QsGK&h1ZNuA@3ss-WXMRr5Qi#3RGx z{pv$sRze$t9ByT{B|1zzaw#ZG%p|)kr7>aNJ^#~XkRMlgt6pDh5QVmAiQ=r4ZMzql zjKS)+oZ%GzFudGQtN@HOgTpu;l#+L}1lx79sj?<~6~4MI+I&8KPH8AIFlb0F5DRBA ztfc?o-*L!_^Q;ImTqIm96rVc-epU7y8q%@TP;xSLSZB)VxYfEi@5WcGgKyiI9_>=N zRbb>=wR6#^N^;rxhY*fsKSS4prdNAJ)=aEKXsvIC#uvL=yEhnr_xoHBE6`pSUo z@F>O!jj08mW%-ExR$pAfu35h*2{*r0&Th_q2aAU>n==cxilfHY61-$t&mtRk4O}yC z!}$6VJWqQy_}h$E1n_arDl#^55Rr_RJ)gUVlV}>}adg%1)~ls?zeLWC$~*wXXUl`U z5Wx^H$??5_=1i>M1p&YO{ji?~kjb!D)Yp9HToTx2HkvqRS^@=;0;z^;z)b|pI$6V% z52xU}DVGKgn8QTdqqC9n_4RRrssdZvH7Db}O-XyGsT$@lm5-K)y8xM3u~Fno<2h#E zS8t>G&*UNv?Y^y=T)^YjzwkD=L?_lUR<0q}{TzNQqx=i$O~0!Y=%6O&VHcDT-sGxb z-6A3kwy)qvYB{*oESf~gMfBL+$Gxk(HeGXQRbf#)CZl^J5=Fya!fuNwr&|y99Pzl` zDkDJb#%IlvuuindknR54n_j*%J`5Mi;yi6I1WP8NQF01okOrd^c=7f^c1XT9%n-BG zQbU{joSGEKv5&YN=x}EI*h^(w@+(H$~#$TB?z(*FNweQm!{vZyw?rL=G6+81uEgWPHa* z0~|uC8LKJN!uF{6_sp1_!g*LT-doD#%Ho*FQG1;!!PRMfhB`Ei9T#%zt?NB|fp2?n zaw=0xPO?`RZfHOl#(-R=|9#wx?`}0>d)-yV^Rt_4;Ik;a#sDQsnG88Ms96G>%0V=mjWvfoGS@6MT+IHl$$0cV=?L_!*h4)XFyMsesR=}bo@$_YjE^QXHP`_}`_`YeTbwIQv!rii1~^tNHCP_8(LgSTfHK*?8-44KnT z7nBg+U4O*~#Xvg&;{O#Xdk{4Rx)|e%+Mz%=&Tnj9P)TI2wIg4`J!r-GmFY5Bx z#Sr$K8&Dr1_A9JPhE3&G#6uTX7^3Yoon*&89YtNaa5|mbqi@F}a*ukuqT@o(O$d3pD=7r(97gn#m89hx8`-b&C z@KZj+b!T!7t{ZW?X@R5#=(n*zgYP~(B3`%GKLj|WO@+ZgpN!g==f+P+V3g9e6voPk zO??{VWfRAPL4WqTptvR$p4)re(SkQ+1@{WQ+<&Ju3SOo(nb}0`e~aKdD#JB?m5wzb zxdZTDF>L(L(OosD6Q>`fIgn|J^*ZsD+o~>- z3VUH8f-YewtNmc);9G)+*a5c6RmZ1k+VfPj$U^P*p{=*?>*9ALBTVuO^ zUzob@rfvG{5b2!rJY{Ll+`) zHUl2k=Mx1V+*{R;n$SQ!AV@J(NF}`u<1jx9r8c>;_eP0=H>sdvcedXd6O$9T;ppFH zIk8s{eVlCKjR@P&kR?q0+{3VLidXQhdnv55h93(*)JW#=VVU$^wseiRF83u@g0E5_ z{oJ_@jQ#XHxq-e@2_l(p*TlWyzYR6khKK21MCvJD(#th*1EhzX!0^ii{$I@Vdn=6G!7Dv71Zm;*4*kdFv_x_ovV3{Jo)${@* z7nC9EGVhxh{%m@`XA8g+*pw;YJQIGF$4Eb51X^s)ZytW;wlre^3(6>b{a1|#YSqD8 zd5fAFCOylTFO}1Zrc&^(1UPa#?0HLH3+!^e);bo2YRz@uCmy=Y(*3( z_Q3WWKlL8KM@d^lr_C_c`S#bs{a12bk?-)MYhkPnS_1vzwnuAld3@out-Q7s7D?js zD0uH2Qm&y45Xm2WPT@?N$@aWrm3wV=H%G!025pNz$iT4Bj`AzSSDM9m*(s!7wo(w< zp(BtbMcva&4yznq&B=GL*WoKL2OoNcMJRo}NV*NHaZ16A1bGg6u_Xh}f_Lokw<`K= z%U_{=h}U`?`P@ghe%pFF=-#ZbgFXF46C4gs=nBEdcR@G zh^zAjblbfy^T!3hUD@QXn>s`IV&6?wKQ^LcO6YF1rxIr27Cw*jO-6+r*DvM?S(Hy+ zQM^-zcE2LzS7e5lFZ`fSAo3O;V8HTbvQq3Yc<;ZLo@E&Zq{ zXs1pAi35V}`_vN%dAlWqnIbSA){Utia{^qonfE(aPNcwm#!GXT+-gnl45=eWPHKqa|&8>Ypp_wNer0Y}0h@aA#}$ zQ2M*a$fdvN^Or^0C91|lLY$4&Rew$d=+X(kBDXr#>1Znw)s`Olv z(g*cLq=@tDz6a)OnLd-p?@IZP)=;!L$`E zUIcQpBh<*b>qaf$`Naq88BEGX)3%(rZ$YWmsvdW`hk*+%e3`%`js)3HxAtmY*>$oR zXDx856Y~Q6b}e`^@#Uc+JA0pEu;+tmkhO z3KxFRZU<%x7NK_*qBKk`U>_?p$ODtbptYm>988?m3i_;vP0Yz1@3|R`{CPuKpwnZc zAUfc64tJ*_gXdCwecV>@b1J04{t7WKV(A#m%4xaXYWu#2mapfm4gulYR_(n4N;dr3;|NVYx7PL>q2f<{jN7MB-iX(D3-EFh;5gzv1Z2NL;@+yI; zhiqSd=6LS-AY~$2mGxp?b#7yB{c4XOe)>&UJLt6JdASPPJ+I#%*2RGOzZO6+NeuYb zwE%vhoe4LSCoLHFDltu`}N`_bM#f z&;V0#(35f`^4^m?!Rb(sscYS?m09*koGHzF9C!*vpQc<1`a2WAh>B_Dy>;%zB60IS z&Yr-iJ8b6xT%J`-n$=?U8r;Rh`~k76a+mnKk1iWh@YBn6rd!6_VdAk}`F5Y%CQIf$sz1U>HuFsBb%?6SLo!^(2zgd{U&mkXj$p=e`DT-x7~P?Fr6@tjM;*v zk^vc0Tadf7fi38bn$92FfKxT3{mg09v4)TUNVWkv<|qtS}qyZ@*zlDO$xB1*sdk zbhPJQC?a*D<2`T8JxzxTYST(s{GgQ-7q+!<=m0a%b`i4`+Z~`uVKus0I({aD7h*Fz zIN=}k_+jKyA7D8O%oz;rtLKesY6BtDX|Nw{J)Q(A`QyP-0}hDR91AN(4s?xPS}3py z=B#G;uDq-ml!Bi3D~wplHToX*h58!NfMJB(sH`RU9+UFLpZ>Zluc&2s3;#xY6{s+8_T5ZP0Vf|s*A+3*HmFrxhfEA?$G*( zAjq_b;@kD8JAyP#haa0ONo5uskO37lQ_7Q5=R2fv2<3UYqZ^M<5K8~B`w*#v=;HIx zPUq2y>djt_FtvVf=UwkE*GL(ED)~m-Las?VQze!m8F6(BA z&ahtCz?ntd8FQ+#WY6B2T}V&P-rN0XtA{vKv7 z+Y1MkIk20io^XVV8%4b)BX*$O+E{gsV_(9znWZ|(a#}<<#1wo7OJA33o9b0|4Yh0y z(z=WHeL_?y8 zSv}=7*$dDp>$j&btX76-*0B*sJZfS+A3~$fQL<`t84kGc=dj9ipu^pxB3#0>GaOe8>XSD_EaA?zkvhI7wQ#Zgg;yKK>UVzBaooiZoYyr(d-Sd4KGe|5 zXJsI0)_iRYl44j|*eBnpf-cjs9Juk}I{z18fx#p1OT+MY4uXdf+9Ew=C z*)FbTi1x@cScsa z)`4u|QD_}AIuG>KX*enLYR;`QW>4OHLJL&L5)^%k5kiXRg&*V=qo>Xtanp9bn1fvJ zsuP_38$qO%{Fj1 z&W}P9&p&E?h}uaoSW^*UlX&CW|D?aDFIHj|VbT}tWLhZFPcuC}M1&w=cDAmo zknWEyw0&Y-n<#~2_kc!GeQ&L;#lrT%L3fXu^NGigO`pj;S8OPsd-=RygqO(k2T^Ft zb%-jwh-AG<$@6@ucKP=pSb@QfGC3w_?3y6-R?XInjJUoh#4i1>q@tgc_aOR-I>-Au z(DkO~yY|2Ko~7Lvu)RJg9B@;+Y#*pUBhgKhQSJj@`;us|YG@rA{QUJRdj;xSsxnU+ z)RRK-_Bw(aJp|Zt+cb8-KBrf&>xA2ni_dkK5j z36Y+6h(5-ik-HhLwC)A3*G$XNH9DB~klWy;HQjy$(4eJgZ81$^DQXg*@iotEek>W- zxR1KZO&AAAUbsW4U6}|_#%jh-J))g8JxG&cDMp97H75x4vzZBAk~|8)33+4wF>uJB z+}%>VxdjH7R8QgE;ry}m(JFLS3{v?`W&R&uyG=Q1el3jQM5_}5lfMb*42bsx=08+z z{caQ}6SRKs)%-cmd)x_I%M#*0R$qHH?l@73RrmgT{T+lo&c5BLHa8OTXk(nGn}D8+ zddkNisWzuHdZyJHhxJIy26-_NUYr(pAQ_TsRI8Pn5Lo&*Rl4QQ;o4>R)41<{XHHqQ zuwdn<^I`Pw#>>#sMS>Kde|`HI@6ojEz|R}M^@&Ly#*X;d>^p^{vDW7u6x^ac*Wff) z2p?*CLBCn%ft(Y;cJD4S(Ech{$ZlF==e7pbDGNutWt^T|VuYELHZe~xQIM3a(R z`!?konG;!0;l6boJ1>q z>~9!5eQsa^9*)!^_i*gVNTX5H*Ov~BGutu43d%edmbP#(cp&|NpHuH4;YDHK zkXiMA^NX9~Xpi*qwBhtD8sD^4XEUc>d3P-FRm$wC(Y54(!v;)49*HZhGExrJwOY7c4`M;l$cv%1DYj#`a||De%Z0)# zdX)0BG=Y5SaU^T=xAwf{=0Zc4AzF9|L3~ytSa2p0<}H9yC>6Gis^@4h-`?n$X4t@Dei^m_3ESoI)M-3T@K&P1^AEHIQQKaiJ%bk zDcgm|m})=59p&&q64dIK>M7fU-rve_Rd|8MyobHP4H^U|LS)je{B92Py**|_y(!Xk zGJYU;Huq+F(Q(xgcm0zo4}H6CyTQ@e z188Z8-&ifU((jn2k~BhS7$tDWfXZU%rs)^TX$7`CXL{Q?JKQ-;-$;yMqgB@vO~=p_sZ~bs^t{nbvg>$caRJRr$VT|aN}hkOa=rO zseh3V|5aXSjQg+xF2FG~?4Z|TXZ|f~l4BDDQKq>|8Iho|C&xMLE`Jnea$vr@7x#%5 z>o56W`!4c3$?B&}A5mN}722s{wer2Az2r)RR&+~x*5Ey_*_z2?bPZL%oeB-=VC{1V zWco6DqU=d7M~P38lEQgYqSJMAPd#zHZspMzD1x#lKlUnX>$pdrjqt%8fsk{wlEn1YV57T%x=6o`)M}S$wQH4<{a1;@;l)B{B!3LUzpXlj zm?9a^@(R>RHy1k@4XMH7gLxaKq;8J?^7$tCHtiS`b53gTKh88?Z1T{ck~{3T!erY! z8n1WuVPHxZ@(%4ea^vHK`kzl&?=7pq*q8EUm)eL9+_rhY!#QJ{PB_sL&jx;HG~>Vj zCJ*5I`wSN;vPo9=l@DvB;2@ox&K%iC^4^TOGz)uAUp#u3l2?|2pSK>131l_D$)h76 z&vTEmbYlR^e{a_?JDJ&t*?m?hw^X|ocW3@uf94gPr+uY5{NhVOuNGns>v);+D%FKn zP&LC`dU-2@i%c)KZ8hX{LFpz6d()$8*BzC3qudF*NgL;$mS?*&Nioe#N72-yJr|ZF zj(NEAP&@-RQ%I-x0@|~5jba|S&N0gPZSiRHh`08vfaFRAT@~m2BYn!frY@;*CMrDG zRk0GQ0E^CfzN$3R-vH;pXlhQrj_jo0G~l=+g!-+2jacL{sfe+`-dwd#a4D(eF1t=# zt^4-po)?FJrbbLD507E1cO`nSVszW~=d6bn4LaanIAn|5ZO90LH`N*+>bDi6_~>sL z8dwwxriT+h8isQ>`kiiFU4Lyt@F>?$^nf6jzYEBm`gF+fTUAsp*fjclHn;xl72q~s# z5|-Q;h$qw;Mt!90MAacI#2nMwn}0jL=GNib(r)UjJJ>+G~Z$;l~2*xLwUbINsE4QqBlP^hs6CkZEIB&lxHb-L%}z+#iDwmg=O6pm8j z)pI&eah$c588LpLGE$Ml&EEQpYwUz3EioxhrmB}-Xt;=udf2z5gOh3@gu`IqBe7xC zJ%mH~q$0$jMuyMn%)0$sHAU04(uX;Z6Y#+=uA$zj6+TZWy!J$LwKOKsTePeG1fs>% zai`c-+g7u`d>h;!1@+W>-fn7mble@6X6iAtP38v_ih|&^gF{ANS{bW_tLAk`o!weG@5j~CUh-9U>*Xz}SvUP0NnSdxsl$3{YTYLo zRnH)gIi+W#JLw zNWb_b!F}Wbv_;$0qg<3SZfqKmWfwi^FM+F3eRh60bQdi1mv~DnnZEaI@LuZeNt{#J zST1%=4rOP;=M68%BA>^rz?)CpSClV{-CB+8j}aRA?&$l)H7|J!vjU!@Cq#zqlu?SP$sn-eQ{!{hPlBdrx3Zg^k#dSBWS4O2Hm4(B7mi_f}4I z@||E5ij8-gUY3;_b}WrNtNMXp^H%<3>T9~88!4`DzX0+Q%-&wA9t*( z=it}|Q~l;W?+#i=^!u5 z*?{TvM}}lpqea6HGR-{eRqezvDyM9%C>nWkcD7%VWAm!7cW8*J;{*S@6cI8t?10KV7OI6#I z&Xl|Rtv}Mt^+eMu(;dBAvcTlS4R0o2bze0^_=kvVm!(c>Nz16Ta!)3|4K9|oGnBYU zZ_8TDb)ZEUWz+3{QPX*kar5x^;8)1B0vf4iXSCkOF+Poxk39dRb^mMJxlMU z&&oWqMyH2@^5o*poO3-FT9}^dJRbRqOdCl>i^sORw5#q9*OAK9gEzTB#NrX2$hFl5 zh+=?JVn;^8yvA;tbK+zwFXXaP8%74y(K+m<=_`Mn@}33QKv|sxbh0aheWV9AOn)DK zTL1W{WVAeM9c?VWdf%eQ)686>?qvDn>xxROiii@!F#bjvN%{+g`kbRBxO(`rSj)ZD z=n(PTAdi;j?ap%7T<=_8*%UamUwtpcT$wM;Djf2LWM96{y#SECQ6#74!`N+LvTLm@ zoG0wM?mE>M-sSbu&w|F|*d#~mB>U*Cx&-*s)sy!1{iFoi0oT453juS93p+m#rw-<2 zYrQy8$)2!AOJY~UPS-P1E`copY;R8C!JjGi-OhAQ9B2i{nqMJT38ci>uozsGQaNu@ zv4mRyqNPV6=2}aKdQXUKsBd8Efk_H4jq2b8QY;>Ro#B$jPh$&v zWYg;}qhf21q`K=j)TS23G*V%+De~rUP~k;%FdmgE5K|!PvJJhn25uE zOC#(h)!T17wTulY`K2lPwfi5=4z|#9m`(Yd!0nC!`ge+8KJd_)^x~Bt>0S-(=HMY$ zS0H|lQ=qlhJK58eSPP?Ky;5co$cOfAk;x;O^e5N4*OnUBXCyaGO0QamAPQp;HD8no z#{Dwxt6y}7q*qt2wGiSi>ejLM{|wY#Sf?=|j`l32^!k{#TY)wdbjKI8;Npz|Y@!@4+Fx|MW7sl3D|yp?3LRG0L{ zS=7%yO3~bSfp#8!9^Km{Co@!oW{M;8)6Y4XsYt!5NjVvcVv=&2KH;;5f=Fd}HncHT{@$+e1|>T$c@G*VS_QxNRPu$=Lj&H$y2w1K=W&Vac4Ek_@uNTGKUda@41Kp&bw}8Q`@djCJ!nOzE}x@) z=T`n_XIrGdX04|b_u>_9KLoVehofp~*PemhYIQA3iO= zhXnDy_81;0v|}~HqBQb*-k@C)IqeA#{T#D5!#;j9&9x-MTk%Jq_g;#tA{YC^hOhmR zgzuOMJ=*6jVS}qR3yO>BA9xKPo)(sc-mkmDkoL8s_G`{ioyPKe7MuScfwX8*vQhB(!n|eqL@b{{)%%msf`4^Re{x{{?jCUWYO0oh9su&7<;3bPUcKDP8N$UWXBl^ys}Q7bQt1}!*+eFx&66Y^ zBEtB`a?yema9aIInm4Iz14PgAmmMZvfYYX0Uu=szruPf;V%jJL%zx?oZ6te*5kJO0 z&iNTWP>DwrTb10xBh=0jyX9j3NOn2yP17wdC<++kDVjnCPKH{YqmN7Ca?u)>(kKI% zY=(0@DO1ZcFKaF||U+c_{3PoC#5Y*?CfP|5c=u9Pua%rf*iC2w0aKG)cZURQQ| zKl0>wObvuZc#hAGJ?sjECF16o0Uwmd#6g(&*1z;wYY-1KiuddnQQqaJ^YK~4Uw5zb z9+Ddybn)EdA3L;in4EDh-x%_k1n>sE8NK4Fp=o>7|1qyTtnCD2|0DVzn%!`D_3mFx z*?{>RZ-PA`XlWgAqSgvSniL+fi#>nLc?$Uu^`frj7bE4BI_AkpS!h%xGJk_*jJIjp zsi89A)yv0?$`pFRGfWJuW=dPt{KI{v6}6P|s1vP}Zrih74b)b@PG*UaxTfZ1rFKYs zNkjiRHcR5O;eJf*nq^#HHRx{2)Q{C;m*(^tE-}Pl7CP_lB<7j6zi&Fsx=9^sIr$h@Z|Umtiar5r)P!M!zdG)0zW}J!5J8-u(;t@M4cBK@A=!U z2_Z5()lmC5!`E^ z|7eIh82^D(0W}A9sAokOC$f|n@tAX|orc%66dO zBZRu&+5Jb2bok2hR3snTf_U1Z;y0HQis_IgEzbv8U#A1~|8vjMp&u-;$pu9<(FCF| z@h7EOXz>JVpQpjVS83Xt!m)64pAG+dA^NOu%UIT!aGzJ-Qd{A|MzUpivqu@uF=1m6 zo&hScb{p_Uxdrz5$s?+bK5Riwq}OD&)#1ruLlSbW;E-H?xAhPW`L=uL*Y#^(dN>3KHjplLNvM#*vP!n+G1^HCl@Wit)_yJoaDjSM_P+r(E?Pxna8Bds`s&zs*bj z^CU8_#|>+Z>H4FBQ(P5D9~r#CGp>IypAi|+Z=N>Llf<5zL6=cl_^YKKMB7`?O>#=a z5RxNStN_O@r5|v36hW^>XSH`a&@BFK{fcurt32+O_LMVs_$)HBEFkC)>71KBJI!ctek??9sa7X4z$S3(LHFM)B0tO3D_tYUHbcj63qTd zmiLecwQNU1p>iIR>-TkyL?6(d+o`FIQdb*u`Zb}pQ+|VY4Gb|v;&-QOP>>JHrAI(> z$A;0neb`6JEm`@!cU`I*3ZypCAnpQL7G`0G8h!u`)ApN-r0gZ@ zTJv3p8&9nM>^ElOQi!BHrrT#U*s>w)0@QvX#c!!_ZKXD$EIp9z@|=zgs~cfs4v_5W+{%j2P3!@tp-nrUNZQaahj zHufb-GSb@g# z_t|3FMZr}H%-6`H(`?w?1eY2&0Vui)o`8!#^&9LMv;^~5m=UIbZVKWLK!l;d zv_54E@(JmVo256w_49jM%{<&*a=r?=qHweu#yV5NV;Qiqy1jq9ChBs5h5;y8DI&nV zr$aAD8M>us3n-@#@Hr%FW>A>+DXCW-@QTI;b*Tb&G>!^bNu{rHx{fn>&IT(FU9q&?Z>=%E;is%fB*mNw zxTDIXNhV!*aRAV#e2<7##E*Rrc(b)}U!aQMyj;gP2xWiOMmpnAJJHr5gl zJ3<)mFLGmjzhh9EcFgJdi(!*dY36=#CRq+Z!e@RH1#zaVSV(qNF7sn&Gfx+`$H#sw z4+EA7GEUpU!@qK7xKlU3>j~@<;wEzF4{(qk--^9@AR^NgRLs!csYB-*YgZ$1k4m44 z;Qq1lrcXQAiROzCX=XlP0OCIJV!neWaM=v|dKqh6%Ylpe$LqJO4$AlvFn(c9lMd}$ zN8>>9oJxmiDDOto^0zN}@^}RRQ**_NSKQ+8oPXP+&gRGLy!3aIV-#Szs$-$UXWLpq zWAFno=X9@i%&iLFQ^PpTZ)?t*Z=r}_<%UcvvP$%)ol#bVw?TBYMxJVDjK!I{GGp>? z^xAs1c*M|ayd$R&Jyp#b<$-Y?Vgu4ZY8=3c0~!`sb#?Rj7PSkM1${~Z__ zrfkLGbm6<&1kB_D{l4{{Qt_gXmU=&x8y;acRFpEid-2mjadCiDEFnCBt zeKmDCoX|A0Y34VLaV&(1hZdC7fO&!})CCv;(D5XYQhHC$5S%VvK>G9|!yq zc0P*0MDxSo3gERAD$bt)_cV90^b#xC9Q_M2(#=+T50pZzoF1TDW-o55tzLPXukxjo z&3fygUqgQX7h8CR!7y+Kf^uqc8Bm`q2*z5$GNWr|2d}R|jIhD=M;k+s?OO*7E505ArRG=+#?Q8LEI@{a5o}giZ$|l;Qf;p{5-iV?ReIHJF9`UYT5>{Dtp0TfL>pL z758Qx>4!jv!$tb%?|g)I724P_pr9fv>=W~lOIAc&kD?3Y=-Lah`F4bup$P=JWrz`3 z3Pxs0JJaB!%Mdd%YM~E912-WDbSZ8U+CCZTJ>)YQ3^TcLaNJiY?O2)pbAn=Y*QvGM zqfDF^cG_MbeE>sYF0Yyn-#~UlI)c|TLbM{{fA1y zY&i)WFzr?w_c&I~g`HlFy{)ztbQ!ttvNj;*ajZ0pF^u7hB z@6}6jAAzKy2O>c3Wj!X&C#?Y3iVq_CqNae#5BX2By?(Nu{KJ-Lnu|i;?$`NIY3tzY zxaoc3oIu|$9tLb9L9T=v-1engCye*urRG6Ns<*3m^+6u9ogs`>Wil1JfM*=M?`idw zc1R**A&}HC2I}Q(A(2{_>+6s8`7?_hyL)-lEkMsMUPS_s<+1XNTX54zMxYt^ijfXv zrII|$;BUGixLP!RpQUmx>T9`S73}yt;Gw$AF83_i1E;`88Kh9w63G?QaCx2LZZwOiO2d;0q_$RQvTE!QP4Tf20 z*dFugr&Q3Wbnm;Fb9D?i-AHm+i~ydCfUwr&*P*$QVZOa`%uJk8K=BE)?_rwP2r;|tpU|^2b z5gsQmh+_M|o3_)EubvOgRz*;B0z!Dc9`04a!}8XbN4+=W6Xr7N0ik77n+8L`ozI1q zC54vMyR111%Z-5*1bN{~m%z^04l7f~A4;351MtjO*X7n9_M8R5v(FzJM0z((k90Ru zZJJ@QBD&Ze}wlw;E0`#Lo1htD07JU5)lt~jVSPDoPd#WC14Lf~$1fGuNPft$6kRA~@K`^{!qc;H&3 zH6<4%+M=nMKcXHTXQLC5VAXv~gl6wQ#eGZ0K2>mq(loi=vKgL0-vgR;0e)Fn9~u7P zB!lQMY6e3*M7}oufix^UQdkmWp==qvP+9iHnz*tUeO>?zE$eGDX))B3JEHzvC@|=| z2H&S;$o=j{OTL%~dE@KS^p@D3V5AWD`tU%eZzUK@>Qb>02Yk8JWE+$o-E>4ydo{36 zE*rORZ@sXrI+$(*d#7ALBJvI~TBs4j^+Ez_6yh5n8eW}g=zcFUnc^tqMWzH;AU-5x%S_0Fne1^U1=e}1KVJKo zGXxwc7U4hqL@|tH$LknK;Y4${RIQW}r=KxIP+9|q2UtS zyqYA=df9n7-0(iP<(ftz2sY=fNWW>wCisd?u$FsH@^vwm9V94T!Lbn+tL~xn;$}e5z?b@BV`AN>rLV}uN|uN z!>MoT|AdD)$Lk4tRNwjhmwaRxL%zFZ{rlxDyl_7iEEB?X(FfxE8XmdZ)gsv1qeu3v`nQPB7VIkQ2%MiV}Ib<~y7qm6~YG*@sq*cS99|6qlRca z*4DRcL*(9)aDxxaBW&0HvLq~pXqlSz>}#+q{)%!D;N}W@OP{n$>$e}*`gR%4q;_Af zX06m{-4}V-@Ud<~p9sb$V3W$cCBfxf06}KXpHVZ}k|qShn}Lq|S9NN0chz#$MRe?R z^fBhHD}NS_^dwmz&1b3ZmVAv=K1=x%$20>tpPGgx=7>_1{-pV|>f^k%iwD>b(}#Nu zCq>^lCsF%`(4DiIjXJg;Mh<}7Ll3o*)izk(%>KiXQi#0f8y6rq_=^oO76!_+|C-00 ztSy;($)rw*CC)JETM7(`R=+8$)mR4DCiU<{83$M3O(Bajc+Mp^xLBR_>1#ONUV0v% zG#~rvg<)>4ku{MIi@i2a54c;&gkgAedWVfnx!4byr373&IId-$09rkmpPK1`Xnxoy z$x$110`V{)2TZEFUA{|4H1%(IUCs+O=hlL$+E!d9jw!Z!f8q4IaQVcJIf08J>5S=z zFvcZWvwUyy%$G{EW_`@j7FE_H`&TVQI$D>;;VbdMK_cC)CO57N(nXSMw}oa-YdS+c zBiy=3w|>(`^s~^OOA!Amnpb@7)l`6e9q(jdUk9zc9wddBl|sb@JAT1px1d0-!>#pE z_$CI`hBKWc^Z6$S>I>%T%)+9wh(lnOaXhmv8<&pV%CM92BF>il;%3Oj#GyN(G0J?O zf{*Bhqa{FLiB*$Z@!+986&@*_Y15EfmQ7ivAcY~su0#MB<`P2NL&%*2*MRcnH#%9! z!AGCfji-KVS>AKqL3oYu!<~l~&jZAi<(8VCv?WIgJmMk)@+(nOXff8rHpBYP{?mQs zrP`NHNpZywI~|`|wxD_*Z6=~=$9q2{w31c*$>9EsmZH-W`{Dt9)vof8*0+K05>E>ylOzjt&RBRteXSc!jnNJC+24Ioz|k0H~;;*O}=*WsVU+%=E(; z8=r^d$FciPKO{dHqwuLvWC<@x1sI>@Q2X$*meElA)?KC=$M(}-erq=Yi~1l%!By;g zvzsC>%oQZTbgu>F372cGCc;DtB%*^VCAl%zmS3D{yj0OWDJ)H2X+3Q*f_PX`*dJ|o zU4^x&)?duAeKPGsc({M|!iUz)j&X5@Psp$vpCy9=0v)(N2p-k={?ZfHx}Zr% zUm2=NBpD|4uRw`C92m@Gb*v5YY3DuG=I&8g@ceiDm)KM?w9Vpwcnm2cOpMv zGkp*itO#SG;NCzskf+gpIn-h3N=S++)y+u67wJx|biIW9GmsDQ)C#r0H~S$w376AF zid`%xpR?g*QVM>~|DK47A4MdU8CGx)gA_W8{rd3QX=1%ZrQzQ=e;UGsxRuu?29UYa zQu3^lFsV0H-$#Gp&kAKuiRvToFHSE^o#T8UeyULW(j4-EYeks@GAihB$Q(-pw&%FB z|2!DfXo-tK^=9)kv~rEzQ(@<(8Py9{KI#t9FX>+c0|d?;On(jGNfrIsLz)aTzCM1g zu-julUV7>jkkP`LXOmkY%^0WG6`!jFdBcW3LcU0a7{*s-g}M;XKn!0LK1RH9Kgsu& zZ|H@vDqTp9%5Xh|(jJVfw+LyZ6WBQ&piZ;+Gc0er3t+@GCTEt0T+nBi%F>9H@nDr~ zwyL&$;C)t^+fnTO4W}P!E6f3ULuzyM@>QVah%dmBPI>P>nrlxLd-_IDMR=g&G5$bW z?NfB(L}F`n8g8WTk03xy5hyO2JeRabKluX;rvE3;Ql*{pr1II=Tr? zmW`RIfr{>DZ@~nuTfl*lwB{rmDgfF`_00hu0B-E8vbxWzQ+c}Y1syn zMUqjiQ4N<9;j0FJ|9vT!mCGOEP^$cQleGqL1WM#(EWGLAh2zPG4GlzRk1 zum`(z!=R>fMfr^DrO4pcuj>)w6z%@Gl0NN;NU8#DJJ$n&9-H8FTdr<4*}9xNqbk0o9avVERs@%7Hzg5s}8CIxHR1ty@r$eJAJ6#UEmk@d|(x8=;A`)J+RQ zM{lVbd?@;Pie_jHo=PA+WfCP$Sw(RRk82bGFgs+Z$ z9&WzP%Bt$jGzao-DA`&BS>)nwTk?nD!N3T?o;R*1{ps+tJ-lJA3Q&7h2IxeSIPlSk z^=j@yr+wXRtsl7*NIvpy;~arzLU~L0l)C!fLvn!>mh?Dc#)YQiB17&V^b913;3?YY z=E^v_6t3r-%IEU8_*0Kj!wt$9N@S*Ia456%>^`vMX@qFDdVHaciVNm=j$X4_I(U6V zlXcgU4kE$-c?veQVpXJ))OXj~I9Tyz@K%up+Kvoa5cYp%Tgf#!3n^*4u4l5ZetPI_ zy&D`FR|*t9tZ>%}EZrck@p$Z^=O5_sQ7GHf8`L1IDYIbphrIu6o2IWwnVOhc@Oce? ze$P|X#9*udbIJGdTdDI^^`7;F$8&_Z%LN7+>6FTLDLONkhyqb)q|URW51$cxQZ9d&#IS5C(It=k191=k zi7bzE=h+#kqR6}Zh}tJhj%{piKlJstRM%b%6B1iemW9P@*XBlt3J#^UYAYXnq_#92 zDiL1#V8<>=ADb0H`WQ5!vTx}A0Rjs&Rj82yGA<_sUuw*|K#6iFW6sYIIaL`f^g{XL znS%{|vd!J$tt5_>u1Ucnks>kWAXbHWS;O&b-%qt_n$5g(l>Oo2QEgDLlTEqRx`|)4 zw9L=do6ec$zcZfk zy%*AvqX(9oXU66~!rs@^^`Nxcc1hZvlSGH2NPbynkvCv`$2n;F8 zzC0lUIeaJX;?fJ}ocIAsLiV7Geh(5h+?5bd#jmzcw=6IoddcDWOI&SAm6WXNDLXH2m zt(7u=AMM1qQQNC&4%c=VZlZU623d$8A7DPIxv9Um;1R@241!ra1UbUYai1L}SqtaX zPo_8^ok*O#0wRGhcE!|F(8oIv{J)Fz&vr{;5|9;Hs+e75ovCyvQO z_O)Ana{s1%eE=3TEoi~_L_k(fWJq)szn8OXCz$7v9!GLOdIq;e4(Z`^iTqQ#mt`Gb zS zzb%MkI)GxoglcyfgEKMpO2mhY@yU3NNv}v=Dyl+rR)*BIxV|mL z>imH#l3CHOO6=CbxsV;IXSltbO9a;@f=9g7XASgc z;}AtW>)P10E}jw{9_ALkQa{(Q9Sq#d{6p-Ln1`*5dQY-rG2)#bHTM)ig;=NpObS0T zMn`w-BXn1h)?5CkPRg^ypbHy_4_fLh?=i~C&95b?SvQ{T%Q;18k1Jc%UiDmh& zx_FKvp#rzZnr5J9eVu=|it)Jn@uMp*v;=;v>mn2bM$TW#NrY_OQ4J_Qb_xVnqkhsF z)(HNfTWZ+Prh&5P;t|}N_)Jzw{rLf!x8kR4N!XY|34KVkfHH;hjT07_qld?>h^&aN zoK|@23}fpoEQ?(@ubI6ou5R+fq|l01oW9oN%wFHcgrT()#a7BoW8p*_(8ZeP_Q`fD zq6sZN47EJ_yS!~Tg5>!Sk9hR83&-h?NE|tPvBZxQ2Ab?(?3VO2Z{DFm%5%l@=JM@e z735vh6+Djq;#_m_fZ9;@+u)zLBYXsb#Rx zR<9Tde?`tovWCEW&jlsB&?(hn(rvb>oJH<=4y_~7clGPmv*Yd`tdOEI%qj9HL$>^AsN5BVA)1?OW zrtU~|Jt+pRMaw1n4bOV|2c`wMAO7*GYP?E z{NGubRkAWRf8?%I7iJf5?rwjz$E+whqPwlf-NXG}nK9K(t@NOkTxO;1hspX;awe7L zAlcpNeEv@UkMIVhIRGK!Pv^B8klPeyq7QKkStU9>3(Y4SKKD?u_+@c_*9FOf)+GeO#)C-XN`^-7%zBKq_HSr2^7UB#~`*`mLEx`9lR5IA^;}g?|DOYMu%U7G6fPHNFUp$hWY4_>OZ)V2z&kyTA;NfyBrL8oxPTe2rG^QoOCq0HIBK#e_ z_0beK^*GV7tocMkg5o=$YFT>na7ggPdR)*x^h)*sk0Kq{xnmDlo?;oc)5LoNDOlgW zyR`LNt7XKOc)IB*&z_m(k!x`WO7=xtOYjp64OQK(j`0TWcQBmEA$gSN`+76ac(ljz zjF2RYa*WP$XVBIgCohSm46G*Jc?qos!kR2ysE|1sbj;&aT-uGKeWl4RW}<;5nw<$?S@F#t)8MzIdFSKa(e` z^z`Rr5A3uP7uYi!t*vGRK_`8#)AeLR+i6k^S*EM>+E+CBx*?uIMDz^G?OvVWN?Z2m zD&=dzDUjd(aO_|uyMQ?e_@i7jVe>Fao*H>Ix&lM%V0?ONX!y|=LpsIU9~sZo$71hl z5_T0l8tah-kPROO~+ zgg|P>QGkE0&5W3By7~!tnN}%Ds$!+&TFA{MRnpkA)GTrKbRB*}r7D<|jB5X(lNYO8 zf>)+-VCahgedT%5L*h215}fmm+km-m|K7cCNZ|m$;4%uE8-* zW|Ovut{~sdHn&#bRy-!JP~CfFu&I8ndN6U}q^F@c&sDR@g$NtUjhyjHZjCG-#kth> zD#F;0>MS_hqfnH8sd~<3bJZtc#QhN$Dhz5&ZF>*S$b_`QRjIpseTr+nhtb^2c-F%L zpr`;A_^3}K0$rP_wwvbs6)n$dUkFlH*j*<}{00zjsHu$T9lL zW~0TaubD^NUBr};m|)|6Im`uNkD0;^wYEmJPq&0o7zAG#s=%t57Sk}`=Fv{3yAWGp z&%-@|U(I}oESPOM(2zJ=(RakcyDh=GtJYRn`OLEGnW8QmHraYIj!SM4hbIJD$7lEw{HsEmnw%4pPBHh+m7+o6Gy5)`?UO5oPXPag zXesE2NQeSg3uN+CrdxfI0CBnoY}$D_XU@5S>9#?y`>U?