diff --git a/doc/LectureNotes/_build/.doctrees/chapter1.doctree b/doc/LectureNotes/_build/.doctrees/chapter1.doctree index 8b96f9279..1771172eb 100644 Binary files a/doc/LectureNotes/_build/.doctrees/chapter1.doctree and b/doc/LectureNotes/_build/.doctrees/chapter1.doctree differ diff --git a/doc/LectureNotes/_build/.doctrees/chapter10.doctree b/doc/LectureNotes/_build/.doctrees/chapter10.doctree index 4b39a1bcf..55719d459 100644 Binary files a/doc/LectureNotes/_build/.doctrees/chapter10.doctree and b/doc/LectureNotes/_build/.doctrees/chapter10.doctree differ diff --git a/doc/LectureNotes/_build/.doctrees/chapter11.doctree b/doc/LectureNotes/_build/.doctrees/chapter11.doctree index 22dd46216..ea8463cf5 100644 Binary files a/doc/LectureNotes/_build/.doctrees/chapter11.doctree and b/doc/LectureNotes/_build/.doctrees/chapter11.doctree differ diff --git a/doc/LectureNotes/_build/.doctrees/chapter2.doctree b/doc/LectureNotes/_build/.doctrees/chapter2.doctree index 6b86b5068..932c05c39 100644 Binary files a/doc/LectureNotes/_build/.doctrees/chapter2.doctree and b/doc/LectureNotes/_build/.doctrees/chapter2.doctree differ diff --git a/doc/LectureNotes/_build/jupyter_execute/chapter1.ipynb b/doc/LectureNotes/_build/jupyter_execute/chapter1.ipynb index f2ca50cc0..965c07e22 100644 --- a/doc/LectureNotes/_build/jupyter_execute/chapter1.ipynb +++ b/doc/LectureNotes/_build/jupyter_execute/chapter1.ipynb @@ -343,7 +343,7 @@ "outputs": [ { "data": { - "image/png": "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", + "image/png": "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", "text/plain": [ "
" ] @@ -515,7 +515,7 @@ "outputs": [ { "data": { - "image/png": "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", + "image/png": "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", "text/plain": [ "
" ] @@ -583,10 +583,10 @@ "output_type": "stream", "text": [ "The intercept alpha: \n", - " [2.00955125]\n", + " [1.99194201]\n", "Coefficient beta : \n", - " [[4.94419718]]\n", - "Mean squared error: 0.31\n", + " [[4.85108001]]\n", + "Mean squared error: 0.28\n", "Variance score: 0.87\n", "Mean squared log error: 0.01\n", "Mean absolute error: 0.43\n" @@ -594,7 +594,7 @@ }, { "data": { - "image/png": "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", + "image/png": "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", "text/plain": [ "
" ] @@ -822,7 +822,7 @@ "outputs": [ { "data": { - "image/png": "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", + "image/png": "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", "text/plain": [ "
" ] @@ -838,7 +838,7 @@ "name": "stdout", "output_type": "stream", "text": [ - "0.005000000000000001\n" + "0.005\n" ] } ], diff --git a/doc/LectureNotes/_build/jupyter_execute/chapter10.ipynb b/doc/LectureNotes/_build/jupyter_execute/chapter10.ipynb index 7b950da52..0a4e596ce 100644 --- a/doc/LectureNotes/_build/jupyter_execute/chapter10.ipynb +++ b/doc/LectureNotes/_build/jupyter_execute/chapter10.ipynb @@ -869,7 +869,13 @@ " 1.10378326e-04 5.08318298e-09 2.03256632e-04 1.92507116e-03\n", " 9.84443254e-01 3.11507992e-04]\n", "probabilities sum up to: 1.0\n", - "\n", + "\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ "predictions = (n_inputs) = (1437,)\n", "prediction for image 0: 8\n", "correct label for image 0: 6\n" @@ -1077,7 +1083,7 @@ "name": "stderr", "output_type": "stream", "text": [ - "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_1323/953065564.py:4: RuntimeWarning: overflow encountered in exp\n", + "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_6316/953065564.py:4: RuntimeWarning: overflow encountered in exp\n", " return 1/(1 + np.exp(-x))\n" ] }, @@ -1655,7 +1661,7 @@ "name": "stderr", "output_type": "stream", "text": [ - "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_1323/953065564.py:4: RuntimeWarning: overflow encountered in exp\n", + "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_6316/953065564.py:4: RuntimeWarning: overflow encountered in exp\n", " return 1/(1 + np.exp(-x))\n" ] }, @@ -1673,7 +1679,7 @@ "name": "stderr", "output_type": "stream", "text": [ - "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_1323/953065564.py:4: RuntimeWarning: overflow encountered in exp\n", + "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_6316/953065564.py:4: RuntimeWarning: overflow encountered in exp\n", " return 1/(1 + np.exp(-x))\n" ] }, @@ -1691,7 +1697,7 @@ "name": "stderr", "output_type": "stream", "text": [ - "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_1323/953065564.py:4: RuntimeWarning: overflow encountered in exp\n", + "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_6316/953065564.py:4: RuntimeWarning: overflow encountered in exp\n", " return 1/(1 + np.exp(-x))\n" ] }, @@ -1709,7 +1715,7 @@ "name": "stderr", "output_type": "stream", "text": [ - "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_1323/953065564.py:4: RuntimeWarning: overflow encountered in exp\n", + "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_6316/953065564.py:4: RuntimeWarning: overflow encountered in exp\n", " return 1/(1 + np.exp(-x))\n" ] }, @@ -1727,7 +1733,7 @@ "name": "stderr", "output_type": "stream", "text": [ - "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_1323/953065564.py:4: RuntimeWarning: overflow encountered in exp\n", + "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_6316/953065564.py:4: RuntimeWarning: overflow encountered in exp\n", " return 1/(1 + np.exp(-x))\n" ] }, @@ -1745,7 +1751,7 @@ "name": "stderr", "output_type": "stream", "text": [ - "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_1323/953065564.py:4: RuntimeWarning: overflow encountered in