*B+CPtRY%Loy0jRrxb9Oa zM8xQH4mo9+l%bQ8oBqShLRu;d6&rNFnx28QG&Mw#%7=15H6YyVGI>U>E8o=C*9qvm zZ$2Hv^TwH=NJVQ?bY25;mJk@D&gLg0+U#QMJhtpnf_Cbdn!tpXolxcQOJ+jyas^q% zF8FWEJfR&tYJ;fdibr{-VtaL|;j1lFUH(c9*5bQw5obpK{7E9O@?aO`li_w^nCY!IJX_vy0oJn}|$Z>>(gz_Vt zQkiZTRd#*it*dP2s>t0v^io{IMxi~-QZ{G9CJ*}cDeRX}pOnUA`7OjyBKqy!r3ot{ zw24L&Gj}!b(KoJc_>|e^%(*+itW6pUo^+_Ao5@oTR8_xJ4=-OC;Z47mhJd46O>T7R zrTDALDgG|nU5`jl7qv|gbwc!&9x%>=J_UxrRy<~M$aX1qIGLh?k0SQfIcJ|$3)7&R z{?cX?ifNgm%tTwz__*eo9#2O%t+@Z9Z^B&Qi~ZiU9v=jFE0L3PA79A%CH>G_fdg*r zO!C&KYrWT$h3a@LOhGX&{&RHMw7EH7Ibdd`!{CPvF_kp7KuqPzRgalNWaAWU7t{4# zX6eo@VZDal*c15Jor>!m$m&VIsy?~6obyH4E(hnyzfE~)F&BiT$S$v}L;?KX$sjO9 z37$|cY`ZSwh_1f^*9@-XEcsq-M5;hePP6*#=W*D-*i*jGS4;=R&7J4xbH zeePqN5SQ&rXjqM%Lf2ZaDNvM<9@CLzMG|L`$Yjo6uH(1w11xOHmFCu(+}e?u?=RB> zmfVsCI^0ADGJF(j-c{9D{4{@UR*Xvu zXChCwpqZdMVgNz*m^FBZs_hkgPJfFAG&J~s(z+s%Cd z8JpgSYbChY>|Qcxk{jq4ID}cP>Xrn7U741Eqg)S2n4gu=I9=;xx9`OGnZK>b=JUcL z0EXo4%I(a8R)Z;au|607b+iq{uvU<*f)6Z@nr$2(oI*}N-pu(gEPzOlIRNB}QvKRW+p#Tn6YZ}@ z^MZeU2CD4{lVy2RNjBD+Na(@IL~OURMq}`ps^DiRRXIBi!kMtMULc09up(wTI77Onm_~ z*9>qy+vX6mbMWXhg6FUr>f@PDR$s$44Y+xRoKG1WtN2tACJx99Jb_w22 zlX|zFy*_*2Z6dOtr@r489Mt`akukb;DMYb79zxV~UW!pnwKdWQ!FPiP+z z8|XOLh)wM9IRVbf(B;kOKgF+84M@*K`%q^u*!@#*kinYiKUouN9Y|?y$nswSv(1k1 zIk$L>Uo(?^7PA7FRIPRsFU2R?e}AzP;4$?#C(nV69t`r-PXi}ZJrrU$;T-1&0_xs5 z$vxu+HL*j1g##T!H`N@~|MBw~yC4U4U$!9nJP=_bL$lq4po_o&umj-K5%S7;kWU?< zbjaf|VW4E)WA=B-`p0uJ(2%si<{&+Ic|~9b=*uws1(~}6&pwn*bc|Nw&JC)Ak+=3>d7*u)w zW`F%O22R{S-n-`2D?z>|BN5iWh?yt@S}2wis4vD>*^p&<22_*W60xgHZlQKVY@h5V zO!8Lprv4%B{9Zbc8PvGBb`Gq;5(N%4ybm1P9MTQ0P!{tdJa8%vXwAhAtD4+OBiP^N zwf|bg%}=mUd+gvdgEP*|`EbybGNCb@8W#relRaQpbAZ^nn&PzW7aRN|7~z5vX6i04 zt$>UeAD~}yO}f|~Qac2$wLISMh5^@XJ9&XEd>69QOf$CuEO3YVyEDId5#U|HFJZ_s z`eo%DZviiO2N_hi)alsG7HS85{{VMU1IV^ZJpW$ahqVlZDz-S!{&}BRkP!t*^4i?= zYX@de{6jYMLGuWZt(ftky?}!DZ5Z5~3fh;!^g`eu765V7JuyZ$o#Sqh12h3~%`=bm z+qV7_sYP>hULUmu6E`_Kw8Ua+uK`QD0>&p6f#ZTllzk|nlI&=+k$lc!ecn@jlhC}H z!R;-!e>;yK8^sFFO3GU{@dU_n9y<)mB!O+*x%OjV0N>fSF|s{a-VTH!f?f-HBaTQn zg`d>2pTbH6!Ya}wKS~DLQpgi@m7tkh1o#wIGLDHIAZl6jm#k-j*r!TjR3Lsn3UK}eBGCNe9V*^G zdCWhTApjuk(fhB=z)Q64W$+e)9B&OZ`BQ#SI!F%Kx{aDxT|QHiuPZYU+BoCA7_ur4 z9>nI7G5S;Hek>^B0C>>gMZMds;K^nf&MWqsxubm}3E_)pRm6BZp*U=aQ7^6^GHw{N z%sxPW6B->m^xHA>X81$f(nlCS`nWilNnK{)jG5aVB4Av3Lj@X8t;M;)W$_fSv}Qd8 zx7o#>r9Ru{me&)8f4Z6efSNzw%mfNsMO^aQ&obC+at{13(#<=~pj$hOxvN~93AmDX zq}xa*aPY5(j22J~zzNys^~t9Gw#mTK&|fyWWq`(?ny$}^gIIWa3o$KIvI|I^f&UjZ9)F3R-F9s#cF5wN z+t0YhRKv079Eq(u`mAOywp9;2T@Tzc7-p}ax{~boPo#h{b_b#p=E32(OUt|x!Q>X) zKJ@MVJ)y*o$9J%+Q{OB~pp6n63Oug>j9ByGiRM<>!bfadTKLQFbrN7*_Eh`hStw&3 z$bATmr1?S^Wo`z{l@B}$UhT?ER!sikiOev2vuE5-APe`a6h~M?BJ4DI`#NuiBCLO5 zX{wZQ2XOB7vBMyR7Awv}d(&S};wT78|8jU@sYpMaxWldQP`5GXpKMqE#_wAOu7(ir20i~EFv9Zf(-yH%6kG@L)PCqle!zmjmf1VoM@_R;i zap!FxMzbBy|4VeLAnls&8g0`uk4CVj#pD}6-u3=Hq<8-xph{=7&^r(Uoa^U8uDx%8 z@JxP8+u}|9bds9ptx%c%n!_FGELZ>q7@QU8vNv+2UQMnCxKPVXU`Thirw0H{-C3Zc z3Efyz<=?j^)f-At>VLNeZ>I`Dy2rO5TfSYc2aAYC$b*A}T7Zme)&qX{pPS?lMW@SKr9*Drq zI`aGDo9XwxKkfPTz&Aht0XgpX4#wUl3?@21%>+;d0Z?PMTjlRddwqa=b!ydzx1)9g z{PiChfYp8LrS#LSw8TFJ2%1;2U46rZAwO2JCii83qHj0^TGGBA$n>6)G^t+;HPk%_ zKqAf!Bw09~s;6@q|I2j({tg5J3aI(nsEAz*HSdx`A>8dWH#G}P=cYKaNQn&9mkxo( zlS85Q)*?e->V5XTf4x{Kv_r8B5bwM}wJOok{|Z^ME$}ins=F;>$Xfu}EKe5{E`? zNV}Rl2XE)VJaVF(z`{(KUMDrSgTeVju$&0+Yilw zFpFND8>OT5PS4D?efynFF2rAkXY<1~Xmj`|>zOc6w*t!_KA8(Fg<(Aa;Jj|S` zbTqi8#zLrPD=@=1{-%w7>HyMkO}q`rg{^CVg00-ePN!KWB5{SuJ@mHOs~Fip zc8Z+9K$qQwT-I;rqG>JzECPs6zP|d(3p*_|lt#)Ao&Z!lR&CuUq4pcVvdEmCH|X>W z4=k&P8vWXK(XY(s6v>^Q?`8Y+7yt6Foy}N;?83VI*#0rgt=OCD7V&xlr2y%4!Uym& z%lrfNE0v>=!SO~-^RB-97k<_>r$K94b!!u8E8hx0K@$zTLH(KsxR$bVRUXF11+2tq z$SBCH80MKOxE1&{IyC4O7bB53x&O$00VAFUT6{2?<_87$K>PWaUMxw zL1OVB`lH^{1&+z7V=Q2cjQ$TJ!vd+$ZVQK#(8w_2yta@NFyij#>SH6I_JY73Qv|Jd zWCpZz-rVOE+^W_MKz(L%Ufn;80+6`~Gsv+xcHh2`Pxf?`lK>Xit(=EQ*^1Y^p*Z~` z6kQ!U@s2ni4fa@~Wk^eUdZ9>e#=WK1aOfUW}I>K=9!@FK59IqSTE3Q?ea8&G4 z|15T@5V+mMj)MTi&P9gVgK=_2@)`fi34!+jZN57dy%U-iG|hJ)rWlq%Kg5HDZV!*K zfpy>$**Y+$|5~pwJ!Dog(D{g36~$u_;^Lo+W^rx+4dl+>%K6#(_LmXJjs}uF&et{f zxEKkCZ!=sH;>!}UFoO&qvanO(N3rV{XcSs}2I%IpU`U}MgRHKiE~ki_X8z=2@o~@J z_8t83J`dP>jAMH}W0W`&{dyKqr5`4R=5D077^NN`c^0DD_c?z0bB{KA%#{%MSy9&R zh2wY0JJ0?Lrfykls5k9s-1;46hLTXgoIA&BL4!t~^NCWJZIk@;>sO(P=;N~8OZ8Jw zmsswNq=Ht`vnw-S40UBLCZbNK5WqDLO8@=vm(atHN3UPp!%)Vd%*fqFUH<-3D9L~6 z>Uf=G11rQf_zID`UBN9<-bbnflOJBb4Jf_=4p3_l7|p%vdx|b9`1|O;_W-ph2+qUo zVw2-(OpvRTGr|`khr2^Q07j+$VxR#>HTI^Zg%I{B2N6 zr-ZP-wEn*#1*D$efvUf%Px6Rl;%tJiXbvAx0uBFN2Ok=rX!EIV2vPk6?o#}AIh{IP z6LBo6RTrgHJ3iAn>Fy}?6ihnt8UY*m=Qubb_VRe*UIa$#SOmdf_`jr|-|I8v2V!+L z4r9dk3$AxU50jiD8Zz8g<{v6IgZ6WU z`8Vu}C7h-tf+=?PtfT!-9f4mn_CHO|VF()UcDZ#ZTS+tynyAk^ z=J!OI-=cf5Lb@UbC(vB~1`B5V7zoNcxx>Xk7wEuEs83r?j{-kJ^<@W>aC@LO&nXag z1lAVy2WxFtr3*&$5A-LGaxXJ^3aiCY@px>zxJ_ZDi4XG8^<5bC?ix!&gWn@xpm`a? zkV9W)??|Gd4f>~iklJv0?ZZR;voc_J)#!OkqnJJW_xK(_W)7%Z?b>=10xp0$|0QA$4Z8NE#zhpm z6mfSqhfw_Lxxr`iN01oy7vwy&8hXqEe?cbFo46gB^#522whr&_1jv3BQ!gd!k6D@H(aO6XU2HLV@GS&i zAAB4g43{fo?#}<7^0IF2(SHFdCR0)L!&)S0I zDuy9QqQbt)*a?o1u!5SBruswl0tQB$4vlx_JC++s`W0VH{V%o6_z{SYSR8Tmqq*jNRnKA zs%*v^u@>CGuaRWtGX9n}DPL+}*|i{?@Na+$20TEcGgP`<+K)L=c(Cqm_k84ruVg$we^goizKV_Uz=KhVB};fyE!H6iVb~ z(U8%^4W{sjkD?T9jeX-TIM0XA~k8wD!LN8{`x^ zfXR}Q&bM^hu8h}s{&*D!=DOX8jkYax%7X6ZsQZKGU=~gmmiQs-1szZ7DCDPoX(QD^4YFtX&2TFYf8TRHpyZOY=qu+GojS)Acb6hV3K> zdh_o12+mS+uy+Y+fbIYYcJqFju#f}$D=x|TwB)~-USRSH-p2p*iZew;xsI#=1v3~w zSCGN>2EO*TRQJQ7U<%y@S1@E^ZE-nBY+m7aj0;&qbI2OD?KL6wpgj85Ey^Ml)@gyI zAefa7yEo)BQ){+vxA-~T{Rq(iJyu;DPtC4plUAK`K06!4uY>t7Ed3i6IN!}Cx5qU` z+W?8?E5&_zq{nLqI}2bjOpZL1e0sIy{r{x;(#clXD*l)C{a>7hpL(l?Bn(a0igCEl zM0fi;K*l$*)4Dd#_VybhE3d289kCCGj}R2AcSxCB)c!-9_d7HYKw_WvFv}zLhHb-u z;KHiowD@1n*Z2Ztd7UP0`r4AgpcQqb@o6mvPz13JJoVP7CD__K`Qdq4X9kdmS{rR7 z(!FZoxuRPxA(tKl8bC)fPF3>Q=+lXZ+EZymPM{X^Wzm$-%lcTJW;gD!TSQZ6&t zPM97xVhqXocL4>!(_i%K?(WGymHjABN)l8@=D&P5g`iV}(b)htjYu=lLCdMz2u zL793Jw|0GZ<~bWQTxt?@7IOMX^_6q``>D?qOu9`5r6ZrYQk>*$D#qnX82be9`j#+3 ztI!R#w}%l4R+<&H9Jhhd+Z$~3OKJBn*jvm6;lnBy+e4MQc#ZqQa*f#P==*1N+Ow#6 z-_K1?!*?~C*;Vsyj_&vhT|Ke(6_f6xP4xynzKMZ0K3ajcNsEa;8Fa7-5)plu9ceDS(-CPhRfV*7ru3(uw*c1_B#OYeGcxc3KVx{FIA zh%*vaaXaM<8K1lm56y&zsPwVij|5GJ?~q>qlgQ_kTVZ3SLM`wtORm5)O!X{<@KjF^ z!W%8Su*=U=6K7XIsU(-{8Iuu@7eR-vcP@eC>JP_1Ycz3yVY?Jx9#P&KnApGOL$@!} zI#0Vje8EkOUDbIsq>p^i@xLIh@L_=2{iqAs#4)>m^>3>XPIn?gJbR^)+j8j~Jq9qg zyNGmFjgW(lL*P2|KgTmG*-ZpqiKf;#o<@6;B@hkM#|b##eqQo=Q7k$8)wjo9HG{>$ z<-#Ie6`eP%*Ke@FUP1U!j-D)yZr~&QJ2wv*o1d<&H}hqQ%BlruY4-rsGApuzauS&j z78TKe_G;5Ino&ntb1@(iLz187SzYNKLyCu4YKhS*r8X~l(Vz#YSxPT?@?@MaXch^{4Q5_R8$|r zIuVHs$lJt?;rkYeSHqaZFZNiK?IRq~u8$?o4fEdAjlX!R0N_yDS)3!+N1`4Dqq>S& zo~p9A=G4#R>AG;UHfxNyedN>z*Xb`uEpDKU+qbVlGc|6++29PCM%dbp{5y^Y5_p{q zJno*JucjP)UGQz(p;XWxSPGh@-da*EiBlO69#eUTd0l3Or5q1dRsa`QCD1aEhgFSb zz*S|cN)K2{lXQ9qhPv=Swdie@IJ*<~si~sQ~`y1eMFW7C!C`2xKX_ruKXvk{pB%Y48r;&HbtBZHTx~ zuRiMNOxvR!{HptE7W$FC#}WdIPpr}kuCTOR=N0YMJqP2Fv%nH?%Y6VOFx5KF;^W|k zq$k8}Blok{KoG$O-5+&Uz&`!cuMGFUoUZVnkc?(mX5XSt4>38g*ggUI!I|ktA!@HL z#P99lWtT3=4z;nC$9m}dSM^Q8c)Dji;%kgnG4t>l&3X=eW5j5 z7+R_7c#;mW9pH7`pg{0TGM33ptKrQ^CGc=JK|+WT)X(PLu@cn7&LA&qX8wm;-tPrr zqBbpADODLK2>MD;{W4IGR2r9A-Dea;ve{Zbh8ha*r5&g!0 z)lD#Oas8A1FqnSyC)5Jpn2POEbLK)f^3_t}L+Q?4KBVNw#Pj|R?r}dhZ?45;bKKNE z&)`6H1tA`t@GpP!!~XW#;(@?ff9NAbhDadKS>kT=FZgisbO?1R;bV!b!-!+BXp$N7 znyXoZ3HEN#g+}Ua7K{7(Xqxq(g-4h@o=EVG?wd~Z7}1)Hun9J* zy@iLP_Rd4Mmw|N5#uUiQ1T6X7$Qw!Wf*K|I=Brgg$D>7)|6`Lh&AUJ}5w##Eh3(_c zV>(YhwoLLy-d&9&hkShb<>spqzC~9-l0bR=32cQH_NnVSKX&cP$jE86W_a^JjA&nBny=pg-TT;j0RH6O#~TmO+# z0~!2XKze6hWA~f0B6j$(;1?~{IT~U~H(q7S$aCm%rMroR1!c}&6X72CdjF~1+HCVC zZ17bqi-E?Hl#^`0g&DlFTA#&aJSm_8ArY#5ERyG=0sw!i`QW0s%JX=Vpo~uo>fPx> z_TB%bzOhhl3OnN18O^15Y#EHnt8`T+JvmE1oHv?w$Wrr@8XZ>`(@Uv>kLgBOZn{k9 za>20yrl`~O+}9;!eBnm_8H%Vw;IOIdyRjOBaxqFZDQ=Js)$|sSfSAFknaZFk5pa)B z8(cBPqhR(T!)6yPh6j1CH}P3I-KTAt|E#h?5$F%R+d@k-q7OMi zM&2aZNGR8XB#L|tG!RQn;s{OcF<2Opv7E=<9bj5DdX0%EuDTvzEYxl&`Y4YBv&C%p zxdntqKhOVH8V~@k!N)*(ojN7iTBBgZsWM0hd!S6siTC8=IJe+*?}!sHgG6n9V?4JL zndBh{%3*yA{o8zpm?q=EkSs^xKcJa=j3rfB&g-eUTFaIHHkmR|z!>1D>zysn(0pE= zu&cTERl=Eb6wb2#Nezapa*b*%5g8KI&mOx6(zsCad_GdN{}CIqt*9$14?`{2Qt8PL z!-MwQKJBMNZ4V2c1&oaDiVj7jE7vlQW6;OU{|VsU8RPbXZpP&7Wx8!_w_4|mTi=gS zZ{X(dQW`-Y2Qg{4;o8gRzZVyNNei@QdfW%Y?kF4>g7)Rsjn=N%XUBH`p|p>>@hbW% zlj}6R0mF3KDg0zUzSK={y?&*p)uqGl>10}*?BrJ^Pda)A7@Ubyaq~k8!_Ft)FL_9{ zGAmB9DLPtzbZc3w&)+sIH_@0TYMR0qvQTub;#^g+{ (<%= grunt.template.today("yyyy-mm-dd, HH:MM") %>)\n' + + ' * http://lab.hakim.se/reveal-js\n' + + ' * MIT licensed\n' + + ' *\n' + + ' * Copyright (C) 2014 Hakim El Hattab, http://hakim.se\n' + + ' */' + }, + + qunit: { + files: [ 'test/*.html' ] + }, + + uglify: { + options: { + banner: '<%= meta.banner %>\n' + }, + build: { + src: 'js/reveal.js', + dest: 'js/reveal.min.js' + } + }, + + cssmin: { + compress: { + files: { + 'css/reveal.min.css': [ 'css/reveal.css' ] + } + } + }, + + sass: { + main: { + files: { + 'css/theme/darkgray.css': 'css/theme/source/darkgray.scss', + 'css/theme/beigesmall.css': 'css/theme/source/beigesmall.scss', + 'css/theme/cbc.css': 'css/theme/source/cbc.scss', + 'css/theme/default.css': 'css/theme/source/default.scss', + 'css/theme/beige.css': 'css/theme/source/beige.scss', + 'css/theme/night.css': 'css/theme/source/night.scss', + 'css/theme/serif.css': 'css/theme/source/serif.scss', + 'css/theme/simple.css': 'css/theme/source/simple.scss', + 'css/theme/sky.css': 'css/theme/source/sky.scss', + 'css/theme/moon.css': 'css/theme/source/moon.scss', + 'css/theme/solarized.css': 'css/theme/source/solarized.scss', + 'css/theme/blood.css': 'css/theme/source/blood.scss' + } + } + }, + + jshint: { + options: { + curly: false, + eqeqeq: true, + immed: true, + latedef: true, + newcap: true, + noarg: true, + sub: true, + undef: true, + eqnull: true, + browser: true, + expr: true, + globals: { + head: false, + module: false, + console: false, + unescape: false + } + }, + files: [ 'Gruntfile.js', 'js/reveal.js' ] + }, + + connect: { + server: { + options: { + port: port, + base: '.' + } + } + }, + + zip: { + 'reveal-js-presentation.zip': [ + 'index.html', + 'css/**', + 'js/**', + 'lib/**', + 'images/**', + 'plugin/**' + ] + }, + + watch: { + main: { + files: [ 'Gruntfile.js', 'js/reveal.js', 'css/reveal.css' ], + tasks: 'default' + }, + theme: { + files: [ 'css/theme/source/*.scss', 'css/theme/template/*.scss' ], + tasks: 'themes' + } + } + + }); + + // Dependencies + grunt.loadNpmTasks( 'grunt-contrib-qunit' ); + grunt.loadNpmTasks( 'grunt-contrib-jshint' ); + grunt.loadNpmTasks( 'grunt-contrib-cssmin' ); + grunt.loadNpmTasks( 'grunt-contrib-uglify' ); + grunt.loadNpmTasks( 'grunt-contrib-watch' ); + grunt.loadNpmTasks( 'grunt-contrib-sass' ); + grunt.loadNpmTasks( 'grunt-contrib-connect' ); + grunt.loadNpmTasks( 'grunt-zip' ); + + // Default task + grunt.registerTask( 'default', [ 'jshint', 'cssmin', 'uglify', 'qunit' ] ); + + // Theme task + grunt.registerTask( 'themes', [ 'sass' ] ); + + // Package presentation to archive + grunt.registerTask( 'package', [ 'default', 'zip' ] ); + + // Serve presentation locally + grunt.registerTask( 'serve', [ 'connect', 'watch' ] ); + + // Run tests + grunt.registerTask( 'test', [ 'jshint', 'qunit' ] ); + +}; diff --git a/doc/pub/BM/html/reveal.js/LICENSE b/doc/pub/BM/html/reveal.js/LICENSE new file mode 100644 index 000000000..c3e6e5fd6 --- /dev/null +++ b/doc/pub/BM/html/reveal.js/LICENSE @@ -0,0 +1,19 @@ +Copyright (C) 2017 Hakim El Hattab, http://hakim.se, and reveal.js contributors + +Permission is hereby granted, free of charge, to any person obtaining a copy +of this software and associated documentation files (the "Software"), to deal +in the Software without restriction, including without limitation the rights +to use, copy, modify, merge, publish, distribute, sublicense, and/or sell +copies of the Software, and to permit persons to whom the Software is +furnished to do so, subject to the following conditions: + +The above copyright notice and this permission notice shall be included in +all copies or substantial portions of the Software. + +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE +AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER +LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, +OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN +THE SOFTWARE. \ No newline at end of file diff --git a/doc/pub/BM/html/reveal.js/README.md b/doc/pub/BM/html/reveal.js/README.md new file mode 100644 index 000000000..f2ab6ca88 --- /dev/null +++ b/doc/pub/BM/html/reveal.js/README.md @@ -0,0 +1,1246 @@ +# reveal.js [![Build Status](https://travis-ci.org/hakimel/reveal.js.svg?branch=master)](https://travis-ci.org/hakimel/reveal.js) Slides + +A framework for easily creating beautiful presentations using HTML. [Check out the live demo](http://revealjs.com/). + +reveal.js comes with a broad range of features including [nested slides](https://github.com/hakimel/reveal.js#markup), [Markdown contents](https://github.com/hakimel/reveal.js#markdown), [PDF export](https://github.com/hakimel/reveal.js#pdf-export), [speaker notes](https://github.com/hakimel/reveal.js#speaker-notes) and a [JavaScript API](https://github.com/hakimel/reveal.js#api). There's also a fully featured visual editor and platform for sharing reveal.js presentations at [slides.com](https://slides.com?ref=github). + +## Table of contents +- [Online Editor](#online-editor) +- [Instructions](#instructions) + - [Markup](#markup) + - [Markdown](#markdown) + - [Element Attributes](#element-attributes) + - [Slide Attributes](#slide-attributes) +- [Configuration](#configuration) +- [Presentation Size](#presentation-size) +- [Dependencies](#dependencies) +- [Ready Event](#ready-event) +- [Auto-sliding](#auto-sliding) +- [Keyboard Bindings](#keyboard-bindings) +- [Touch Navigation](#touch-navigation) +- [Lazy Loading](#lazy-loading) +- [API](#api) + - [Slide Changed Event](#slide-changed-event) + - [Presentation State](#presentation-state) + - [Slide States](#slide-states) + - [Slide Backgrounds](#slide-backgrounds) + - [Parallax Background](#parallax-background) + - [Slide Transitions](#slide-transitions) + - [Internal links](#internal-links) + - [Fragments](#fragments) + - [Fragment events](#fragment-events) + - [Code syntax highlighting](#code-syntax-highlighting) + - [Slide number](#slide-number) + - [Overview mode](#overview-mode) + - [Fullscreen mode](#fullscreen-mode) + - [Embedded media](#embedded-media) + - [Stretching elements](#stretching-elements) + - [postMessage API](#postmessage-api) +- [PDF Export](#pdf-export) +- [Theming](#theming) +- [Speaker Notes](#speaker-notes) + - [Share and Print Speaker Notes](#share-and-print-speaker-notes) + - [Server Side Speaker Notes](#server-side-speaker-notes) +- [Multiplexing](#multiplexing) + - [Master presentation](#master-presentation) + - [Client presentation](#client-presentation) + - [Socket.io server](#socketio-server) +- [MathJax](#mathjax) +- [Installation](#installation) + - [Basic setup](#basic-setup) + - [Full setup](#full-setup) + - [Folder Structure](#folder-structure) +- [License](#license) + +#### More reading +- [Changelog](https://github.com/hakimel/reveal.js/releases): Up-to-date version history. +- [Examples](https://github.com/hakimel/reveal.js/wiki/Example-Presentations): Presentations created with reveal.js, add your own! +- [Browser Support](https://github.com/hakimel/reveal.js/wiki/Browser-Support): Explanation of browser support and fallbacks. +- [Plugins](https://github.com/hakimel/reveal.js/wiki/Plugins,-Tools-and-Hardware): A list of plugins that can be used to extend reveal.js. + +## Online Editor + +Presentations are written using HTML or Markdown but there's also an online editor for those of you who prefer a graphical interface. Give it a try at [https://slides.com](https://slides.com?ref=github). + + +## Instructions + +### Markup + +Here's a barebones example of a fully working reveal.js presentation: +```html + + + + + + +

    +
    +
    Slide 1
    +
    Slide 2
    +
    +
    + + + + +``` + +The presentation markup hierarchy needs to be `.reveal > .slides > section` where the `section` represents one slide and can be repeated indefinitely. If you place multiple `section` elements inside of another `section` they will be shown as vertical slides. The first of the vertical slides is the "root" of the others (at the top), and will be included in the horizontal sequence. For example: + +```html +
    +
    +
    Single Horizontal Slide
    +
    +
    Vertical Slide 1
    +
    Vertical Slide 2
    +
    +
    +
    +``` + +### Markdown + +It's possible to write your slides using Markdown. To enable Markdown, add the `data-markdown` attribute to your `
    ` elements and wrap the contents in a ` +
    +``` + +#### External Markdown + +You can write your content as a separate file and have reveal.js load it at runtime. Note the separator arguments which determine how slides are delimited in the external file: the `data-separator` attribute defines a regular expression for horizontal slides (defaults to `^\r?\n---\r?\n$`, a newline-bounded horizontal rule) and `data-separator-vertical` defines vertical slides (disabled by default). The `data-separator-notes` attribute is a regular expression for specifying the beginning of the current slide's speaker notes (defaults to `note:`). The `data-charset` attribute is optional and specifies which charset to use when loading the external file. + +When used locally, this feature requires that reveal.js [runs from a local web server](#full-setup). The following example customises all available options: + +```html +
    + +
    +``` + +#### Element Attributes + +Special syntax (in html comment) is available for adding attributes to Markdown elements. This is useful for fragments, amongst other things. + +```html +
    + +
    +``` + +#### Slide Attributes + +Special syntax (in html comment) is available for adding attributes to the slide `
    ` elements generated by your Markdown. + +```html +
    + +
    +``` + +#### Configuring *marked* + +We use [marked](https://github.com/chjj/marked) to parse Markdown. To customise marked's rendering, you can pass in options when [configuring Reveal](#configuration): + +```javascript +Reveal.initialize({ + // Options which are passed into marked + // See https://github.com/chjj/marked#options-1 + markdown: { + smartypants: true + } +}); +``` + +### Configuration + +At the end of your page you need to initialize reveal by running the following code. Note that all config values are optional and will default as specified below. + +```javascript +Reveal.initialize({ + + // Display presentation control arrows + controls: true, + + // Help the user learn the controls by providing hints, for example by + // bouncing the down arrow when they first encounter a vertical slide + controlsTutorial: true, + + // Determines where controls appear, "edges" or "bottom-right" + controlsLayout: 'bottom-right', + + // Visibility rule for backwards navigation arrows; "faded", "hidden" + // or "visible" + controlsBackArrows: 'faded', + + // Display a presentation progress bar + progress: true, + + // Set default timing of 2 minutes per slide + defaultTiming: 120, + + // Display the page number of the current slide + slideNumber: false, + + // Push each slide change to the browser history + history: false, + + // Enable keyboard shortcuts for navigation + keyboard: true, + + // Enable the slide overview mode + overview: true, + + // Vertical centering of slides + center: true, + + // Enables touch navigation on devices with touch input + touch: true, + + // Loop the presentation + loop: false, + + // Change the presentation direction to be RTL + rtl: false, + + // Randomizes the order of slides each time the presentation loads + shuffle: false, + + // Turns fragments on and off globally + fragments: true, + + // Flags if the presentation is running in an embedded mode, + // i.e. contained within a limited portion of the screen + embedded: false, + + // Flags if we should show a help overlay when the questionmark + // key is pressed + help: true, + + // Flags if speaker notes should be visible to all viewers + showNotes: false, + + // Global override for autoplaying embedded media (video/audio/iframe) + // - null: Media will only autoplay if data-autoplay is present + // - true: All media will autoplay, regardless of individual setting + // - false: No media will autoplay, regardless of individual setting + autoPlayMedia: null, + + // Number of milliseconds between automatically proceeding to the + // next slide, disabled when set to 0, this value can be overwritten + // by using a data-autoslide attribute on your slides + autoSlide: 0, + + // Stop auto-sliding after user input + autoSlideStoppable: true, + + // Use this method for navigation when auto-sliding + autoSlideMethod: Reveal.navigateNext, + + // Enable slide navigation via mouse wheel + mouseWheel: false, + + // Hides the address bar on mobile devices + hideAddressBar: true, + + // Opens links in an iframe preview overlay + // Add `data-preview-link` and `data-preview-link="false"` to customise each link + // individually + previewLinks: false, + + // Transition style + transition: 'slide', // none/fade/slide/convex/concave/zoom + + // Transition speed + transitionSpeed: 'default', // default/fast/slow + + // Transition style for full page slide backgrounds + backgroundTransition: 'fade', // none/fade/slide/convex/concave/zoom + + // Number of slides away from the current that are visible + viewDistance: 3, + + // Parallax background image + parallaxBackgroundImage: '', // e.g. "'https://s3.amazonaws.com/hakim-static/reveal-js/reveal-parallax-1.jpg'" + + // Parallax background size + parallaxBackgroundSize: '', // CSS syntax, e.g. "2100px 900px" + + // Number of pixels to move the parallax background per slide + // - Calculated automatically unless specified + // - Set to 0 to disable movement along an axis + parallaxBackgroundHorizontal: null, + parallaxBackgroundVertical: null, + + // The display mode that will be used to show slides + display: 'block' + +}); +``` + + +The configuration can be updated after initialization using the ```configure``` method: + +```javascript +// Turn autoSlide off +Reveal.configure({ autoSlide: 0 }); + +// Start auto-sliding every 5s +Reveal.configure({ autoSlide: 5000 }); +``` + + +### Presentation Size + +All presentations have a normal size, that is the resolution at which they are authored. The framework will automatically scale presentations uniformly based on this size to ensure that everything fits on any given display or viewport. + +See below for a list of configuration options related to sizing, including default values: + +```javascript +Reveal.initialize({ + + ... + + // The "normal" size of the presentation, aspect ratio will be preserved + // when the presentation is scaled to fit different resolutions. Can be + // specified using percentage units. + width: 960, + height: 700, + + // Factor of the display size that should remain empty around the content + margin: 0.1, + + // Bounds for smallest/largest possible scale to apply to content + minScale: 0.2, + maxScale: 1.5 + +}); +``` + +If you wish to disable this behavior and do your own scaling (e.g. using media queries), try these settings: + +```javascript +Reveal.initialize({ + + ... + + width: "100%", + height: "100%", + margin: 0, + minScale: 1, + maxScale: 1 +}); +``` + +### Dependencies + +Reveal.js doesn't _rely_ on any third party scripts to work but a few optional libraries are included by default. These libraries are loaded as dependencies in the order they appear, for example: + +```javascript +Reveal.initialize({ + dependencies: [ + // Cross-browser shim that fully implements classList - https://github.com/eligrey/classList.js/ + { src: 'lib/js/classList.js', condition: function() { return !document.body.classList; } }, + + // Interpret Markdown in
    elements + { src: 'plugin/markdown/marked.js', condition: function() { return !!document.querySelector( '[data-markdown]' ); } }, + { src: 'plugin/markdown/markdown.js', condition: function() { return !!document.querySelector( '[data-markdown]' ); } }, + + // Syntax highlight for elements + { src: 'plugin/highlight/highlight.js', async: true, callback: function() { hljs.initHighlightingOnLoad(); } }, + + // Zoom in and out with Alt+click + { src: 'plugin/zoom-js/zoom.js', async: true }, + + // Speaker notes + { src: 'plugin/notes/notes.js', async: true }, + + // MathJax + { src: 'plugin/math/math.js', async: true } + ] +}); +``` + +You can add your own extensions using the same syntax. The following properties are available for each dependency object: +- **src**: Path to the script to load +- **async**: [optional] Flags if the script should load after reveal.js has started, defaults to false +- **callback**: [optional] Function to execute when the script has loaded +- **condition**: [optional] Function which must return true for the script to be loaded + +To load these dependencies, reveal.js requires [head.js](http://headjs.com/) *(a script loading library)* to be loaded before reveal.js. + +### Ready Event + +A 'ready' event is fired when reveal.js has loaded all non-async dependencies and is ready to start navigating. To check if reveal.js is already 'ready' you can call `Reveal.isReady()`. + +```javascript +Reveal.addEventListener( 'ready', function( event ) { + // event.currentSlide, event.indexh, event.indexv +} ); +``` + +Note that we also add a `.ready` class to the `.reveal` element so that you can hook into this with CSS. + +### Auto-sliding + +Presentations can be configured to progress through slides automatically, without any user input. To enable this you will need to tell the framework how many milliseconds it should wait between slides: + +```javascript +// Slide every five seconds +Reveal.configure({ + autoSlide: 5000 +}); +``` +When this is turned on a control element will appear that enables users to pause and resume auto-sliding. Alternatively, sliding can be paused or resumed by pressing »a« on the keyboard. Sliding is paused automatically as soon as the user starts navigating. You can disable these controls by specifying ```autoSlideStoppable: false``` in your reveal.js config. + +You can also override the slide duration for individual slides and fragments by using the ```data-autoslide``` attribute: + +```html +
    +

    After 2 seconds the first fragment will be shown.

    +

    After 10 seconds the next fragment will be shown.

    +

    Now, the fragment is displayed for 2 seconds before the next slide is shown.

    +
    +``` + +To override the method used for navigation when auto-sliding, you can specify the ```autoSlideMethod``` setting. To only navigate along the top layer and ignore vertical slides, set this to ```Reveal.navigateRight```. + +Whenever the auto-slide mode is resumed or paused the ```autoslideresumed``` and ```autoslidepaused``` events are fired. + + +### Keyboard Bindings + +If you're unhappy with any of the default keyboard bindings you can override them using the ```keyboard``` config option: + +```javascript +Reveal.configure({ + keyboard: { + 13: 'next', // go to the next slide when the ENTER key is pressed + 27: function() {}, // do something custom when ESC is pressed + 32: null // don't do anything when SPACE is pressed (i.e. disable a reveal.js default binding) + } +}); +``` + +### Touch Navigation + +You can swipe to navigate through a presentation on any touch-enabled device. Horizontal swipes change between horizontal slides, vertical swipes change between vertical slides. If you wish to disable this you can set the `touch` config option to false when initializing reveal.js. + +If there's some part of your content that needs to remain accessible to touch events you'll need to highlight this by adding a `data-prevent-swipe` attribute to the element. One common example where this is useful is elements that need to be scrolled. + + +### Lazy Loading + +When working on presentation with a lot of media or iframe content it's important to load lazily. Lazy loading means that reveal.js will only load content for the few slides nearest to the current slide. The number of slides that are preloaded is determined by the `viewDistance` configuration option. + +To enable lazy loading all you need to do is change your "src" attributes to "data-src" as shown below. This is supported for image, video, audio and iframe elements. Lazy loaded iframes will also unload when the containing slide is no longer visible. + +```html +
    + + + +
    +``` + + +### API + +The ``Reveal`` object exposes a JavaScript API for controlling navigation and reading state: + +```javascript +// Navigation +Reveal.slide( indexh, indexv, indexf ); +Reveal.left(); +Reveal.right(); +Reveal.up(); +Reveal.down(); +Reveal.prev(); +Reveal.next(); +Reveal.prevFragment(); +Reveal.nextFragment(); + +// Randomize the order of slides +Reveal.shuffle(); + +// Toggle presentation states, optionally pass true/false to force on/off +Reveal.toggleOverview(); +Reveal.togglePause(); +Reveal.toggleAutoSlide(); + +// Shows a help overlay with keyboard shortcuts, optionally pass true/false +// to force on/off +Reveal.toggleHelp(); + +// Change a config value at runtime +Reveal.configure({ controls: true }); + +// Returns the present configuration options +Reveal.getConfig(); + +// Fetch the current scale of the presentation +Reveal.getScale(); + +// Retrieves the previous and current slide elements +Reveal.getPreviousSlide(); +Reveal.getCurrentSlide(); + +Reveal.getIndices(); // { h: 0, v: 0 } } +Reveal.getPastSlideCount(); +Reveal.getProgress(); // (0 == first slide, 1 == last slide) +Reveal.getSlides(); // Array of all slides +Reveal.getTotalSlides(); // total number of slides + +// Returns the speaker notes for the current slide +Reveal.getSlideNotes(); + +// State checks +Reveal.isFirstSlide(); +Reveal.isLastSlide(); +Reveal.isOverview(); +Reveal.isPaused(); +Reveal.isAutoSliding(); +``` + +### Slide Changed Event + +A 'slidechanged' event is fired each time the slide is changed (regardless of state). The event object holds the index values of the current slide as well as a reference to the previous and current slide HTML nodes. + +Some libraries, like MathJax (see [#226](https://github.com/hakimel/reveal.js/issues/226#issuecomment-10261609)), get confused by the transforms and display states of slides. Often times, this can be fixed by calling their update or render function from this callback. + +```javascript +Reveal.addEventListener( 'slidechanged', function( event ) { + // event.previousSlide, event.currentSlide, event.indexh, event.indexv +} ); +``` + +### Presentation State + +The presentation's current state can be fetched by using the `getState` method. A state object contains all of the information required to put the presentation back as it was when `getState` was first called. Sort of like a snapshot. It's a simple object that can easily be stringified and persisted or sent over the wire. + +```javascript +Reveal.slide( 1 ); +// we're on slide 1 + +var state = Reveal.getState(); + +Reveal.slide( 3 ); +// we're on slide 3 + +Reveal.setState( state ); +// we're back on slide 1 +``` + +### Slide States + +If you set ``data-state="somestate"`` on a slide ``
    ``, "somestate" will be applied as a class on the document element when that slide is opened. This allows you to apply broad style changes to the page based on the active slide. + +Furthermore you can also listen to these changes in state via JavaScript: + +```javascript +Reveal.addEventListener( 'somestate', function() { + // TODO: Sprinkle magic +}, false ); +``` + +### Slide Backgrounds + +Slides are contained within a limited portion of the screen by default to allow them to fit any display and scale uniformly. You can apply full page backgrounds outside of the slide area by adding a ```data-background``` attribute to your ```
    ``` elements. Four different types of backgrounds are supported: color, image, video and iframe. + +#### Color Backgrounds +All CSS color formats are supported, like rgba() or hsl(). +```html +
    +

    Color

    +
    +``` + +#### Image Backgrounds +By default, background images are resized to cover the full page. Available options: + +| Attribute | Default | Description | +| :--------------------------- | :--------- | :---------- | +| data-background-image | | URL of the image to show. GIFs restart when the slide opens. | +| data-background-size | cover | See [background-size](https://developer.mozilla.org/docs/Web/CSS/background-size) on MDN. | +| data-background-position | center | See [background-position](https://developer.mozilla.org/docs/Web/CSS/background-position) on MDN. | +| data-background-repeat | no-repeat | See [background-repeat](https://developer.mozilla.org/docs/Web/CSS/background-repeat) on MDN. | +```html +
    +

    Image

    +
    +
    +

    This background image will be sized to 100px and repeated

    +
    +``` + +#### Video Backgrounds +Automatically plays a full size video behind the slide. + +| Attribute | Default | Description | +| :--------------------------- | :------ | :---------- | +| data-background-video | | A single video source, or a comma separated list of video sources. | +| data-background-video-loop | false | Flags if the video should play repeatedly. | +| data-background-video-muted | false | Flags if the audio should be muted. | +| data-background-size | cover | Use `cover` for full screen and some cropping or `contain` for letterboxing. | + +```html +
    +

    Video

    +
    +``` + +#### Iframe Backgrounds +Embeds a web page as a slide background that covers 100% of the reveal.js width and height. The iframe is in the background layer, behind your slides, and as such it's not possible to interact with it by default. To make your background interactive, you can add the `data-background-interactive` attribute. +```html +
    +

    Iframe

    +
    +``` + +#### Background Transitions +Backgrounds transition using a fade animation by default. This can be changed to a linear sliding transition by passing ```backgroundTransition: 'slide'``` to the ```Reveal.initialize()``` call. Alternatively you can set ```data-background-transition``` on any section with a background to override that specific transition. + + +### Parallax Background + +If you want to use a parallax scrolling background, set the first two config properties below when initializing reveal.js (the other two are optional). + +```javascript +Reveal.initialize({ + + // Parallax background image + parallaxBackgroundImage: '', // e.g. "https://s3.amazonaws.com/hakim-static/reveal-js/reveal-parallax-1.jpg" + + // Parallax background size + parallaxBackgroundSize: '', // CSS syntax, e.g. "2100px 900px" - currently only pixels are supported (don't use % or auto) + + // Number of pixels to move the parallax background per slide + // - Calculated automatically unless specified + // - Set to 0 to disable movement along an axis + parallaxBackgroundHorizontal: 200, + parallaxBackgroundVertical: 50 + +}); +``` + +Make sure that the background size is much bigger than screen size to allow for some scrolling. [View example](http://revealjs.com/?parallaxBackgroundImage=https%3A%2F%2Fs3.amazonaws.com%2Fhakim-static%2Freveal-js%2Freveal-parallax-1.jpg¶llaxBackgroundSize=2100px%20900px). + + + +### Slide Transitions +The global presentation transition is set using the ```transition``` config value. You can override the global transition for a specific slide by using the ```data-transition``` attribute: + +```html +
    +

    This slide will override the presentation transition and zoom!

    +
    + +
    +

    Choose from three transition speeds: default, fast or slow!

    +
    +``` + +You can also use different in and out transitions for the same slide: + +```html +
    + The train goes on … +
    +
    + and on … +
    +
    + and stops. +
    +
    + (Passengers entering and leaving) +
    +
    + And it starts again. +
    +``` + + +### Internal links + +It's easy to link between slides. The first example below targets the index of another slide whereas the second targets a slide with an ID attribute (```
    ```): + +```html +Link +Link +``` + +You can also add relative navigation links, similar to the built in reveal.js controls, by appending one of the following classes on any element. Note that each element is automatically given an ```enabled``` class when it's a valid navigation route based on the current slide. + +```html + + + + + + +``` + + +### Fragments +Fragments are used to highlight individual elements on a slide. Every element with the class ```fragment``` will be stepped through before moving on to the next slide. Here's an example: http://revealjs.com/#/fragments + +The default fragment style is to start out invisible and fade in. This style can be changed by appending a different class to the fragment: + +```html +
    +

    grow

    +

    shrink

    +

    fade-out

    +

    fade-up (also down, left and right!)

    +

    visible only once

    +

    blue only once

    +

    highlight-red

    +

    highlight-green

    +

    highlight-blue

    +
    +``` + +Multiple fragments can be applied to the same element sequentially by wrapping it, this will fade in the text on the first step and fade it back out on the second. + +```html +
    + + I'll fade in, then out + +
    +``` + +The display order of fragments can be controlled using the ```data-fragment-index``` attribute. + +```html +
    +

    Appears last

    +

    Appears first

    +

    Appears second

    +
    +``` + +### Fragment events + +When a slide fragment is either shown or hidden reveal.js will dispatch an event. + +Some libraries, like MathJax (see #505), get confused by the initially hidden fragment elements. Often times this can be fixed by calling their update or render function from this callback. + +```javascript +Reveal.addEventListener( 'fragmentshown', function( event ) { + // event.fragment = the fragment DOM element +} ); +Reveal.addEventListener( 'fragmenthidden', function( event ) { + // event.fragment = the fragment DOM element +} ); +``` + +### Code syntax highlighting + +By default, Reveal is configured with [highlight.js](https://highlightjs.org/) for code syntax highlighting. To enable syntax highlighting, you'll have to load the highlight plugin ([plugin/highlight/highlight.js](plugin/highlight/highlight.js)) and a highlight.js CSS theme (Reveal comes packaged with the zenburn theme: [lib/css/zenburn.css](lib/css/zenburn.css)). + +Below is an example with clojure code that will be syntax highlighted. When the `data-trim` attribute is present, surrounding whitespace is automatically removed. HTML will be escaped by default. To avoid this, for example if you are using `` to call out a line of code, add the `data-noescape` attribute to the `` element. + +```html +
    + 
    
+(def lazy-fib
+  (concat
+   [0 1]
+   ((fn rfib [a b]
+        (lazy-cons (+ a b) (rfib b (+ a b)))) 0 1)))
+
    +
    +``` + +### Slide number +If you would like to display the page number of the current slide you can do so using the ```slideNumber``` and ```showSlideNumber``` configuration values. + +```javascript +// Shows the slide number using default formatting +Reveal.configure({ slideNumber: true }); + +// Slide number formatting can be configured using these variables: +// "h.v": horizontal . vertical slide number (default) +// "h/v": horizontal / vertical slide number +// "c": flattened slide number +// "c/t": flattened slide number / total slides +Reveal.configure({ slideNumber: 'c/t' }); + +// Control which views the slide number displays on using the "showSlideNumber" value: +// "all": show on all views (default) +// "speaker": only show slide numbers on speaker notes view +// "print": only show slide numbers when printing to PDF +Reveal.configure({ showSlideNumber: 'speaker' }); + +``` + + +### Overview mode + +Press "Esc" or "o" keys to toggle the overview mode on and off. While you're in this mode, you can still navigate between slides, +as if you were at 1,000 feet above your presentation. The overview mode comes with a few API hooks: + +```javascript +Reveal.addEventListener( 'overviewshown', function( event ) { /* ... */ } ); +Reveal.addEventListener( 'overviewhidden', function( event ) { /* ... */ } ); + +// Toggle the overview mode programmatically +Reveal.toggleOverview(); +``` + + +### Fullscreen mode +Just press »F« on your keyboard to show your presentation in fullscreen mode. Press the »ESC« key to exit fullscreen mode. + + +### Embedded media +Add `data-autoplay` to your media element if you want it to automatically start playing when the slide is shown: + +```html + +``` + +If you want to enable or disable autoplay globally, for all embedded media, you can use the `autoPlayMedia` configuration option. If you set this to `true` ALL media will autoplay regardless of individual `data-autoplay` attributes. If you initialize with `autoPlayMedia: false` NO media will autoplay. + +Note that embedded HTML5 `

    3 zYK+3O9LgMrvQGv?`a!6*-$EE1f){7lABJC_H)KAb52VJmv@43OJA%)ZvrU>8K5eW_ zEay}R10Tp_N53*S>u7i}ChEfo-=;Y2lGnZ9!KjG>l)E~}S5&=A$a5GEMeU}Ib%~*z z(nz&9Li|C0V(264AS0HUyuVG7y;kzQ5v7%<;m;L98C-hC-_7FIxn-;# zH=Sem%k>4tl$Fao0IA3FVKEh^xz(#abjxP`z{KbF`KNpK_8sgp{_iDdag-gA=d7Ih zx7@mi+cNTe2Pg9uihu0Exej#BN02_M4`X1S6TQHCT;y9Oq$o&cby9A?`IPXH(L zlW6NdHF>%re*6CX8U z;Ca>vS1|6c#oW>X{O|7dv2Xsl`&hjXjm zSUM7KB+)&gZ~U6xZBwU*i8kXaX9M^58yD96Vs~*?5tiHPl#Y0C}+XBkbxbhyH; z)Oac9qhr%wgGKRX%}?>yP_xx1CkjPyXcg!(BY7k2GV~4P=9n)llY8uGQlS}<`Or2g zW>}-!6zWZ`r_)G9z3|Asj`MiMLUJ3z4aCN^a2hAG^O+*@&A6tG@!tf@yScVMGj22F z#Bi%|MLLJrH5l3H42f@PGBg71Xx^~tC;ZC$v`8_!!|eNAzWUUVm4SuAEk(J}vNbcI zAMwboOY=OH9hJja)yq`>{#^RWx zD49r$5#8oknVe3pAgiN{PiWQw&ZcT{V4@$BmDnzoZ?TFswg(mDoT0j% zD$$s$hx7DGfJ3KPFV|FT9By_oN4qNE^5lbPxnZ>2csipVccbm_phb(`GI|tm;xlrb z>|pC}yJ9_W9@)6XJ(WijxoZ7mdrv3Zp+S20af8;bqK7EyXQ*1AotRp^omf}yKWO3q z{9qtZT5$cpMTty6;1_HVMt|vW{&K@8aNtHU9Qp^7$b%towuqM<20C|`U;hcXO(Xin z9KzBUozJqYu+CyYl){QZzB4x*CgiFaAg!DXXB`59s6`?M62Pdm6Wh$oz>wBsP?qV0 z*NF9?w;_Eg73@^OGrC&AkIbBiNHUFF;|9mBh=cziqpl6*t`M(gM!lrLO5&=bG_HV- zo%yTQ)}JlU$N5GGuGH?iEJG+Co%n6b=wp(L(!<{eG|kBmVK*93#6hzXBCml^>}Y68Iv} z1FUGwdsGZTc0qGm!we>y^Z~tyW+1c(?K7QbwaX!({7KT*B#+@^ex04MMFW?SKzo}S z5_AqQLP?c0V8g=+VT-IP;uoTmbW9%wUx1z(mAWqF79AdQT|27cl87d%0h$O)k{8Ny ze%`@FG~PT0PFOrfJj0B3`*9qTHF1wcLCOK|L7T5M25=Qjf+5MpH~p!g*cgMT%xWIz zd$J@Yj8?Mm_<6K)TSfmuoCUyaMwRr~rl^W4o9~E!t$BRRcC6Ygaif#=Lcp9$4Ak99 zk&FA>YFbHW7wW|JVJDChpA_5TwIUv)uU({qoi)*!ayR0Xi^$H7hx8a!KW48So;rVp z@YO2B$WV2Y`;*0rZ6#~i#>p_rH&>g59nC$T8obckgcM;kf`x1A7}BMJiaTvzXkWzQ zlkEL-yc^oux2tiz9a5_;plwa!>-X^rt^AIO_l@CkZ#*P)UI&eSd0?M87;BP zA(sQqYGeU-moUS{5DRgMQ5o=AXh(v+j!w3mK^3qXg)F^Gj67xQPjt;(@*Vq7qTj;h zXz5^{4HUXaEkEWyY%0-ut@V>a#jCAUg4@EQnoRCumJb&k!bGG^kuj$swnv8;XN`ojoLyi*@F5 zcx2r6`*tp@kV{?UF*i|T8`BuF=b~XzfVV998{RGCPiX9dhUrEpxTUK?(b{(`=16R* zH-EXPPZKuUO8fEEB`?;)RK*m4&q5eCbB-X+kTNsrb5eJY5)s?H8Qin`uqLdSkPH7X zl3(suO$0{Sr(_M$zo!IDroPCm?OBNqX}h!ZZ+YDX;J6aQK|%E6>e~K&->#klLS%$} zOaXwgbLX$ecNhQ)q=BDV+NEy@P7@Q@;fYhFjX|a0YP6zJbvx-W?J~J3^X2vZqcE!O z!n5r__G=Y`QTj_`PfP$)iSLF`UbhYXn5z>~b3rFicsFdLa5h1MQ7OjzqUIO+l-ELT zU;#1;77Z1v8!5bNH8E5+p-uLBnO3+EsIe8#l#Tg{!k0+6$aR6+w_tbhFP@)xu~cQp z^Xm>7jKH@b*^U9jpjkAA2`Gb6T5ByNd;c%GI7M-!;pEKciP4TwLsl1*M3Ebj?z}^4 zL9BBJL1;G*p958?Bn7P?#v@|}vr|K7-oi%7rWiQIeYS@;sLU_|P~Rv>#5NUJ)-&qu zx?d~dG_Y^uZme_Hfw|jAFY?60JG>`!gt(5Fg&36u#yF!Av_0W}In`bU^y{{nb0^ow z;p+BZTUEm;DK2fUL${lMEG_J>1Qcy_7DT%p46DWpNP9{@tXDc+z^^ANiNFGnf zq|lw`Bue*=dP7>OF|NILY0SUBXE)(L^yoi>(^S%tOd%=AXGxQ>&`2eQL&~!pbll-$ zx-a-*BqwLX)JDJ&?m8)`wA0rlRPMkgaV+xLK~p3Fx;eljE27c0Jg;Af_2WYPk}LV zd$5Baq4Y5AMN)*Y?w3b`tlgZLhu|CDk#8e&ONLAxUE&~8VD?zI@$$h@PD!1BnUazh z`LP#cBmp;PQcdTQVw`N)=Npcym|WPs_zWcKJ-+n?!H(}Rg`|w@9~Pic-xEQpZKwNB zfHigmY947wP1*LP`8YNp#3uEBMe(PmuOH4cQ}3w1=899A_2$oY z89&ZRkea?)v#UD$1#ki{@N8b1a!HfTK6hf5*IC8BG+}h?(@KfWbF=UBZnN4>tkaqJ zGt>OjUWH!ub(|v|QqILL3W>&*?}9SyGW``9DW=ae-s4D&-iP!NZTfjF&t;cO@Iy0c zPY%hpY?bUmj{M7SU9{Eh>?5uH%IaqhjUSS#+nwYG6B-~8X33kGMam3qLrhc>gLk?E zGmT&z_P8<&{Z)m_z~?LUE7ldtQuDVPQmlMZjg7f;wtsC;c{wZNewUH`|6?-C(U{D# zwcGRiQSb&b>r4Sdm|d93Ed85;U+p5t&_n3Uk0MrB2pRl}{IB(sPddSL4O$y)@33@( zUV$VS3O?_Zv6O6E456G!tuz{FI{G02ArW(VORAmQ0Eb1lVA?zA9ln+b^dLdBU!q^f zzLH(I^ZuZOkn>#{@H6H`aQLQ0ZwF^8rNa7VItGF9B^S?E zvY&=>=W+7WT-JJ+Yg~hAbJo_GJ&E826MkxE7!T@>@R=2I%%_w!#Ezfy+v0AP^){a6J9n3w;qmP zSx=t8qVC@W9L5yS%@#&htVl<6ZyPp33!bUtNFNM=Ssn%EL_(VOAlo939Bx1vP48RZoR=Vu0(Ar`sPO_qlygBLnD9 ztoeXqyRiMOcwn|irgmr(J;c4mkMu4CM0ce9}Tkygul zO9rFjB|Sk^5yGHro(r-CrnK1jlFMv@6Gq{(w-&}uo>%_pHVYs4LqqKhT7SN}j#+0D zcIvqwa!~bFytA5JjUPG`3>mg_%9YZMTP*XthBc=$#EgWg`Lo2m?G>Md%_l!zjOj4B ze`-)WNcZVJ0ozZyn9z{B)nX(#`6wQrtN?d>syvCwGmX+N|047dGFKu zf*1VzIO8k-NF$`L_G`K}@2^urp*I>JrtX#kWyK1U>rxEt+Dw;rsMUVlLM(P@1;Kt0 z*IV9`W8$MBn@iWQk(L99r zqe6EX8;{myl%zNW^Yz2N>7`=SMdH;ZA?&4QFV!9mut3on2|z33K25e_!C{M`(pP zHr)NFoZ~Hkg#m^u#ZFa2zJ_mC=J&G8VPPDFj~9^Pm(O#|`ugon&r`HZ90hiy)C2-;V{k@Xq~VfzFD3ojY7& z+h1&(sBM95`&QWvYZBfyr3@}vr0+JLi+-kkaHVEGH*GII5p3#K3EeFPbhUVG4Fwaq-8h;=_r2W; zH`OpirxZovi^HL(rZ4V)iK;NwymVOSLwQU4?{~lbFUGsyPAD5J{QZFN0RY}f_2}uG ztZgH1&0i(jiBpNa&tC#T_~ih>T}cPIMHwd99oxowTMvzf<%A8*-k`)@BBEbE;=Gss zBERC*#j~LMsGh5}ggXUt(RP3-)mLI~h#FY;sQgeUO}`IsYA5(Pu{8$XU@VNcrG7db zF$X3o+u<5mW=FvT!5Wf(5UGL%-Is0g7ERZ;5M2iPS;O}Gz_Enaxvb>gvH&&?^d82^ zG_BbAFGS2JC?PApFB{}uXO^DintaQR(qMYK9aM{P$=eY<4DyngTIad2y;J&aC;Cr( zej;|?5H20^a^1ggYMxgx zhKTleWW*hsl4DlKl`<~JcT%^y%t-Xy-W^`e{+)ZmO@DPaoolurt|hi!`L679uaHnk zD%nqj>)uRzJ|$PxoN%3>K5pq4-L0kA(21RM-V_u1h+C*zu7a1IpAfvWPjtR&FE!7# zZ_INbIT~t0w0F%nsHE-8CiWi5*}BgO0n zO>M^y$Joz4{yq(Dsm>3^V-e=ZB>BR`WfWz_xe0s?jx^c8e0 zdcNGD=Z*`f@Q!sg)*)$1WliAC+x#Ln?g_*Vu84u#UWPwpfj$sY&h+-7sv@;R4YL1Lk$js?9# zD{^2-IS(;h2$4!}w@~zsE2G1@U%@82PLQsY2B`*(i(+vFjLyBnRtW05CI%Pd<1}8W z?~q7ZMpf3(UendJa#jMn4?Pt8OBSRg_^CEUB)#5#P5uXFMF)36CTD*noF>IHD()O$ zrmg{uAOhu?O6_?#(IdFpm84K9C4R$+DPYHK+tV=SkuSEYQT-x=&MNk{q*$>m9YwKG zkBTvZZ(@tG3e0xzYwlHV>`p_Klp16KbkVu@JfuHkC0nbX#!Cj@Vpw8)B+kC!-&G(* zkGNp}Bvvi|M`LnCRs6AGl_to;LpXz zA$WP%IDoEBQUPBFS$y{_%ca*NHBPD$_6?K72x-N+2^@a**AJywlE{)(zbd^GPq$ZM zgUr)0Y6+J5%aFTqY*4o%m%+mKbupr|W_re3BlDS}7JnE+A6C}jL-gb}7LW$(&DdHt z9#Lu&IC)a*uHF=Mjxm1RM_y*g6((Xtu-F?z;^)?M#RO)-Z+28tWs8-Z!!;mz6V`Sj z(Qul1@9R8hy2Mq3t_R?!2odMvC_zEyiLZC^$TC&&p_+3Zg|i<}#lzlrb&x$l@?sS< znJDvvwy)p0`VUr=Tn?AmHxvDk9$fV;zK;%CJVwN^8rwqA4ZGbwJ&b4zdmq$l10ziP zuGZmh?-5Iu>9xcwHzufj6t1CCHhN|t_Mp_-z|Id>#)Ht+PXe@!T2I0rzsUc5|21zYk#IslkY zk_zy_KuMc{DB$AQpq)-<+P!V6B_`ZK>SLBhX>s5u_U6-TMsKtFcCtX$G4jm`A}co8 z;h_&NGOtykY)3_&4dGDeQ^yu@w>4x8bYZSb@D=1HxfQ!m`z)Oa;;1_%_i3Gb8W} z&z8x|uHA+v!|-6XVUD9%P~v1R*}s%=qD=2GbP7+&2mTMA1-Qrt8yR}`YN&pDG9hsx zt&|AN_qa^`vEW&BW52h}_+*a6=b_$N4{A-~b{&fDPCDZeGjeK#TLLgh_ltd5zNJSp zBN?OcE5&OXdy^#dhJTNk&iY3mkL3yXzs-KVG2sq#f-drBdHY{ZhE5)0NjruN|YeT#p>DYeb|(!CUiE5C6?(WlL`oucs++p!hlt6 z2Mov)`Np6H4(s|UpY(S~CtT!p|I(Mk@^I^%4&BtmYFNKxg&Rxux4k5%gf~k%&eRdc z3r!Gr+n+&`xK&xZe6lHH&R6R;*{nA$>jb4bZmDI#tqiU>{uouAnc)1>OYfvOhQ~|y z`WYe3_d9mk_G8R>>8!35jD)*3`*x3E@X1^ zgJwrhTuqwX{o8CIQIFW1xsWh`*50ClZRlpT<@lqZWm+KZRe_KXpyZty+uKna z)Lofjti%FhhXE};tUQwz?;iRjY&yk#N8fFLZWD`ct0)=w4w_Y24>1|&yjx~JQ!!~= zG_WAV$PK9z_2-~T&WBSD78(k)x|j*xbxmg6TUUyTH8NrINiQF5jN_v>P3RfWUUA9d}uKVN)keM<~h-4E6vFPMkvhEG&vl(@R38*Q|l z6qBKE%1kdL50D6X4?cYsCw6J3uQDyVX>=Z;nQ zKmt+?vTvVB@>k0PK@-j*JM=jx;8kS_C-xzuEQT?Qj?pZ;gD7D_RkUCUPZzyS(=qzRSfPKO+GWHUzqtxa!fy(@w_x^JoXnniV9C0JTLc<>@< z)+A(ndG>dbjX+fQph4}k`i)xLu3=@8Z66mu_y$pmAb+8E60m@IH@bf{#CLLAY-}Ug zsQ$%BW}9s+58ZHp4!U9Obiym{n#$PD2fu3@@@;ZDl1(X!ul9&{9R!%UJZ9By#ugdg z(rVonAkw+J>u~`T_r&;!vA27>e^8|6&wCjSZPYd&J^_hMc!v^bPm09l>ox~Ic3&7o zdA<4j#@MI-P9WcqwD;)7QBcN{b}0MHM7>WUVMBpW!yeR#8|En5Hyucxu1?kKZ)Ix&NV}d7&xc=G`w}F28<(J!`D`(V zoxtCiNau(1llkMIEbm$0xWVs5nAeQ48e%TwgrAdjoZ>zSJ{cf$A*@l&`~xCdzW$?8|Irx<)YKQF8>)rj-3kAIq!W{eTl6S~;maAktP^ z+Y=KUvLmovR;}>Oh&*bQ2eo(z=>NH&2Wg?ND214D*DcvL{<)5}HcLXG0=t8z=JqL{ zUfvIHg##b;ZdytHdo1YeBR|aL`(Z>0PN^ zwtDT5p?9=T)+@uuAH#R!q=qY8IQ)iE=mt`E@F8TkgM^z2_uo(Z@>IHXNwt+9^hbMU zMR8SYYuMD95v5qRwqkSQ&hf(5+;iiw^-2n$X<^Qv+ME1QUz&y7MDIT6cAsdDa2*7jN$lUwoAX%w z(es()-#x3t#1gDz@3OC*Ml^DI-r*`cDBHaQ+BkDFbNP?61iJFS>dt=O1JomL$99p~ zNB!TUr?5H;%K@tb>&2kAcA3&lay#(iL|%8C2n1TCo_K|4@fz|zt)aIMZ9%CR1=KnJ{p|UrDlyVut9mB|^Ejp&k1nP~F0@6Jx03?5&y+H*(H) z-{P(XoWW@eB7(u<^xjExQ%=Gu{RUw$h83szkjW9OP3b;^c^0HvtnQ`O%tm$8Tl7La(cfB9u zOM*c)-uW9Adr^M%f2dU)mTqkx1d9pL86ERIqUw)}#UnAEltQIkp%sbLQ_NVc*@fCv zMgE2MbY(=oMZs0LQ>q#K@MlfY#oUbTx()g-Xd~C{ZP0;>$!d9_Wk}iXvoP^_r9*7! z%>S*`iEKO&XN-{tTb)C0zsJZ(B$Li*wg5~5@GM@wFY!W_VJDL)Qm)SkE?9T+;&lkS z!bdrjl0VU^l^LzQ$}5fn-bw2~$@r@cQ=oj0^}M-ts(_k0)r2o~2HF6Fhd{3|D-kT^ zl!8=vLelUddXaLUYNPx!m6DY^3Rq78eq`VmcQFy<675hdmqXf8$H>+VC*k6N86})y zITm_Wh?#OJ!Q&`AI*@v>w~|@Mu*DquwSrP6S zPC3R&P!vBh`gyD6Mu9066xuc7FFhAuHi+Vkxdx+Cy5}kTcTx=K_?x<>OWrY@3G!5PDmsXn!Z)k)%h1Uv|%1Jh7}IrvyAy>g7zEua(p#g z@vC8_J)^;+f#CRnYXhfB-m4s5tLa&KHWHN@(yTgj>Q&q39nvU1*KwIo-Mc|=>V?}O0mUFToqbS`?rmY;SoL>5zUwP-S>Wf1cJh~5ilp#&wTGn3qUd?-${8l zQ@|E(s>(IME?nfn(xPWgnwqZKL+vb^SY87kEv#n4ZX@Txrv_aQ>2-KX8op4|<$ZYi zl+Bz(b2Y3Q6qftPEo<;sGVcki>vaq|=-z}Mhtv+J$z;FodzsVOMuMD!Am$FTA(VG6 zQ9F0k697Xa?`yGWIGs@V=oXCvBbARB?JIj@4f|sphDil|Vf0zC%W%wIelsIZuaXKK zET7FSDh)y-mEa>{`qZkS<|I z`5+W49|p6ipX4E7uv18zS5+Q-ZSNH>TWk*(TYGHsXdvM?Q`5G8&zL8v!Bn39^Ae4@ z)ApiaXv)ekSB$GNg2briPRvB~l;qW%A04@)WM2)YC1c{MP!Zk@BUco?Q|;ZIp_w;X z{+xXh_G^xpxKOE@8A2V^FGBXKg`8MJs20-YO2+SqmDZVZy#qDxN(j_4LJt!eRf)2_ z5;ge?!MI&UK2?uk>$n-*>$cG9LTl{{dE}}YBc|-H#(N&3m!l1VJ&6W8FLut4wptmn zDAv4J>;EfpKJEV$7BF7!qaE>JyMO5Z@0a_%K2z))NEv|xuh1D2MQP9A;XW-atNoYd zs-{)ATCz^k$!YLsDWK>sx}2;)xnwY;`RuFsB-W$!S`A=3R+){=h>mJfVO9n3qKSnr z=-_KGJ=4!FNJU1`AWHaLa^>tE2%Zq=y)F9|pgdN`%lz#Yo+W7DaMSNE$d`I!02MI*abK6*Cdi-_F0Z9&bbt*Zy$ANMw#*(O*86l=ElPd$7TXhGhKqtvU5 zCdQ7Uy$))H-9#jr)>Zs^@%^4L@N6XrB%=du77M6Owz&V-HQjagLMig}?M`^pA5=m@ z5WGZN3pAJ5^&tdz4=c^wSQyVkaXH3%f3zVmr4_w+F3VWRq9URW@)ZWPZ3r){Jh=p3 z(KGPQ6fEYV0wOo)=qUcLjqqT6czdwHW5L1upE4x^qCf?SgsqQ>P+iB^m=f#B9;ep# zkN_teyWBY<{sXO=1DS2YIyV}A@F&sBs|#np=|_5)P7|BG8+U0xz4%?SYo$_JS7@9R z3R7Is_Ub{!k8HcVq8}&v=M0P1{>`2L}sC%HhcTMwn8<&`b zrbKa}cl}+7($N6igLr4T`w^^R`cs3ZHp-R{xwh{0{gC!~jmP^WDTC01jpIB~N>A~lv8bEjfN8izOoar~L9p6~M zf1x{&;*}$p*@}kZ)^EAa6o8xitr0gkcFHaB*H&1Qchn~cc48ETP-#xp*9Y2&d$Hzo zv!bmPb;ZGA7{Z<#;#4U=+ga@dz-gJ>5bu<$W6({^9!ShU%WKIXRL3z2rn4GkV$0?< zH)MoDAiMg>AI!M{>SM9Su-Hv(R_+1A;scknj)|$3%UEhg4M$S>Gv?qgx?`G0S)vw7d3#AowL%_+>p2yvZucPCxhPDMwR5$h9OEq)1 zo6hvVy(5Hr`N>|hg<%}Q;mQ+5);P*(8<M?x@I$~D;X4mgyw&dCkake$s$Q}2+e>$7uTS3#W=%6*-==Toi_1Mb zsP*}M8~-n4^VRyuUSmyP82jfdT^P#*J{1t%ig;D%@+=)qanaMpgPzr9oz7um8QzIFi}=#1qML66gkkKKDK_2Qh=5c!%~y(->zS z`j8yeuGIZ=O=EbJdCc8kSr#Z}_??dsU`z_C44-o1Cx1K$wIVWL^JJ6Ek%r{U)!i%u zD#=N%moazhgCy<;V?9bv<4A&@PefcB3QLmx$WIDfzDB-l#8a&J!kw(_JyPDm=t8XX{oyxzh(FP6gCz5)T!} zZf3%D#Ujj%g5U42_z4=8k36?m6a#ymbrhdzz$NX+i>KOGU+d@7lzx!7{=0VL!b_Ry zR|4fFjhDe&Xvc?dR#fM<4Q`bl6IS+2>^0i)$DR_Wt6iH48`ZqxVnhAB-1bE`?e5M~ zQ-|+&7u9CTN9c2h+DdI%$-)TV{3^rOJIFmn+%;+LSx+AUhw|JuTjw8%r9Vv0)Lyq{ zclF)8v+zE5*u(C>1$W0ICXRHI-fI~?|NQBzIDp&VfJu!2FahxzoU(%C!8$akHee8R z$LM4@KYl4-*rwCQ#`GYsuIWHahyeee2SGIfO14?B^9iFs(OGBuWe9 z6=6D}L|mSey-bv^ixEF3QsxzptFkq27ev`q>caRc-l2Aga+ay_V&)2=nABDAHqPK! zZGY!$_Kl_WA1ww3_Tn_ylxc<+)OGO6^TG3~SG7oc44YR-tl-k;9sDY$)TVURrA?p0_Gnzr&davTc@3HEwS6oO@Tyi>W zH&_Sh5{#6jnopX0G}XQH;&DC~QzQ&Ruov&&i*hEW9ozg?GFxi4KP-@4V2E?k)ogf> zXAdu6I8A+E$$#RD17aCo19cAO(2S7VnXAmfky|p9!{q9n5xP|OU9fMm5+vxye4Cj% z$6CPO82taf(Ree)8%^vDePi{H0mUXfJnOUbz1WHC==X*>`AmElB}%zIOP+nj>vr|* zW8tRj+1e}bLHv~PcOVgr2nu}i1&D~Pm%NX9gMO`pmjv~p{eX@+PiKeO9u~n`LN2p> z%!y<>rFF(DBvYi(Dof+RMO_yWC+xvpuHt3UDj;u&Fe4=SVUPc=D3s}0+-ILr^VZbB ztLtA9xA|V5_M`v`%2&>BLEWmVIn)=w*?i`-XT38(0p;$Q0X0bDJ zqWn;3W0F|tV0E~qBCx-R)C!b|8Eq2+Z>X|PS0A6V*h>m-jFw=X-TJU_D|-u1TT!)~ z{B{>_S*f=nPBTY8MZ}w0zZ>Nt>g+t&UxXj0l~T*VQ{;;`FQ@lrb@ON3B_7y6t7^vg zCg#IKH64e^MknTam#}p`C+gv z2lsre;HS-gAIQO4LdHn(2Y;XEC!(HbywBq0sKdSZX6#*`Ss3R8^y9(5_0G9Rgncey zV{rLz+>|@Q;09rBv|(>zrAL##+G65hoK&vuj&y^yC8c@X?gRm1CQ27#r_P|C@&SDd z5UVoH4N)vaDYvnWNAO}uYO)-{v|1<1aDIgK2H86OFse5;btFoh`)TGD znvF}+mSFCykqmllQc3(JNV6L?@71?by{53gPC*Q)_c$Jc3LLD`cp0}og zm^XBqE8)A`q}PmCH}5jZoZa1dfsHQe+pJj!}oH*^`)?1*l`-lu-xVVXha7Hu5aBQOwt#^Souzx6$E@e+q<@|+j{G4 z*Dz9)V}fk2zrX5(@&Ya?V!v*Go%|U)T;3qv7eQ4!d>c|3avC8zcf-2yC`KPG|CK}6<%LYq+x zRA@6gPb`oBZ)o!t9UY$jfW92`+_L~kf%Q?-t}Dqnsv3kd@0N07eq8#Lc5C}MoHQ+fsrWCG7&wDIPYXTuQKqvLiEdS0uwC{Spe&vVNTD;ZaxQAT+%BXvJ1&KQ@dRM1Ly zMLhn{=lBU9$MYPyNqdLN(B11bh{K6%@}U_t-c@SFr}W}Nlw>h6%EHWZZ03|R%j`A! z-zXAuMQY4`-#uUGN5GKz0MPb;W}(j3*p--_!^OQagOx@7Q~ID^3CrdZALqC+!k z6Rk3S#vJJbXtoF#92V18anC#(M4DDwNNPQpm43MwQh8dqrhRTCO;Yj3|4Y2WV^_tO zjh*I=b0R|J8WfbnGgx~xYSmXYnzjR;ysK+>173C(MkLvYOZXKbPQ?mPbORNJM*9Y%9%$lW zdO+yqprKAMMMx!iwd2Q$|IDM>;IB7Y~4=-Vjlp5@}>G9%M)wWZl%%|8UJb4ehZd=WAriz+nh z2)V$+)oEcd6sK0o@5vA9mvY#a>Cej8ncO^Z26uREP9yU|cvFu*P4nj#@L?51OJ3?~ zt&58dAj>kEkgYO(OdwgmC`__`_~PNTQ{h_e!i7|MX|J5U5N;O&_7A&jkKm>kLDXS4 zG5K9V8>S|1q|ZjQmqk*p(|!>7Y7%tVVWBSYO>3G;dS(BBbsn$GD$;jos5cSe-R0<~ z5E2#_u&ZUNyGJLHiv#*Qspci|cN}_P#;(M1%gKdIW!`4`*G78M3B9q~=GNMO^fz$g zbioG^5XI7m@3J7)^y>8R*el_#*q!BZ;2)SV^wcojio71|IbfVwO>O>-Z$Xe0?XP-a`*Y4Lrh}UTb)T2YO`ew06m0c{h9HdjLzCOmzEWOTLTR9x4X3y+0$NJ zm^HK@%mErGn5AxygSVjhGK!AkAG7knJ7_4~=G2={AF5bgJgU0xA*NU*`y;u)8 zXb(V2HclKoJrV-1ZU29+rOTL8D^kU&hK$6&@}BED5a`T^WK4(7P#KPC?P`qK9Lg)<=4wL0!+!Ri zD9+)pOhUhg<^-36pFPjf2gHzf@7{)XppE`a-!Gh|Vv5q)x&-PL$LXcuZMtoi`~c7R zQGKT@e8qIh3ioIg*qNdkMAIjkD{kMhO#5pFGR#DiU*1p7mQyHSrc6PB@hivtY5L@oP96L8Htl4OUu3`})p z3F+upL<44hr`D-PwAT(7>{etEs{(i|3z&%CUlYwzZ&_0H%h0zJjPTcsk;qW^mXwgJ zp7u48@ElQhT;G?geelP$WsxfKn)P9HwC6&YvyN*h$MKkPHpale#3ydyf}GX&>e~xY zaL(QnG@HY*?doKz@~PiQd&QI$G+yJ!{XK3i*U9 zP#OODK*+Yvy0uB0IJ#77SxRL@V1#*uzOAUFXLNf%4E5MZQqa5q|0P0@4&oE#a>i#T zxY(mkhkD2AS4LzoX3ipS0#g$eQ4`i z2h)PLVr%d^_}+kFl;TIiCCh>=`lT_8!|)fQUge9RyJ`{TYpwMCHJSdEAEF@?zTD?8 zN@3_xXbgwe%Ym+=pS1S}M4x#Micd&BxR4_FNoYondUdpG-(!r9PG8d@FodKitji2# zj@U;vG*I`uLrqQFvqt?g*)|D%HVOzTSAA*tAaO(FMv)lIlPWJSw%Tph__CL6hq|er z%J%i(^akhtwHgL37At+ibZa%{+(~_Q%X4ryckF(vAGvD9{KJLG#7K_hg|Ljs*+d4t zt)PNoF=*AUx|THHy>E|toE9U0xeqMh+te-w`6l{TRgW&PDGsoA#Z%_@Q=l;qguI8I zn?T8ME8Na_DH{96<=zP3BB8cTGCNA96D zt<|eu%@nRp?|%U`)E&nY-?iwkig~yGE+j6|(S0OEG8daO`YR7-9TM$1e71iYNeAV_ zF<8DqP3-gVJ?nrCSQgT0+V<~ItZASm!|uy?R0kQ>_?0VxSI$8J#WBA8&n zmv6T2v-gNdp00UFU2m>x_&_l76H0F6j{Oq7_5NsxAwywE-QpFc*L<#y6)YALB>?H-pCx ztP5k|i9FfhAg*Z?sazneL{5gk%3}E2!D-KA$d1l>Nq4GlO9uIc@sp~QqJsO5)oEgu zdQ|o-A>D?%IJBdZk67-1qhar8fp!j;EbxLG{Zu4wn&bJuXIjcLyCI9kNMQ$pTmxl##5EB{K?c_BK>eZOmPBGg=hCkY8Iir&Br8>>{;%* z)3e34G3Nt?uxdFiRy>#-Uw(6Ot2*!osh)oC#T22wv3)tXv)rlT+c<$b0=DFz0c?-u zf!sTF4O?6W&P=ol z8q<}VsmEG;Bk-OXFVJfz%gNo<%V!INaJn!wQCMSS2-x8{v)p|RtJ!2)7S*U{t|cuB zOXj!Teu5Rw%0UOLp{j}q74BK&+Y;?e$mq>}Nvg>POC|Kz*I^Uf#OI5CKMEi1Y6X{~ zEua^9#(aZ{5+~n1yD8_1F=OV%~$xvo)9dPawu9O+w0Da(r7{w&|UmZsVRMmkmSOW77UBKLcX z`)C(J78snEJ*zG>2I&9Cq;s6ZBqKhpN;-$26TMT}#IswS6=(6|Bxp)frU2 z*t{RkON$7yO$Fiv8&q#h@A#s+z60~;!9<;b@D2=-HNh-eC)9_pasG9^A;Q}|ZRrDFHDOneGKzwkiN1aph@)EPEsPkam9Mwc;AhtZH zLp(DB2iXfweE8u4Fz(zDDP8r;veO~aO^dDEW)!xw4couU-g+yS$W5#8oU00c$4@^- z1fFERLv4VaMPfZMTJHA|I{gK;JDQj6oZy@dub%5fc^o7`teIp#4!zKX)&I#WAHEeS zePx;7@wLUITJ=3_-hIik;z)^6%h%O75%luo zhP1SA30cxnI=MG4lgXA^VR zzLbRTb};08qsGs_z4q)>0oGGq0L~F~8}R^YEVjhDQMQuF3c6UXDd5FSGjuOdRvB5Y zs{d|maARFXm})0YN0hj>tP)2r@$p@wQlo#RPS#}v)pIpjy?QC&3hnlZG>#JFC3>2z+{^o&wDyEW4GxSA{x|o$zxcuNz(V3M@Lno2Nqewg%LLp(rs9@h}RGUPp>mBCLRhb&4Mr8{@8MSeM>uBtC&Jn z%jiGRc+h=taR`5ZX=inEoR21z0oEK@Wg}XUu`5WR!#$sk1avYd^r*hLUlm|!6vvB8 z+3rYa)SwwW^R2@un#KEx5r3P#cT0_IuyX@KVnxpAEGtw`IwV0ju;eIyp zs9P73ZF+cn1;-vz9&V(2{+}wxCl@L&^aP+9nH!$^o6y5Xi|F<}_XXxnzZNms<4F{R zyYaCrXHM9H@3GQotd3u^A8nv82g&i>*vRxmoc+OwQ{Uxvhwe}%y_%-4hkx7-h>qoQ zqIo<}G-nm7MVitCHeZ{aM&s4-dImYqqpQ5qd>gn_ydt(0{HhuX6}N7pNlQ!cnP5tN zbR~9+s{1kxEag#-fit;IdyU3*{JMG^wK?jw{pKk71i)#Jd+y-Zp%1pNfcnRR^3%uVZe3wgYiz?EFH*g^A8QvCEswtt5{Yp>soL@=L{E z(u#*$!@5ui>PR&Z>|P-2S$|%`9aj=>NxY5C-YZ~Eh&ftdmeTqbX4d>B`pF?Xhy2Hy ztZ2xAmZ#lho1pP1?`{EsPalG?>Q+S9TcwRw2@MJ|CiA`g!&TVPA0V{&)^i~-o4tU9$7jhj)#$8ik0P_WALY$!iAeLbN8<5~Op^b8dsL0nZm?NWRR7~^vG>Qi4wtv!J z%Ta)5moJ`K{i;7JPc=CoNKoVrwEQB5Tl}WDHjz6PGE<=NGYP30eA+GAA?FaiR_{fo ze#S1IsIeM#({_7s0Oil`B#SAcMg{C`K~jja{@8gsitFTg6ASif%CL;`(hx^J$%x%;@9X|7BzPo5a6K$rrDCFoKbF>z6Ug74l zY!Nz&Cra{>;#@{Ce~uob^$Q)3&7-J?c{}{wI`g&T0cdRzqfMNsB>70Xyr@0*FgeT?G|sA5ZJ{<*YB?09{89EPYGvl zu7xXYtNp2XbI(5$$B=x1M?0q_pRo`nF2q6$zi4%Pz?l}ugZ@5QNPLNbG)%Mz@?GAH08#KH$XTw0r9NOEkn{Pu< z^K0$@P1s^}%MSJ@-17CqM5|GtFB}1NZn;J9qp?l-MUP|67g7uxzxaZEeiEE8769Hx z(IA8_pvx#(l`j@@7>gkam%NEza}80Xb3nZ$@VH;=RLR=JlJaw5T`{&dsOBF={pM&= z1%@q&VlcfNZ6kW4t8nQ>(neF_lMe$6;*}2Pib=%RsFEs!lzl6^^G}##Hby6q=RChK z^p3(zVcUC@gStJAsF3=#9>U=@yd z^P!WaSycH2wD=V+dY<)8{nrP6^da(8HdcNq%yn|{@~UUIHhen`rw-&$ z-iLpMBlrd@y$cpC&6J+FK54oVJAHqczR*$m$ee{}1pU1768$< z&2&FZCP*BUEcE#M99-^7P_Q>Nifk*5%skIH3*F+Ul}W8Iky&C;n@yO6-8){fL9kgL zf4iVK5$^wHAP=*{b1=h@cBIXmjfm|jA#%lW$(uN**mR76 z%}>q`0V{%j)o~tsPQHGb`f6V^CyI^=Yb_pa3LZ!dr7p0Zh(}|Rji6qjI~_)w?}w3h z%;V3OmZg0fel%W8*x`sL)KR4oPrP!iM=8{K=B6>RG%7#z_q9uNFPplJcd5o6(puV* z2e`C(Rwrhuw-7cRa}`tRMthsED66Aam5FB~Xn1YtVvUUa*7DhfStQ-LHovqEh zma6p*k3*Vn8&#MiOkSWH*+B<~<~I}$4U466_VSTmD|RgZ_FwtuH9Hxs2wnMK{njE7 z=uk0yccv!r%bX{DJos90Zij>Nqxq|td7LfylaNwg9~Gy|XTiF}RMUBQB6k$BlxH+> z&;#iHQW`K$^9AINsRVEtoiwGZje47SF1T7% z=E^C_^gJSWP)Alr#))N*{fWc3#vQV6iz2Uo_>qa$m<`*hDvSc>-e^b1Lg@2;^o0_R z(owwMKHua%x@cp`k0P1%u_z`vE(I>B z(PIV<|^?0m@;HWtirH67^4`%YA&u?V+hRIei_&r)`{?SGFr|JGVtIe8sD zdKOszQ|e2H2hKcI`~2<;mn|dnDJ>1a1Ns!XQs$tNZs@K6Q1i!Q*E`mR^=GFT*}QV> zwlU#g;b=&nOYlWzc;r?D=WB#v61S$qcH=1E;4GzRblsHnaTNL~dV+aIkiO0_2<{hJ zo(tBk*kS8%tX2zwr~;1Z9T?*f{E>!#25tN@n)vHxYM2?F7mHWV5&eqg>7h?6D*Swr zoP1R0K?OH$)d%T69UVG=-kqtHArT~8so?`t^{1@M z^vX#7Gr<655WsuSsj+`%DOHsz%-w$VkK{cAKkPo4U(G}!w3Etgs^6ohoALN)@Gko zE~YlCN;B;H-dJJ$X*sJ^+0#FA@L@+%_y`Xdov%kdJ9Cu9#or6zds(zKE~eePZT^BW z!^c3p@sWM87|MxjVPHSVFW>|4Vi5Em;I$I?DxLZ}Ao2Oz3a$McbM1M|Rg&!)1pbcGebe=P zukQ58tj@LSOWms+-W-?N1H(V?!GoA$b=}pTd9pNu+ogJIGI|UVO6fD{37s#tCq{zK ztw$PYU#f|oQ2%rNcXtrMKf=C%i|@@yee{0$`iW|Kl5KzNGZBF$R*vn=3kqN34(-2+j zXduViBf+bki^NTz7a9~3qWhLZSs98Ljove@5Nok3V?i>Rh%^Od#187iV3X44I0-** z<>_Z-mN%8+5S0!YNtSO*m}I*JP6((h@1>$#D0b!;=)Pb{O@D>&9*`Ek_qFTq4qz=bn;P}X_gK8C2KB?pa%t<9;#YbhJ)1s#2$gN|oxYXOHQ<7a5cusKuEB*0E?SY&XINn#I;0W> zWB+Nkqc=@jkfA|hh8J9K$J_cWeY@LB`b|x?&wi(K@yFKNbbOV!#vO8C=xF}v4p_6d|3fwRi=Eiit$tn!7#%c#nozIx(TxVQl$1@W?!0|mU4O!^E&|?s^X3a=_1NV64xB_3y8F8yF*DVQ z-vs%vs_^T%Mp?BmD%@^h^{9{bHSdG-J%*Pp?W0sbHPd#s`mi53_TQYnA|o)jb-_^= zmaIV+^>RY0wQ(EtzI-l2?$4*4%1?8X`^MMh9)E|^RXZ}of}S=9e~`*P&(AWtSNBc2 zrCA5@tdo;i`U1Gbuq=pgB8^v7nG3ZzW<6WZ{9vDKeBJWeli3K3&#QXSzMKZagD#uGYJ&S%Hoh)4{QgZXc@c0<_;;>8Gv*pf(Jn7EFwe7MeqnnO8 zbbUP^$KC^84=C98y+dJ85$C&{#142$&dd$gDtwzV`uiRHZ-|W50aP)JSp&i?){|$_ zkd1b}2R;kiDKf9WjeYjJE8%?a`>s88_ZIgy7{!lCxfaN@!Rdt`-DizK=6}GF#AHvytUrVO{rG7bzGo#hT)zH^`UBw@o zx`fb&0`uLG&|C*K8tCgp+66El-_(H}s+#i2aUv4yWOf-0FR6)u%r78=^1if*_uB2eA zI@(NYI#0DD>_&(c<#8o1D_46PPLpbG*LPEFt^v1r$Y$iZe*n%e!AfQ;0u(dOA;)b^ z?C&rmpaVVCYTxWKl42)pKVZfv{HQ@`OgAO>0Rc?1H(;$i=iFMIF{6I?04t&4Xf(2I z#3wMgCgIFcra~UR(Pw;NQ=AQAGiyFe;x@%@{N&`TY^#qX+xlbO_rI-81HQ3xWd9~rN_-U)-|43qg1@AIO ziSbAsHnI6v;9fTwfoAUg4#B9hckWSbuXUd-nlsKO3~cLJ8hKei{kKZ{6l74r{jZo*_s!ya;CM&5y==>UmL7=|Rmr**pRJ zyv5YJvc_lTXViEYWu~KtZxS|rW1CrO7}!D}?_BqVm$&*H%0@dcv6AKBqNb>-!^s7{UWCm0BMjIX)IcqD)`6Joq#U6nl$^p}M8j z2?X028K|phxqaMclt9M=`GP<`Yqw0xMX2F<^-?x;?=;48SP&LXxij+H8^`CZ#Y3jB zD<=~AS}8J!hmV(QQkBfwgyTtpDGy;GNslSCEIR z%>O_22=^&ky*beB@}U~uc9{;nbQy8k2}eDfD63%N<1J=E@gXMmm}ec&rv|m%A$fp+ z@~5XpWM7Svv-hNT0>D!8Wf%qVvy%%z&dkkzYVhro|_% zhxsPscb4`y4%&A7oH%&n%h8C~(MZUi{CN53qs|zf*&L6bxAGo9E#LQ_-bEbnA>-U6C9HvIZjo0y4U-SpA{omRWFhQAO5?uj&N6D>;IkYzu_pR3!_ zI*irru1>1UXUI;G@_l7x9yNZEDwpp7Z4`Tr^o|9%vk{y^>p2l!;F z7v-E}vSKjMwE@~1$JhoAeFxzz>L|9)^s!@poWfb0{&oKsVD6Jn=x4&O9}5!#;h$u( z%?#3;(mf@#WWyLWH)=)Fk2qI&De|g%+cG0nCSDWN9Yi)Y72;r2=bj^SO9lGT4oWS6e24rB$U{qP*#6y~8FLq1son%JZqIPYa44e5 z14lMq@ijaJD7!xGvWhPTU;@5jM(9Zz@xi|Wqy)x^P;Zmh8`DOkzkh{Qp}Ds4D~v~P z?fXaBm^r?Md*=ruKpdTYhY(Z@jf87VbucK#qk|?V&`CKe1j0V;7U{s<`PGcJY~Dge z!zAuklD(&5?|YYhcoce-H)A$glQVKYhy(;M^~+4kYT7EsCM@YKhzj0$AP{BHZ&}1T zicx|AO;Eg07NMd9rL8c?2dlsdAlU79*W1s+iWtd(s6VM;B_hkG64RZE*76p|O#tQY zd_vH-hH*PXA9dm(7d3+|&{}A9XhBOhSKi?;#VPn*zDw1GF-2xaO8(ImEB$TMmCjgF z)Ue`~ZOi52!$FEVQ113VQa>xgoc@Iu8iO|@3TJn^!+&42N!;n^D1g0 z$M92ZGx6e?p6PQi3RXN#L^3xla+Jvnf7TnjbOJCsiTPlAq4J z=cdZ+a1E7@Z>mr16{w1s+FyfO)*Tbe`g4tpy!~DsbHp{fSP=8C^Ww3x`BI+`^ZW+o zP@9ct`pN$p)9OXVhMnkj7kRS!e|xs3V33kpt5(k#p}tz$Wh#A&gD5Cl2MlwP$X<@g zI9r<^5uy>*x*^OV_>*3SfLMSYIz~D41*}PFoQBxLIKP?sv(i%aO?Wz)RgSIjDKh)N zUh{D>egPdzhWT{_L>aU{>uUG8@+Z1dgjzwnO~2&K=d1|62ef8!485BD$VJ9$b%I#4 zn3*;;Mz{ms!K_2@{o91ms@4kM_O-zent5dImLN}znIQx1utTLtuyH*#%*3#q<9AlOrLM&0o+mV_DO!HhswBY1k`9q8IHNHyQYK_N#yK{| zjl4{t*v~8bxhrm%EKL6)TH8C)!tuNtOdf2ZIX_m*73Y$Lj|MAClqQ`7K{Wvx8rvt{B7d8;5BobkNZt4vwMDJU}aEymjFJ3S0FB7)(O6-y% z+go?L3?0kVEQF?1RF|AH8V^}82B9&@cbxdxPC=BphxwR!IkHeDfW8zm2D$JXy$7R%glb?2sngJb;2xI!_k?RWS3Vyy% zru(taF*Y|?(jq^FBdhw|YNQWk4<&B)$G6fppZn(hyA)A+2M8Ef{yT`moV0pdpk*KffbCgU;{EV-l^30vX9u;!Qi=27 zh8l9g6Z#SU2p|JjyJA7QaGZreRQ^V6-4~nPBi(S8%n(aP+FL6!dhICQGjZ2maTasO zDY+`q0cPj(z^C!&XJ4~qxhq0Ju*WYmE#T_=JU{xLkqYBsMtj>{(|%q<_%kY!ijxLKGX=#upJ^yEd`bip8{dw@3q?bKF4{QlCR4typcuf_C)AlnR7UlpSeWtS%ppvqm`X6F6fS68ZA9!-+nVg|!5@MHqTa#-WCScLl?M~&8-&%Vo* z&5+%jTha{VlM8dFoLJ&!$fG@nN(*K88JTmsXFffqX|pHP~jQG5;dZ;|qS zqYpzt`LQHk*P)Tsyt4CFWU8ETAJrfhqXm0~ScC0>4|vZzL|NJ%bY`G`zhfy^>IUgz zA6%~Ora%HY^|1*3LerX3{;AG=yi<{G_9Q}UziTnfao#vF_G&kI-!bP&IGSYK_MY&= zC4byrj^s{P3Tolqa^x#k5EzJv(#>fneMXkTO}Ni}snVo*e>=(zer;aL>P=s|Y#Bch z_8_f8=BFrZ{J3R>sQiQf-nl_V{4*OBpnv6rw-v`=s7N$k--sdQySd}Sljh96TKm#8 zb;yK660@-y%z#>`Q&*RwvTQ8VeLe@i+qlJZ#5%y_c-);X<7Rzk>Snvi*0BV(zR|zv zpRmQnply|RfMZ|z{wdv)yJG1Gn)z`)CFqo)4ck#GJG$Q3*z0jMVu>h z6_9+ypH8HjHHN)1y_dHKapM`TMj*I&wgct?5vxq`Vd~r+)&NEG@|1CNSr09is9lef zM@2Bv#@IUqlpdDqEuwh^+#|`5C0L)?mCv2AI*8w}wzv|0cg5ca;Df$7&hj03K_^3x za8bg`K+m+w&rhhsemN|O3eZwF}r&3;0UaZWuEcYn1?Kj7Ip`NmAt z9(M*xGFTOhf%vm<+Q=V9aPRD;A0-KoZTob6br;OEx(*+>{v%|@UXK2$TFe-4C2I3r z{9q%~<2=W4O0X;qhd&8l(W%EF2Pgu`EPahF5vgkF`cUP!S%k}SAPt?I`S*z%Hk&g3 zsA%J@?&-u!^;7@zoGO!mK61nvSA zi&$gmI>emxx=o?Hfg##s+E+>fwpV^?#bnrw2$tSdyUoDP;ENEmOlwRm36W|!HOSM# zlM*x&-hq#H{SysXH#arucMAY4M0CrTEHf0q&c2XlmIxYQVp&p-KKp07O!MB5f@Vz0 zwV#pml&xX?s@BCA3TK^mS=UNlg?rBGN#k0-Bm47Id0rivx5%HZ=p*4PMhTzpd0q90 zN@UBNr>bqIWbW^6F50W*THx904QB6+>z)?I1VTDa`}$8;`>7d@tB1S_39p=_T;(0g zRwaksePgyIyN`UsHN9Mq86DlOR}K{6Y?qm{0Q!eKB3_fwp$|<+wrkgD3E+Yo*eJD~ z_OmVjy+FfweVCz61=iA*has?3QI5qxasyBFs2ld0DDxaFLsjvyk@+Q3! z`!{;0Gk0?_Yj|_IyEv_EzQ%V~HfesqWC|IXUc!BqS}Q~7KA<-hRprYKOj zs&vTLy9+0rz8j@Qd@l`$jjv(jqTtGTR#xn z_yBSF@jDzgk|2fYOuJt}@|^Y!l0oq!Awb2uw|ufBGgFn}!K=dk9JB z@w?PeXSYlyjhCIMF_xa<-KJM9Ng~2f=eXKCBhiCtn5SzE|9)rGd8zH@{SI>7UoZc8 z4G7=GmP@}{Wd-6>T$6uKk~>7_{ku}T-%Y*Y=Sw4i5HnYviIlnm*gstloL4e`wL6=` z;;ke@YCBhQdzn1jDNN?f6kjBMQ?naMUgm5u*oX1DJLDvhuhGyn-^?^=6O7A(+={IX zHtfhivmbO$jV$zK%47M1>D#8C2w#`qSLS_}(K*4o^j-wg7tJkyn{hwRQ@u=#cnOSb z4SgJ~S1E)J&DXDVBVOT1BH=3!$+n6r2I#2jM4TDw4k=bIt%|4FLeBf1h{IX{tB;6# z7Nkn`>2O<10)h2M$V9XQ18+4~UcCnS=sIhF#-U^RZ}=VleR^1nI{9dMI8P821!e1I zqY0akp8QBXsh=86Y-wF`ZJjucV;<0PnRhQ6>h9eEmVz?|In9b+Cz*V=|Cul?AAza0 z=okIMQL~fY(AXLv(>+U_KX1ITxEs8rrjkKL#&tkvBrFh}Fc8^VIoRz-4HID0TjXRQ zn(!O5)+98x@+4opJt?Z&09i zDJ~*vfD|0DS;p@3axoi!0YM$6Eqa2VIunSc%;B2{u450bgSCsQ7H2g&6`J! zJCD$Y^;6JdH~=U2$Q5d1d*PD41dn9Ym9O#5S>kt&6_ZbS{$DB%>640sW8>(rH^cH2 zNJ)eL=K4t$B=H-ms@Hk?(dlWqPmU=$l^DL2IhA>@7;MOJ->rFHnN1K1Q1jibNK#SR zw8Qf4N5SD+DZZQ(vqT=+{>#6(sRdK%1=(mdeqMdB0 zJ@&kIdGpCmq>fBT#U3cHI~rN!Dh+f{HD|~^KY(XUQ^m>8TXh(qt{26`>}8QPVF(fo zZ!fV@k%gFt{VoXml;_vyNhm>+zfTQ4_G(B+PngS9r{y)KPY?EKU|AoWjs0%-il}+S zaRFBPu)Z=BQf=MnK(hIHIme7BBLW=cY&KPfWH&s`hkQYGEJ@qD9e?#-`%2FDtyi~< z1tyO!A5-Lena*_O&@b~ZB~9M6gYB^_Axtj;G!HnbJF4A9b6~_6J=NR81u(Gw`Lz*PFmB}D&`prgoXq9>4s;EGXk-T~cV|=gvS?x1puT6j2l}pdMlVvsG=qt2o+BEpLO|;oP z4ODEiZ2FvAooUu#p`Pw6?U^PJG z0B@yXjV|xz>IAyr^?n0-XVvJ-QT5JAGsdhO!I<#geQke#p5jEkr^M5=b4C2_)wm`A z?eKN)-Ns+iRFzUm{Vna1aHAwdCP-=l_Q4>NFZU)b|CCMzSJcN9cS%jv0fct7pRVCcXf1Ysfm{ShF!9vMGl-6aIkNZ%3v$_A{p{-@bAO(mX~(c1bc~1+qG{0KH~2bKJ+;}9L>es;|`=mc}&hM5NpS^%)`Djme^y1@_5^c+jf!NcMogRovZXz zN4G!63&jbg97(3Vnt*m%>#m1aCSM2Pao0r`gl!z1`4nS+3YkbZn*?)g8O3P}fAwyx zUw{^C0VCZNY(9z0C^y+m+n z4CEaE{T@}fD_2-9BoT?NGhqEJ=I&Wk^xJLiZ=_bS1yEmwK> zQSO-tw*7R!PAR)nX&!Y)C>KGm@GS{XW<=s^>5|9gB{y3PpGxCDcnZZZj_Go!qJ~Gx zYaW^6d~$fwR;Nz9Aw}Hu7tXf|-FkSmq0Qs_XwBQ8$!)fq z)emoq3q&j5VMJz=x_6%nul`TGNydSCUOZbYOKX1(P(}bj3EwnV-IFm)&(xYu8dJG| zql3EpI43oCS806)7JSQ-PrXz^;F|(e0qSgJX;CNYoZZF=*=7H&FQD zhxRXEVk**~^mfFbz8_59`0?4mq-Dx62|!!<`s{v|62p!^VWdFIdAdO!{1REMkjHbz z@Q(k%ZjRBmztIN!cIulZXRu|N1P{v#$8-M2hdYAeqKn7iCo_82JwLVOL<_R@AD?A6 zk&9X-J8q%>nP#bXng1k9F!yGD{p)+P0!GdUMkXgz9R3yykOdQf_Wk~0^UW^ZVU05EMUrW=2QEXRP_&upD8kiQ9dXw#!D#GJq{_STvAT4|< zsgt6~)VUDpdbD85*?K}w$dGln8}joagzr)coml9ypi^%Cb8Q$coOY{U`29<|4}Jn7 z!Y874(+{IGCDgPf@?G|2JLGIeK^9=z=WX8EPom$+&)pU&?ljge|B0v`qPuH|QU;uv zf4$V$xqj*1%L&#`aKg75i+RhFM=0?W?WXjxAY0fWPmEi#RG7|s9CeH^7kWVlOoykhYr;cl_VE>kAm%NSygY!7I^ zJ&G$$V$o~+rMtxgZE4%Rw_rVXWU5W{&8Tl?3%aE{j(lq2l7J^k&>PTq^4<7Fkicj7 zdx;Rw%VFXe)1Q4Y3W*7uc_MD-V79T!y_9szv1CO?kQTPuXLEsg(xTK;O&|;mpHGqA zTz$lB41x1~+;)_@dk-$b82>ga@(@E?E?w&aEOtR!@D(LmGJp|nBcO4MW6EgW(01^` z!|&{;9^jfDnLGf)@4ejTQnwg!i47>937v3RWcflM0^4U@4I0lFL4%1@n=XP5S{30K*0~y{-SK2!bzul^p0uqQh*3+>pn??= z`&6?}$^DJ!fIDx2{u|#}?ZJRXsa63nflhO2k3^@_aotX{=-9v_T29TEw-FkghFIgU z7;6R{S2|&RNEkP1oB9$XP}3KYg`0H1`nXq#%lpj?Ja=cO+>5%ruJVEDuAaMv?} z!_*a&b0~hCqu#HFuayo$F3u`>B9nSmmB>fahm+H~EuM$H(+9mo2F2ax9QsIyn~6)k zs-L{-@`0ZcUrq?nrTjU1`UnlO4c>!Ze*i>eKgxoCi~!ko`$pFkir;AW4xDIwzC6U& z;s59iVt;5=w%Me^zx&IC{c2S$F#jywk=S5Bgwb>Us}a3J(e%(^_zUu~{}4@&hvWJB z=zmrQ#q7X@nd!*I39RtR84+daN*~y+@&0g(azQMG@%YQcez5P)%*c@Qa@G)ZR~Xxr zt51&EbE{tDN)0!P!Xil6)9uC%ABP0K{#Oft-DB!~=SGyVuE4Fy{81{xn$pf_s8tM< z5F(T#4FP^?SdH0knzA~o({WjMl2||DsXKFCFkn_Nh-T}$U4^rq6K))+3pj(Mz$pPVixvSouwLG&<>~s0SXK~ zkiu9lJpfdy&IBAS2IP|8Rtm)M=OmE7iE}F3b>6!qGmD-oI6C1XKvjiXqlLvuF}tfD zwt`57W#dK{Xy0=E1u({ZK+aKNT)_-4pS(NkE0bnL-Wg!a>B#02y>9tfdn&o*=H;_A zuBPT;ssWp6(-Y9a$o}g~YKcC=OcDC9GXdLQEp%yL4Kj*e?0I(~>jc!hE-9e)M8qKj zt+s|51FZ+ppf5g9#5<`2=)X#4y9^=kA88glb2z>&vPFNX&-~|Cbq~NDN`0lmcz+!J zo$y>O0EjMAYPH^Fm@7+G&4GQ6;!#vUFQnKD+)q^GTmB%k4c-Ti)Ug;nw&Y`veQ$$3 zlwdf#EQp;}&kpsj&=)ZCKW>m7#;DI6!vdTB6dLmXpzf-?28E~^=YFYFhpoiQ;FHKp zmo6#pb?OU{j4ugiqE~A^>#m&Q_m{$^e0I%ldHHObx~rOD&JR6>6H-Sog}UCml&u3c z767Pt+uu!OOuPHI`MSgeBi$XoD}GA@@2?GL6~~|RlG@zyU)vElyy_lw*k+v%+_E^{ zwlEoM1ggD#68@T#QeZ^;+e2xA`}{n9E$`>F5M9}|U3jq@UG!LX%)x>O@?NWP+KNuN zpXasB8rw=8zrKH004Ed(?e%`x(0|wt)Ui*;^T!fu6< z>9;54&F2a$hY0;@QeB0uY0;gD*+)!SRmD%QimTJmT{oDhaktkg|Clg8+As&q)zmCL zTGv0`(l>!ZW@f*?Iaz&k>stT802P_6OF>~KYu)DBh?H%D!CN4EADG1dJJ|LjwVCaH z9Vnif5m}sM!-)12w-yb=Nwda~`Z6_o*l{fYEfKe{?{{ivMs?eZp(9ZkSWWRald z+9VHq`)k|!esd$RF+G48sdiR)M4Gzt8uf68`Hp5<{N8cUQoxZk98~NkO`Aj~MXg!> z{oxp13SvI^N}2ba{Oh1jDFuEIw%w>7mUxUVN3yNe3#LzM>uJjTp`HfLE!Q0n{^<60 zl+Ry3r8m51B|%#={Ci2%)wyC5`dLfCAX4|sN?#sAgSi+n#n0)>U>=r)*%tmu@u?Y3nE zc^GLPE=eBV%`X_n`v-4%-qw@>-hFVbqaF(G^*SjL9XAu|o;CR?Z2vHb=alnA3Jna@ zSC}3v->ErTsO1HX$17Z0PwKt@_*;7Kou02M4(=7WB*4=z$u`alWq!e~xKE=0^ZR(D zPKD6FDAbsHon*He8cb5D!l}d zVn;ecARwJcCv+*&L7D_mss-sqq^k5@lwJh`NT`8OLc)E*`QGooXP-01y<_aL$JqS0 zpS3dAoWC}QUtuKgelXT|2Kk5OIq^sDGc}jLWOVkHv3TOgCq;W-OM*GBREjAV?TLOCBFMsX)l>GNGbRWF7RN7wyc&!Z3LR+s;IPhPcrBnwO)XFn>fB-dcqAh7F- ziA*If;OAa$ILZ^1FE6xAX)ZW5#}YeL1)y3YZ2z>&kvR z2PiUCpl=&*4xKD=o_OPvbB^IRwTXW}mG(E`vNm;tMW}q7i0glpl`UU;f&37%CLfKp z#otYoHE=#1e2@k$dj;ldVSJs|(m*$_WB^VS4QtliL0T>Hs*uu8XElwVUKbQ zD7^`-hy?!UOKsf(DOsj7&xD`%6dFm!Ru^BOJUj`?zD#NBurtYalDy(V8Et>UbEC^3 z+kdAkYLMAZZIapW&+5M1sa*GeT@Y^;V0z#Tzqfxnd`Ync%lwr21*S6A8S(9~fw*w} zR+jhp`;&Ln>uYCgB1I%i!g|{93N<9sGiJDHKe?&Wx5#e)iw1YlJg`1)td^JcxZ)tu z$m~^+@_RT5N|(A)dT_9^Rz6tArD!K;8_91@-|RFdw%1Zv6A?@r*!?AVhfv z{%dHm0&{K2jzQ5AW2F{3p!;nqoD1rU)0lD0;Oy#mZDF1`*Jmw#*ih2-CE}8pco5veidCi z31GiCHL30QT%`yH#b!Gb_R3Jd+78=X;tPRNc7}fM67d180}%&p!fWv5g&QmPnUwC^ zbl$lhV!GRX$!a#qef|Y?ds}TA>ANpi(9d{bUBibsWtz+ZFyNv2#F{>r65H>(7=ZAel;=3lAuTLPe5_x}7}MkeIm zYlkM+h3`LMc%1Q(Nl#ReVSK(Sb$wb3-%>Y8|ue7T3qlu5dN42 z4tP^}czlRs{1$k8u)RH4;&KBZUni=E+S^9U9RMwT)wAMlCkNc5cO;2ym)2~MXPd}! zC(atWpa-6cEJ|f;wI7X7+wd%duy9fcc0f+(hFGbP9Kg_z9(O|QjVx5k+V#q7YoOrX zlX$wRL_<`5&XD-7wX|J|IJD;)ZxuBDU7FE96kIK@$Q4k7zfZi_-ho` zEgouSJ<<41FsW7oR*7efB|6^x);wllMeH{W2_$CQZ?1d;gHx^8qI9Pd&(sb$c=u9| zKlYTW!5fg43~XB=E$Pi!GpMank$_S|M_%yo@d|E0Dd*;u!H_GDAl(ix-2-<+Fh@Nh2umFq&u%)7MkRYPm>xRQ(n2yyn?u z;L^_y)OzjLgSpm#urM4I$Si~qL%^KT3cMNu+PEG$08x%2sO{?<^n}gzZ;&=bHa1hI z(+pXtEf2PST&`_R;W})r=#)9d9(4nZx!`0}B!D^i@Pz)@C1K3}fHPPj5=?oG{5c-D zeK_Gc7OwdZU02HyncX5E!h7;txAhdvE1{X8Yvh$C3}D>xBb|P%9_Z&fI>;vE_vCfR z=;Gl6lE2_cPnU=q?1KqBDCuO@IqfiP1IH33Jvs~;4g2vAyQ6C_lZ zj@UHDK#UHM`-CC^9T@T0cP~Cdxjj%_N$m2M0OApLvfAMxrd}5G%%o4?(Rt{oHPXJ7 zJyZVtQ-#@-^}gD35!bnrG&R2>LSEc)``!Ae!r~VXF2klVzELF)AwYQ<)^b9Xx|=)> zF5QacIv@tlpM;U(w_J7o)c<+bnJ~(oFW(g7B%{WC_ZFPXI1nMV`eftSHw;i{Qcq8YE!$%jm&?Mk-&!r#SaG7nW zyFSbKISIsC*#?ibPYyAMryZFC(r=}#Uy6YMpb`+!c+{cRf>v&?7t4{N`hr~USKr8_ z8D~n|Ys#9rZD5%;VOAj+T{kjQ?^O#D_#hrEJ3t<;^d4J2Qh0DlQ&TA^=Sz;ThGjCH z=<*mp)aPY{*n_~sG~gehq0*g`Fj>PB*m!>Gu%FPs<2f4h&&abRi|UL8`S(383QE`_ z$FsZ^%M-MPY|RSvEC6WZwzkX!9jU7f_QE#9Yu0VFKu`&$$ZsANYVtVf(fP=SIjMVm z*71VLJWBgUjR62^4LIEm5II2qz*`{>TdtUAO1T9@l`n8QR~yU$mpWkmnaJjQ;F9M~ zAU|KD#GDg%rM42G2&AO$leNp2S!Y3BFxtRSl-FTPACR=%rqTqqHi6CHfkgyYk)>r| z2ioXyS0jeQ?pq>dd#@iz9GDq0!_Jr_SaM9BTsw+BVSKcwlXBTOsMPQUeQ*kq^I5X7 z&gLnm{{Onjt^`K_*ZSlc*dvg91uTEiDYql5y3}#b(_+ctNFp%GBzLOcJ- z=TGYrcLW43I2KX{_1Xnf*4X5{I%6$huJ1>yFnO3IUEnBu`3%KGf%x5O;-idt zS3dEs*M?6jP5@9h`G=9bhLqcbw0Wfb+;j^^g>g19~rk z42L2EG8)9^&pj=#Ew9(w1~I?6CD`*@4rVoc3tR#iI;ejS`~5gM9yswIo;w915Mp&C%t5_DKos3=Vi9QxsOKKRbnn&yLWY- z3~Y=tO47Lp9>~(|AjeG7Jwb)aLlxVWlgBwX3@+Jg_+bg3{fx$Txi;N{#qVxuNN-NrWqQpfW$G8m^Np3#9YI*#b{F%60sIywGT>HkGoO6R+AxXPJYmX-P z^IwTt=b1N>h}~PI4dQ-423Ju~x`vs{K)oCg@HL^HEB*?fqI?5?xy~Xa&?6gB`eKt# zG%%ssV3*NT+KR-LnYLcg$JxsUZ@M)mGKIktKNt)|OrCs&p)(RV?&yEhS${l(ldId*9tek+?=AQUQ z{IA~s6e!0+fKFJx<0I^p#~Njz2(v{ySM{Qd5Mhg#?hA~>xR!A*$j&CbS%mWNFW)yd zyc%>_>DYA`y^qFK;BDl4?xm!t#TyNxr-izdawqqQnrE~no_tpRyL|hqc#o<3j{*iw-V9~*xpsRSO_YTA8^DpHUHde!w* zv6Jg%XlqGx<1CnKo9Ln(^YW(YL13uV<`_PTB$O_b>*BxeV3Dc*tCSR9HoKnh^?zjR z|M#1;DUAMkByH86IERT8c0&@>@|L6tgz9a*HN&OW$8f+X39s+%s zBP%}sS!XuLtN~$h`19$fiB$<=oFs_wbaq9efQpJu?JQOq<&jDmTK*DuHd8H|(WK^D zmY3c#>P=S}+wF>zw-Ost7?sk6Sa%4I7L3|n&bjBvlOU6`80qOa%Xl|F_*RIeOwrzD z<3gK?{wix!5aX0vx5%x5Qr%|7rKhcz1l%gaW8J7~cge{Y~!}tZ5G@OmDkkhe6#q& zD=L522=po!Fox|hpVu<3H^@*-4nMse-q%G8?34b<$Thkx(#cKFRss$lKBnvSxFVg+S-{5+|Gj^dx?*vRPa5&WAndKvk7F@p5BsntQWjl zmlP>WwUZ+wjEdA(Qe}AKRo0V8q%9WD=6&C*;EkeTQ?AE140n3J{T=NV;_XELAo)c= zRac{B_2!I!@5#A+#+0UGCHf(+$7NjYN>k_lE(ON@(#$nH?J@pej*YDCfdFGAbA;1? z>8a9(8bo&lBt!J4a_M#SR^-9r^*1965|hzO>Ym16R6Yr4e-MD8KO+q1>n|cad~>ts zPs?Ie9JY1@b)7>;#2p6iT)Ifl`I{J!!&x=3ZQZU|RHevW8H1kz(Q*L>COw|GVNTPn zr_}SEi9F;oqCY{<5gmKhmhLqbKmrb=}m@;hY6J5M-6c+FinJ+;=f?Ma-ks_9_DO%68rUYUa?zME%Xi{n`AK2n`wf{5$UEc!fm!)Xgrf zhqlJ{fyP~Z4WG~we!Qg*+#lkb7`~cGouydPtze;-T1;A(wVhsmI+eFhb3e)HFuSHu zFFcIXLx(qh-7IrpyCxZKgn1*A!=7f1E>s;~B489~i|Z_tCN+(Wr;k)hQs6VUkaxV7<|N(|3IpOPSTm3W6TmQhNE z_50E581EW3d#%l7E9|{OrJW%dr&9_Rb`Z~coWaj3Vj8&N)O_{|o-v0s`WML8((L`* zn-j^Uo$?$y*>u7dS%WUWJmg~jnu{Iz@xLIJ&@u_~fPS%v%$%;U# znLz@)!MCs~D$iQyh~N{-cb zz%AClLz>PBu)Q9$aCpeAXty~V0Ki-bAO_BWKWXPBvw3Y*CyZYX619x0}D&0-%~DO;-P&WU-7 zF|V@h1{qzG77rh4&j$*c5uMgt?{W7Dl}}3e+c~$UeBxifcIn_-)R00{zar|(f0{o4 z>g^)=DgW^Sb_#ZkUI-@B$PY1@z1Jr8yKUXYM=J0GXjuvB+#Csg%}agd!qyuBb84>b z&mzVl?4w$rWH#|raN{hvya23pV!vs>k+skKxx}Q^tvkhmQ!DHn zkYKTU;g=B6$<#5*x(VD)@_tT^gYmu3Uau8RJgdoKVZ2E3rr^Bc}9s^6*i#$A` zxkoJkmxbWhN2gO)yx?Tb1g_HZ!!;L$ixfBkln_e4ezY*<&wjH8#x#zxCHLcb?R)T# zqg;C<{ndN_KQ5yWNG4rKGDPvZ^XBv|uV~RfEJb96rr&97g%#R5&h@9;xZ5l}s7RT@ zNDky9AUaRB06_lE=COYG;q3EfV4T+dvHV1;^>Dc7ju~sphXYFmnk1`Y=re_AlsLN+ zj@?PTmCC9WF}1$@i0KXkG^ZZ{kmqGRv~1cB*TtNYHNQldY8-Q!P^)&j6a|7^>i`la z6>OJ?(yKa2IH`~@bAXJ33xMuemkb-M`XZZutwjeZN;HXo|8y_l;0LFE?&sq@;L%UD zb6{Fm+9?}|TLT(*Rtk9Yp;o6mZ{RF0*89;BiOfR2&rW)Ki^8-kvS{D$ernDo(UY%a zZCS{Vgrkg5YA&^aYk4c7Uq*@Th0V3oV3si`Nu=|H7NdQXD?1E?FBH{e1znq*h7G{QA%SB9$4qy*&g+Uz^hgLqRP_LO8=(l5&SEaYnm#l7^@w-mCBM z@XEH@2n7N?bu#9)zP%~DsI5+@0x!D-{>}L9+J}d}AuBB>;GfD*k==3b%}#ii+8Lia z`)GqqcfoQYf|@hb7?Q_M*Ka!Ypkn;!%YaK!qulbB@eS|g8j;qH2F;)G@DmpJ@7iIh zbLU!zuq?cUzAv)ForGHi-->5&`u+M?7!Y9jK`3 zRD2r%xks`ce<$sF1Q!PkD&myHQ9DrT&;y_dg-}a~D8CA}=)!$$HJ50x1odc=3idHe zYmM8iMBHUUaDnV&oeUZ+(aSyK3vv=v@-J2fcfT_4tF$Xu6TkG6H0je}$~lat+A&Vl zJF)%8;O!jGwG<8U{m0@^QmxjR{3ZZq-+wDVR!;8clo4S3%IL38J6mQ z7Zc9p3wa;I1JVAt#MdD~ECnLl8HeI`mzH|mhICRH^&9CPfVI#s@4HRA{|+zSl})-R>NJwHj+@+v0Yq8$id)WRTuyak)erv;7jBKp8L=ir z^LmrWQ{}0mM3IW&*XUn{5XNouWwm`U3ksHLxS*=^j$Q^b{lPS-HAU~$PYpp2qd?HE z^g35AJNIFfXVOk<=VglPlP`O%a)|}v8s^N2sJv0KVo6U!nNxxjSVX3_#RE}OA9j>X zx^Kxe#a@vk#erRhe(4)m?)?-a!itEI&v?kKdQLZ-rMmPky^Izp;b_%ma)SkJ5C-M8_g39gouQ6U3upx3W|F@SX z!|SEe7c1uwJzr)NLtboey!f8jXB5i??M#bY`n1@-d(5TrZ&CeCKA-0RyHh5>*@)q9hq$V5buf`NiVxeM#!3tc zSMPuBBy89(b_EDqbD3h`7#ooGb4Ll`3W}S}=?+7mNfkK;ADTUsKqLUlLVZd~JnX09 z^cSs15%Brv3a$rfis~J`39yanxg(;I(w}oUMhy5g^J2>^3_YVY3i)0fjXx%$oxcJe;4*LrftJ!i#`)q3{>$Zs8{ z^o+`sQ*2E2ZM*PFPX=PhCndt+h0-Z1DHsAejXEUGBAxodw9aR!;*b#&5+ZV9)z^HQsZueDq8&0Ysc_-F z<1amz*L6UEe4W$<_f%cJHWc|bQl@zw2^mp z&66S>bgrMB+%GY!mufs8s&-6Hk!0Z2Bdf)P`uwU_0|~W(Q_p$u_7am)L!^-iJJjQo z+!2x7j>X{)hKmV`YDKLl@2el@5X}LA+8NBkk9&Rv2WjQ0*Tdd@`$qBg-&>s|K(AkV z|2og2+0h@U8-sFMw+OeceeU;;L>?HQtpCB5VMsUV0QfaTYbxi6dC~_+sq z8W`>*R~#Ru$v%K9VI1mG|NgJ?`mb5JrI5^}Jj{NvN#;^wHG_VzO}y`wiIvphdmqC_ zyCG9=lg^hJGs1d5EGV}TCSl})_Kq~sxBc!T&0xTG=|I@oyuDueRm0TJWkD^RV(GIPcRyj?oX!+~UMiRMqI`Cpi z{ENPm29dgRtbwD2|mgkCvj5s1Q%>~mALtoPgadW^9RRX+cTrB)5kE@#AyZkko^w82UR=PXteB@YV8B`^-CL1?;Ql(bV zpnU{gN8*R*r>ZA_j&+7^uzA>Ux0xy?J$FItoYH3p9jF zWK~{QeH0P!iPP#sNB6Hivx7fBb+4FYNDpJ2 z{%bXMAv;WJz44Bx4MTwn*F;Vv&(fBOOXF=nT%R<5Y(l(5CNd5uBIX%d*PKuBKYGqOn^Y4N2~A?b|($8m*)2f5yxAyJ9Q zL3uLp{`KquQW7Ny_05EOZ4mYGJe5*9aea=op1*K)_=ayi?gim5G$?5QRYatsfO7bX zi=$|!&f2FVDK^pPYgn(N@|~GQF|4OJFa6E_Pn^TziQ4#hPBTd*Lx zm_)Y$;yN-J0b!x+EO`C#G4vQvz*O$f1qKdBYEM{(;bg)@5m^;A5HIn=ES?NmoN;)+ zoCrEg+=)lD+jo~(Ga>C9pDek|R-QeV_TZ>fu_NMX*@lsNJ1RSPnWcLK#$Z!c(-xR01=QCLW=cw_D6t~q_is}fH*aI9>odI=X)3Iw8Qx2j(hD9b z<_;fy$%$nlRLp7N-N3pi6O%qSd%r{7B8GDc&N&t?Oc>v)KD{r|ZpRX|P`(K_tChZUka|0aZ-l0iRFOti=__%k+dTI55BK4Ks*!KRDQ`; zP<&2XIpc&1j-_#z%;}weOki|`u7rWbdrXI#{_7C(6qtxbnC7w@Pt_`x;Le>GHCpWt z4Sb0yA~f=wT}$Jde_=BFGOwtuz=!HGsWw4{y|~bdw9DAS{jCVuJAi=!!oGYbHq1bZ zTj4zAfx?nXWLUyV72;VS`XdlfptkaI8Yer|-rMAM@+EtYJ*tUbXY9)Fc+dVS`Z<-L zFgup!TG}=7BAfe+1AvgQ=}h!GQ&`E)arbV|cmIvLjy-|e}yMll@Y-VmS3TSc^6J937GyI<_X#CY-mU_|knw*0QFiegPH0r;ac zP=7xK1=Zw;r9PqNsX;Rxu#_uVUkec@KC-wC)0umE#Xh4vi^)>p%z|O_(J#X)CyW!n zK{gsxKG{@vLCEc}^u7ga2knVGSeSdX55)ktbre0(_F!2wRG;ett}_6n{&DcL5s>7M+MV01Ci%CfH&P>DXS$C6gQl(sTiSNt3aT`W*dQJ)zv%YOU=2K{CGIS%`FUzXMs7J5<*D1cY|39D15ygP z#JxE|*Z+7xFlqe-m;~Q>IloVYaQ3ZE+62;V^&7Xw>|l>_NQ|s!?RIMczqj&HT~n;6 z|6;6kO_-1ZIhX{%I&!#|hY_nYw(UGO)g~jx<7vwcXh5BwAf?QtIccU==d!Cc5rg?i zNZ)Lkd`X(^!(y0KW`)v`$~Vz9+^!%0r;s4AbTdtJ28u-Sl&H7q@U0Qb3Q#?rbyfLe zC9Y_ehkfkgFwxT-FDyB#hU`rNL=jz>BFr%}4N^Zv@Vg@D>qtK?K#8F^Dsn$i zFZ2fo1j{a_-vZwXj>D)^Zb75d88g(rC_5*s9F@VlNhV$p9 zby}}%FYZBGmlsprLF2L1I)^wFGQQqu+5!uZ%~@iD+(_Vn+P?-OAfG@epOiFZ=G>VA zA`4>`p(+h@b2!A@vRIF4&BH~kvH9PADZqcE(vzW*gGau4r)Oox)dEzg60`>=YImsD zucZ{|py^{EZjfJ+Mb9hbWjE;z1lk6ofTaiH@T88!ue!XKVC94t0~PqobOHC6 zJMB~lY81y-HLzH|_!a3f!3kjEG+P7QFT&=HP&n!3N2Ep71eX(p=)^?tv!Dq@NhVX#-;ZFay<5YSd ztDk1iMz5lrRXqU+p-MeRVO@=fFfdM(HbZ3-i5018q)e)U+{C&E<`(LiG}yhQI?(!2 z3@82KD+gZ84YFt~HGL`;S=}4ln+y_qKiDd@Iz?_vd&ku>iv-=Dxd2H+ztlBNDl_M7 z<(Hidu<><|u{Rq{roLvb85eBESf}7gUHmz^1&8)HEpLWWMAr#evOe4LkrP!@Msmq8 zao_l#y2=}rYe4`z5rfOjjv!!u- z2Ur4O0)5CIr=Z^eGg5s7BRG%dZPu;_5H`G~zr-e|SIW>@69mz*Zl;q603$>n)wFd1 z@rdaDnyJfq@SO0n0fbyX{CY8L=hdR@!4PC4(yzlc`~q6*?aOtuQAP^Xy=ETZu`F#$ zVvl&_v+$Cc&j*b7ezbs3 ztcP4j&29GHtr=Ns;PnQDPd*GWt~w-lUl-K#p1sc+=y~U-@;sUrY=k)WbNOd^yAs?; zetXP$CUK{@Ik5awUyfG(P2O&QW5xVzruH(iLNY}2NkDgndqDfQE|pmC-(|q65|EUN z{^koe_AZvIl5}-*QiPhRY(m$e%onxqQ>nx*(Udk!`cRiM+nvZgx12do|9<(|9H%%h z=jK&X$ME07-fyx=M0B6g@sq7kq2gpcj466Jaog`OU=NUYw$3ZV%DAM+t3ibBZ?KlD z?ly`smOEMPWuMrk(Ou^ZJl^KjB933XChPt8=Lc;%eyX~%Sndf-wK6~9;ShX!pea`B zKCnX0nsY+aGrr%gJo8YuLwI-*f9TN+Km&r;C>{qT98(&PV;cgeO}o)5Y8mRl3rWTcr%HS=5rbGdXWHP{uj8H^>V0 z;iENuCS31OwhgT&Vpg&Q!IpK%vJS~OE4OS%GhaI+-aj*LIu{M}dIQ}xhwHehc4B{Z zf6sVo6%_?1IKawG{_d6o5C({ZdaF<6J{$dS*hT7Da>?6<_+I^2$>VP6>au6paQ_{fv3j|_MJ8Z5tEnb_ilZi*@J=?|Onw2) zvG4YvUVn@wq)W<5dnCLNwj#; zbxW&sIk(Y0m*?)eabC99qnGI$A~q7m#no`mCdQeo6>{J*i99pamg_(8{R-})HK|W# zWx^m8OI&GmZjQw@a(~BH#y8vou?=T#tTm-AE+m{0T$2sA#U+`=fHiJv`U6m0pDiGR ziq_;pUIIzhE7+oGeyWlfQQVKcR}zR?_wasb>lj?fjCnIj60awv5ninu@J&7AZn^=> z-8g`Npx?*FS~ISEP_os}jW1aGYAA8Iq0cmybOwftKfnjgp0eAkLWoRa2Pgw%1$UB}rl8O5JX!&F_ZN)R84+)vnp zbh>evaPHR)$ao+JzP*gI9m1^ZeoLTSAfTl@tee_f(#_F`T9 zfiyA~EelYr)VPfYouvh=ektXK5@BM1?|$_Hz-*X=T)Ze7sVm#r=HZaHpNx&}7gMZk zYflBm$I7W&L^flQ4Pu&Iq}SV=)I3sm2%1h=>wte5$a@87j7Qd^4UQzBHFZh+i@7{0 zf?6?87zM(~5$4t3dL-z=BT z7(1F7Dp|S>#Z7t?<0#V%l!z|?vkn~*Z{K0xGkMOx{m>&s693VxRATf$PwG`m09gjw zoUpNZ3R*D*ohgDu@KY5rWC9ygHcq?C)X$9k2+5s6xgyC3Nd(7(Z55g#;izQRF2*j; zYuq`y`s%1rdjeFEXO9O*!1=uhIS+AQPh_g&{5DXhuR2^mM|V6S&2Zemz#<;NKj%j$ zJYxv*e02u;yFHdoVoBt)kV6{i!`nyViKq{H?TNEo30Y8o*N*asJgZ^g20&tl1;DDz z&SXG6*+-5<)rKGNeTif@Z);M#_r9q~ISDnd0+gxmjd_%me#=xi~#ze3|2LgBnv;KfRWWcV(+e zIttxX6Wc$Bt(;-lQCgE+=BDyOPg9?ND>T7eROi&6yg39vMuqu9Y9pa>&z7mbNys58 zXB5NkgHg*{K6j+n@NN%ZE2qU<;71r$Wy+)bMyPKhGxspB{T^V0;w zD=TwxVw|CmkIDx+{U>ud5gRFW#sLyJBc;~NoIhP83}R8NPM4+_O{`DsEGjnSgf^sA zR2IKF+?(Ml?d|fHZ9UojbNwW_W=+=aN4ABz^a3m2%B%2dKR%MkpeJ-gVF3QP4uPT&*#C-i^t2A^S- z&RtiVP2860U^|O?G(Z^}wvH+qiI11LTjy|>Co&5ONb zs^e=#0m2GmkqfF z3|Fphqv190xD9I{g;y&0a+Re->2t1Zo5|ZwCGXyx1-vsSm&Wf+B72eBj({-eKaa~D z3LKYg&fU_FWU!4Kq&dKJt*#U+y|9!rqyfje4hwYLJ^JLG;`cN0SjM~#8XdhV3eF=Nd%KokY&ox;#!@h5FvnFHF zXFxI|i7B1ubH8hv7kpcdn};W|^-jwWNtKWr>$*j)74s0*FmmF(I>!zu zF(k&;FN*6{jlS~ZR;0$2!(Kk;zgGU*_Bfxca(>@w7s2kQ43ypD$}#NiF_8nZo)+5J#@%Z`*r7`Dfb-#l5}p|;ZzDJ)koYY&o1Qd;_muTO3p4}On8 z+Tn!fe(9SHq+t;hKc-77J&DZTG=2zK6+qc^9?b8mshtlZQIXGgc88a zny%mDyB-cGaE1UJ;3d7J^_(oAn7GKAmtSAeO+F_0{Yb{8*#izob?pje8?A6URocF( zra$$SZD#i=O~n;sCl#TQg7C+9{x5>Twl&M%L;-8Vqtr$A=rGBRmyd4KYSJZ7NS8{K z2+>Am`=rnh)+Ee#9cku=FTnu5rbsRK!WR9C1Oo{pCvSAj;m0VJTE?S3Wl&86_pkUl z!J9LJSOL_U{iMI8q~{F3&b7D6`t$$3V48AzWgdkldA5^Y?bzaSqm&gar)C5GAmm< zgt{6Qd}HI%s)#ZVZX>iQC$zPq`80Fw-%#$Sw}GEb=_KRY9w78C8v=8Q!HwM1hjnnx zz!LA|8zVpklc1r@mhk31kUer-g!8%e4^P-CA)LSoGCnA;D|6hC$bdDK?{4WqLEA@; zZ2N!GhZfSE4$VF2pI!&;c12NA0yz+T34 zq9QzwG~bN}X|}&ImpbiCmDpfm=4@MJL$0~cpOjS((M6X^lP2kavARf~MKbs6#z~nD z#h^0kZAenj$I@h{5Bvz_8?(#nT`Y93QQRFi+kUB{TgP_MJ|fT9Ua#|0ei>(|`R+e~ z;#m5=@n_F8yIx0p|I-Z~8E&UUPYc$}X|?vEmC%MFlIMTn8vb35lS)ZJt3%rb=AZr3 zlg7XB`l76X$?w4>xE>#ad#JTkoI*!j7^-H3=Ue;=r$kNr=F@#PG8^hGLs`)*`&%uS zD=NyJ*qH+}EZJr35%_4~0=NbFx$pWb&qPnIa-2vDytWt+Nay6m9$t8|ihP2jkuNsixnvBMw}K z!$uxT`tJq~v>|tV6+6_*%tqg+?qBsFDN*$22C6{eD_L({=N z-vqRiuBNU-Qqe(>F5V6edBwG8&*#qLc^m5`Vw9a)tOWHZeBzNS9pEdiWx6E(9DjFc zget?TgBhNaLQ1k=&K{omS3uf%I^A(gi>YUD4|_*VT?hEqXM$7aqwFZ0N7e3KqXDRV zRJ)?}gZ2VAcGB5jqUL3l*Gt(^Z$z>BN+M{CWv#+#+9h+Kf5HYwH#pjUCK=?E5pN%l zJvY<@Zwg#wn;Y*Af+g{g63c);!|W>uU(nWI0AF2l4QHM1cvkPWow`e>nF?IsuE8h1 zopqN$zneoMu7ag3@93wv_lV{gk1B^p{WF9~vlAA`;u}NMeQs|259CD^{hq%ORBpOQ zN|%_wUCu7rOCVy#qT4n6#J%ptDIvk%u9r+KmlMkB90Zf9kJa#gp${|u}T zKagIR720TG4!I4#-r=BZuTRKtZ*ZGI5LUJD1B@gsIS_}FMWn(2qqc*5Y{SF#b`~8v z&rW%{DJ*9jKFKD2A8oidHJMeTPPdAA<|wC9(-Xt5$g}fMJIai?MI&K3!(xa>B^n+8 za6Y!D=z&xJ%d+c{#<=?4ZTJiiDr${R3qz26Z~~-bBzwndi4#{rtupaGxogil?1Dk| z`$rVRO5*0r9~GP36xUh5-?9pbm0=D;C}fZv&s9L;)bFJs$}NJu9Y>$;!Q^`!pr+Q@ zIfxrUtse9P;@`ej`S|^RU+o1jFJrny8Zsj}kuhF)p==>GueO;1yz;}mGFY&XUb9m@ zmy^eKXTu`nJNM&U!{2;{$IwJ&o;oDZdrh3Oi)TKOjV+!49;?kWs~YKJZbo~af|vcl zZKD!yMJKhe@YYI2j);S9nKy!~kL;8OCz-pRu1Q!u%F%6ZnDecPJ+d5JEfuDu2v+Xg zf%o6q%bpif|5}ssN-JU`r9O}x(=z3(kOEy1vwUDnR672l)Jr05o&*fzd0tZ|`IsI% zV;@ZhPD(-LN@?#KsaLaXKIbo)^8R6N-%~;40g^J*7$xO^m-aJ_D;*d7Vj=yq*T&c{uTx7rv%3H6_ zsA~BHsM8FeduPo?pTCIeI$hg_HGoI2Uz(q2rs}UL1c{@7jn1o$jZc4*9!H!x2*-rFZnQU%XY*~`fA;uTs|ZoI#=3V zE<_3bHcTvae0kA94kYG2cl71*>dkt&@6I>&^glrpYl@uWK+R%YOZ6>5Zf*fc)vGuC zU8;-%L^5V5ACC)t6H+(P)wlo%`zh%?b*H3fi~USA-p8{NDG@fCusc7%cS2b2;2^7y zo(I*{$+&ZR#)rzc-?UMOIwaAO6HvgQPw;5j5%?yZ@1M_zf_i&OoSv79x~j>bzTMK{QxzZ~ zqA9(i*q)$v^qi;j2y7*@4x{VV+UlWG0@4L}Z`Q{|BMXVwv4Fu3^q+uPOm{ahJzYqEbF_hbZGwMS911S+6_BFH^@HPBEN+8$GeK0K8NG zthu{WFS#SDqAmGg+;1r=D3#36LPtz&o6CWq z%japl$S%Edo3ftYb?TN<=k2EJ48P7*UfKF%bd4&hP6Vm=hoB%LQwaT`?*@-g*r#v) z^GmeVCH}GMBNOkxABt?r)_euP)?@OG+wnn&oE*oW4^K{-@f*hi9|ikP4u;(4n3YxN z=pNmh{?uAh;Fr4`ahS=^$5)8pdtU5bJM9WHx^lG@HRjShP@97`i80RA;**|}{=J%P z%Lktw>uY%Na~ZvS^KB62g&Y6#uRjc;^CDy0FI=wNQ>Hj~<(?%i^%mz2_2pb)(=Q1C z>i#B(VBXP*d^Lb0DAp=kqhU|A4P<&cmp7mRt$G zL!s1Wqw;E5(46N<3N8a)E#21b^8d8=-Ri={Qmp)$Nl2= z;`5nvo$Gzx=Y7s~E`0%C88i1!J$LA3)zQ<%hCDk9I~GNqADB-%R%yz2x=kTSFC>cl z-%rX1xJ~WWWKs-;QP@nvwd*KTI^*rA*)c-wMxvA&txA5i!kWpT^?|C!;*y6;m)Dep zV@JH0Jx0^gV5Og?`qzKLO13&zp%da#smD3?ga0{Hz{9)0!1l~(-7G_gM;2*Ty1VQp zPRhRFP2cI|<>j+X(mTcBlZ=WgWxKg(k0EIz<`ODYI9KT5bI&`nvCdaEol0M=6((q= zWT9)WMA{SDuj9`jz0o7$5n`cpNK;uMq=sc!=XOm)m9Mam5=A`|F^!lzJL#xa_=dKl zq8=M1#rL*2kFHCb4<-f?)f*;jMSBYZ-E560Vy2z;-(awHchc?}O1f@IB(4GU^*^s$ zhKrCLBIpsHm?sD!SUy1)VbkdySEBYFv=VdW6AW>qpDkfI@v;?a=0k7UGITB65DhEy z+f!X7db*IkNlY`LIHS13ShYyIeu`dZjl5j!X^}hE15#RtI;(q<-QSWWPz{M9)$><- zK1onA0#;ja+0RcvVj^KQpmfJCr_{KHh~6o?vHIP+Jtx+`8!^EMOv%h;AnC4V=rCRt zx{Ig`?}HA?N?%F)o$ocBfs#vm^+;Iiu;uf?W6`d;Krxr;y+=zF?T#FL1deYmb=@)W- zLH3l}raI^Y!eP$>y^7AOBR8sEjrxO`T8!XXbPDZ3Q3Oo!3iAi?=DJr?)WJpb<-hFc zRR3`4*e`GUmoqBmh4O|3#^nS({CKAi ztF!91R3|-MJlWF{B!i|$=p00p)e7lUZ{ySQK{Xvh6i8lXv)Pgb!Y@j;*sP*#dMNj# zU&p&qZq>j=!-hSw7S0wE7G^3T<`~Shy+|C%8^bfrilx#SKBVB>H zA8Lg}xw!;KnR>VH5+>-dB$Zx15!hGGnHe+a;wPN0MNu_xuqDOGkQ4kjio`akM^rkB zr{ZkSojW)3;|KFNDU`mdj1xqiS7I#yw?KftXjTCaDlc*}3N`$pmW!OZ(G~{xYJc zO(@UJ#n0CHn{M8eH#{FXF5rReLN;%* zaV<0^h572VD{HNLCNzb*D_mW&rE28I_X5oe1Nq>l6vq1jNzClj*FJ*-LEg~}B=rH^ z->^v6uZ%uF&~sQOjyQ|A*+8_#1dz1<(j%{8_b-gZM3Asd-8dZk8D<8K(V~R2cj|-W^ovA{z$cQlf`I+8xNR;*ydh-4i{A?fG zM|zO(NyFnJ#oPJ10_ew17>a9@sH=5Uc}w@VX>Dt~?E3CmFu|f=fMP%8fIa+Q5Djed z#F}>*d%#tRlJ}&RnXXXl{OJMo9BN=ff_E33Vu)Y~(@|`TJ@4E@r`gM- z>U2>9T1lItC(sIeg;Li&uMzaDjxQ!?9~bEqQMwosd0_4Npx%qw&(EYRloumRVkULv zcAmptIcIpy`PS#kVp9V()Efy}<+%ZTFD97Rctc|@k!MGH8EPF#UQR(0>(gf_N(ts; z8wJe`Y|v4)t`w54TgZ zMt4+|VpVdcGp-ol*;@ROFQNS#^wQJKR{Vl5r7Jw?yll;hPT?TD_= zcP#eZNWAVoFOAwIAu!8WSeUxdrAHy1;xRXt5vJoBDDBjPcs1R*gByE$uHH;!zoRAG z(8Nqn`zP-e!#YS+6HG&5y{$^q7}XWQj44k%JC)RPp1s_xid-YBR=X=fqO^9xKazR@ z-+I4l2djV3`fv8|(?q95fI{>iyg>X-AsSW+{UP!e-Ol#&2GK{IA}fobYrc4<-ahA- z+)~y_-ebRC)};9Gxhxja=>f9YUfVLjG~W~5u8hhSdcIpY(#-NyU7vS23;AK>6RsiT zspr`hNLKWl;7Nvbx-)`lC>Ey`sEGOgQeU^7PWr&M@7h3)x`o@Bvclz*sQ)BWh&UFI z(kM%>H`YzLth-n5mcvMDA2d%(JjRblKST z!sZPQJ}>2jP<2S!d{FfW-aI^n9jmePvdboE(!Hpl7<`ed>Dr#6zR8Tkt9o!1B&%`m zjbp>ZyNrNd@!NGwRtpL(Z@i}$-7?hpf*8*-lvY_Ga1lvW3g@XN@X4B2r@DVTeom!9 z;PnG!Q*@Zb)d*%xME0biIc8s7;cd-Rv$!KxiIF<3+VsXr)lb5_3ioPg26f0F*G+=N zV?HF7VH{a$qcegkEQ>=E>yZ>!zI}!XRP@PnllZD&{Nw(LzWz>9rS~BFw2V^hN|I)D zlvLwU;;j%1tg%a<@SRxuB$;oUtmlMrN3#TW2TY9X3Z^W-3<(< ziv<-wGLsTf_eJT+pB@@U$1g}`d`G;U?65m)O!Ti>0I&iSwW zQc%f846x*dp4+}CGzIjc@t;(JsHIpwF$!I8td+vJAe58xQZe;F(1RT zqk=MdKkmIj%)!adr(tNKcf;Bk!@7|PxY_*;;k#QN|GjOKlqwgJ0G2*cE2|8?s%nI zbUx1YUBud=ufn~$wF8pjY%RBjhdl&_Y8>-*T8lC1N|Y8DWdEeUg2HLvn7{*u2?Dx@ z6~{g{bkmi8yTva96#{c~s<$A~eg|zq=B_k}~=3L5yz`?IfD+1)?a(elFv6avU-zG$l2`8 zFFh;772JKTbfu3JcxT>#hQWrNA{05DYR8Xy>HrDvCa*_H7)n#{<_$D7Yvyn+2aAbu zJ#RJrl7o3;hoMWP84j>dz3Un4qh0Aoy$L4u7-;=y%a^lrbIT;1n_HWZ_hgBOeAEL6 z(eraAHdz4mLyK|1*Sy0kTI@qa^2vAmEPq;|g)WTL61y&RYp=1PVybxD zbYPrL5B8~m{vjLaSUvKl|A-;le7wJ0c(#I4gPZp$d4%tNJSs?$GgmlX6Bt=V#2rlO z(7f(i?HwN6a5|b9kHQ2Hp5)E+o4*#u*RaE;nm z`#-sb_LJmaB<}g^ufOgh=eMZk_f95Xx9Fs0FEv7LDDUErOAOE7K2qf-Ad>fTe;P`a zv0mVM-@;U_Z8;(+`{NBSEv=@D#>z(Byd;Hu149#hxBa9<@?96J{H8Yf=aMZNqn1>V0)71^oz#`j1{Pf4r+ zrKGLTzqV4!%{sJyLqfgtsm+UUbir@Z|58syaLL?wS8|*KJ}>&Xt0Cc~e>r~hl4Qs^ z7jhwTq1+I|_BS7UjC}55#HZIji$WZSFdv85lYJx;_~%PVhq8R+zMK*Da9o?1A5hl* zEkyH){PvAMqI<809F9oLzPj1u`e-a_g!F80xK*`!IP0mL)M^*v`qj3N@l{N>Er@7} zKsC#y&H4$sKYMFj6wjpY7~K`vQFwF?vjiRfKHxx10hM@3`WwDx{$W4}p4P{H5xl7D zbY-*LV(g7z#kM$~O-uuw1~17tUA5ZKMB8shMn)y}g=9S={hJeIozK*(dgS%b?#40? zq#7o{<=^>p8%)B9US+R6u#?D^Rbh~^rz2%7cgN9;g%K(Wx z8zcnYy0mDn5jNa*SJgRtD1W-p%OWixHFDZ4#!#&15e#p)ZBR692)s<)jaR4pAHknP zKJG6=kpT=SLJeCQqf<2-G^qxZiJ%V!Cu;+Bv*-QK{!lR&p$zs_XhlU?=4p6bzx96K z<&etGcQE9B`olzbcXt7tr>%MS$js#d|6|~4N*A4b?V-s*+yBq>tW_aIff+rY7%Rh! zVh z(^UI0CrN?SXC=ZTFP%hoI%I!#D}>uZb<@!*r(6rSszeD`x;uPQ4z5$0+k)BjvMJGK z8{Y@d@vOX@s@8Q%sfsSuRuC!2B>D*rDz#Nas*Lu9@YOwyeF(nPU7sZ>!VRi;%7e$L zUeSwcxh{)G#a_+_$$ejam@Tz|BBd|?oE)R67fzatu_4}NI$phklR+gW%9|8dLe0*R zj+MTiD=^8B%2KrXVfc2k&))B=1ZHaO>eaUT%$8_vCh$&Bgy-`Cmj~ zC>k))QSVTN^>0J;q7aX-?#-qfhOqHIU&(@|7`g?W98E@+J2XnGhOc+%l2+*qlz!7p zQm&pvK5=2T9aETz6p}YpuX}3tTxO%(n@uW5U2RoP8gE?u(ifN=9b=o@)zwvbOmlDI z5JcgStyOw*3N0FS-U@9K_9B${K34=2tCqexI9tPQ00xyrO<^_M6@>#!$N zKXz+9PJP=KEN=I`lG-H_5-YkG@y5~b%B{U9a4OU>KDLrRPjSmDo;za?mM!BA$v1d= zSM^v^*s>K(MLmS%;sHM5ohR+xDxfa@2ECu4Xi^BfA!{>r^l!P&X=f-;q-fYdS6A1- zLfNqDoMT1G-;C2Gc}=rji26EZN^HV~gKP@Q;x(=wVKtX!xP^AY*4rhVG6! zHsMFGpYio+DU@E@?i{>5Ta@NYx zCp{}m#lJLd*=i|u<}uq2Qj34)lF)F(vMRLl=u}1SZzD~_Baq|_WQYCWfnuSYQ30Aj z9FmB33WB|MO+aUTc==N%aALe z7_FN!(@am(eH!{)ov#vEjeJ}DiB*oSi@S2xY4|zaOEmEgpRlgh@KxM$*Uq7Q*d#|<$wMP-Com0 zg@v|iTi!l8ZTRYCPzqlA_YU|O7o2mE5GUg01{p_w#|h(up`pe);oD8MqJ34X;o{)^ zpM_d4D*U+#AL$z{z$?Pr^5tTCwX znp${3U0-Ftj%#Nur3NIjIE}j9a?;nQS5s0To>tV&U!`vA@d9$rip8)s>u_kXltn^Z z1-r71T>K}Ju|C1{o8f$Zl~{ru@xa4=QXHjEOE|;=v{Y~-y$YWt`v=^|N>rnlsN7fi8&$$?3kq%B9ZHsH ztykz(yxCWwq`)UsD!lvo)J-9y=rmQn3*kk`FW1xc(`!iKTTL16b|3d3>vr=!D~^t* z?edj)W4P1kj^L(&INCauOM&VlS;v!{Z5+QjHN742* zqbcHclyy40P3k1ZrsUEhZ}krE+Hhxez?c#N{f)oUz%&L~5paWkzZ?@06mhB6hc_Y= zG(Mq|1GDpAkru`?Cpg}4=0xbUdCIQGvDh_zj;MseNQ!T2pQI3lmn~&$NUJH?luC+_ zuArr$l-N?DG)*z$TWKuZRU6E=3zNfkQ&ZVFr9JV_ew~lHwcLUe=`Lj*w`Qbxq zKGXIoK@M-o)UOr6K_|XRiG9;(X7o}Sm-z7P`54Daf{}PaR+PX6Nh;jK^0=a+!f;zu zY-NaMdjIHam5`UeLE~Q{V3i}3JcyPS^|Zg~Cq9hl4e!QvV+-s1OWUwgV|}iM*GSiM zSPm`L!tZKlU#;(_tQW}MRPMt3EJnT3`VbQBS~XsYHydh$4b5%DyV}r za`ZkUGbBBo*3S}D(s;>-jg6Y#gH5@V8#TR}yk$At%Je~EM9m(J4-a`7v}*+rhUNh> zt)+36{xy16LP)MLwIR6i#NiuWSu%Ql@=`{ZwqKjeE=BwHny8&XE9}9>Mo(+XePhG; zCtO!CR?cZ5;uZS{R26cI&4W_BjEX+ZG%^+{9SQ|Ck__8`lIn~R3UWVMS*|0WcNAVG4!f@mPnHZ)) zgSw$K*>!;Vc3w%au$YxU=~@bP0&EIt^wMzu)`cey@?!4RX}?(Spt@nK;QB^>ny@2& zL@aCE;nN3CmW!--kpRtg&^TSVZp|2CU;D`Z{rdx}c~Pqq z{}f`pc_&CvSYZQmTMv}^Qya|QS%0n%)YMP;lZLWk8}(C9S&gL0@sn6Eska0NM8eL_ zE_ZxdiqA|apz5(*IFvfL6BhR{kv90e^T|O2F?T)qC-hi$pmE_qQnonw-7yD*tUOUL!SCVoL z`Dx411yrflx5!%@w-tVlOMbshrro{hfU_lBn5cDa8o@Y%!{IzVJl>nRg#n1!MZf1M z@W?C*tfB6G5v|Qd8<^Zq4 zMYxDaTO7EmRxKp?%}@bu0%7Sed`vywPL-wmD9mG}#DsP7N11>7+0U<@Z00>Wj}Bg# zd0p$;I6{n{tfe&a%a%qyM>mRw-V&XTn2Vq{?w7}qo7K=q=bQOu)wWc>^U_QYZQwnU zzZ$LbLcj}zY`NIP!)s-XKT>b_EGt(-@q#FRqCO{u@BBGuQS>)s;xdJRn%!$n zgY|W7vito1U%v9wrnsZMaI$W=CCjd)xI(YUbs)b_o7_)GJ-mu3@K;7W+SN)hV1DX! zg%oO8K#UFNV(=38VX`1Ysg$p#hUbJFAKsjgZJ$K;#>`txp6FZ0Zza<9%KMlPx#8Vt zN&MkZjYYVex#atRNt;Xyu(>tThKX63nVBg(*EWfe?8Ln`fEUEvGD`33LTw>ZqIWnyQrGi-t*GU zdd<8?D@W7)W}L7>wv$3}cn#xKv6aX4_@jwC!`X0)+rS?Z5_^ulI=dImoY=JhXWVhE zD-aK^@7B})$J$y8YxFLPY$j$bA>&Fm`n;Gy@s44OW5IRy`Q#4^_FrY zhwfA!Fj>}7Fq9dv#&r@qt|At9&@eARkB30}{@wk-tiGJev|MrKG7~Ea=g#LRcdc10 zD=j?~6clu{k`Sy`iqP<=0^h17%c6|obhqP{Qe7qtiZA)j6uPrV;`n=~2J+>gM@!6s z=WN$ZrFqFEuVWI{Bv7MG1<@z@$!~Tn^v?Rg0e0?qPbG2jlk0hY=N5Dh%QfCuW!7QV z+A_rR>xD+4SI=EC96+MD`BR}mXcz@e^h{f*{fHJzq>r}PRD2^fu;XC zU~*H@fSWD)0`w7{pPw3Zl#`RIEc4M