exp\n", + "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_6316/953065564.py:4: RuntimeWarning: overflow encountered in exp\n", " return 1/(1 + np.exp(-x))\n" ] }, @@ -1763,7 +1769,7 @@ "name": "stderr", "output_type": "stream", "text": [ - "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_1323/953065564.py:4: RuntimeWarning: overflow encountered in exp\n", + "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_6316/953065564.py:4: RuntimeWarning: overflow encountered in exp\n", " return 1/(1 + np.exp(-x))\n" ] }, @@ -1781,11 +1787,11 @@ "name": "stderr", "output_type": "stream", "text": [ - "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_1323/953065564.py:4: RuntimeWarning: overflow encountered in exp\n", + "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_6316/953065564.py:4: RuntimeWarning: overflow encountered in exp\n", " return 1/(1 + np.exp(-x))\n", - "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_1323/1630775253.py:43: RuntimeWarning: overflow encountered in exp\n", + "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_6316/1630775253.py:43: RuntimeWarning: overflow encountered in exp\n", " exp_term = np.exp(self.z_o)\n", - "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_1323/1630775253.py:44: RuntimeWarning: invalid value encountered in true_divide\n", + "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_6316/1630775253.py:44: RuntimeWarning: invalid value encountered in true_divide\n", " self.probabilities = exp_term / np.sum(exp_term, axis=1, keepdims=True)\n" ] }, @@ -1803,11 +1809,11 @@ "name": "stderr", "output_type": "stream", "text": [ - "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_1323/953065564.py:4: RuntimeWarning: overflow encountered in exp\n", + "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_6316/953065564.py:4: RuntimeWarning: overflow encountered in exp\n", " return 1/(1 + np.exp(-x))\n", - "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_1323/1630775253.py:43: RuntimeWarning: overflow encountered in exp\n", + "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_6316/1630775253.py:43: RuntimeWarning: overflow encountered in exp\n", " exp_term = np.exp(self.z_o)\n", - "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_1323/1630775253.py:44: RuntimeWarning: invalid value encountered in true_divide\n", + "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_6316/1630775253.py:44: RuntimeWarning: invalid value encountered in true_divide\n", " self.probabilities = exp_term / np.sum(exp_term, axis=1, keepdims=True)\n" ] }, @@ -1825,11 +1831,11 @@ "name": "stderr", "output_type": "stream", "text": [ - "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_1323/953065564.py:4: RuntimeWarning: overflow encountered in exp\n", + "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_6316/953065564.py:4: RuntimeWarning: overflow encountered in exp\n", " return 1/(1 + np.exp(-x))\n", - "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_1323/1630775253.py:43: RuntimeWarning: overflow encountered in exp\n", + "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_6316/1630775253.py:43: RuntimeWarning: overflow encountered in exp\n", " exp_term = np.exp(self.z_o)\n", - "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_1323/1630775253.py:44: RuntimeWarning: invalid value encountered in true_divide\n", + "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_6316/1630775253.py:44: RuntimeWarning: invalid value encountered in true_divide\n", " self.probabilities = exp_term / np.sum(exp_term, axis=1, keepdims=True)\n" ] }, @@ -1847,11 +1853,11 @@ "name": "stderr", "output_type": "stream", "text": [ - "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_1323/953065564.py:4: RuntimeWarning: overflow encountered in exp\n", + "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_6316/953065564.py:4: RuntimeWarning: overflow encountered in exp\n", " return 1/(1 + np.exp(-x))\n", - "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_1323/1630775253.py:43: RuntimeWarning: overflow encountered in exp\n", + "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_6316/1630775253.py:43: RuntimeWarning: overflow encountered in exp\n", " exp_term = np.exp(self.z_o)\n", - "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_1323/1630775253.py:44: RuntimeWarning: invalid value encountered in true_divide\n", + "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_6316/1630775253.py:44: RuntimeWarning: invalid value encountered in true_divide\n", " self.probabilities = exp_term / np.sum(exp_term, axis=1, keepdims=True)\n" ] }, @@ -1869,11 +1875,11 @@ "name": "stderr", "output_type": "stream", "text": [ - "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_1323/953065564.py:4: RuntimeWarning: overflow encountered in exp\n", + "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_6316/953065564.py:4: RuntimeWarning: overflow encountered in exp\n", " return 1/(1 + np.exp(-x))\n", - "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_1323/1630775253.py:43: RuntimeWarning: overflow encountered in exp\n", + "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_6316/1630775253.py:43: RuntimeWarning: overflow encountered in exp\n", " exp_term = np.exp(self.z_o)\n", - "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_1323/1630775253.py:44: RuntimeWarning: invalid value encountered in true_divide\n", + "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_6316/1630775253.py:44: RuntimeWarning: invalid value encountered in true_divide\n", " self.probabilities = exp_term / np.sum(exp_term, axis=1, keepdims=True)\n" ] }, @@ -1891,7 +1897,7 @@ "name": "stderr", "output_type": "stream", "text": [ - "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_1323/953065564.py:4: RuntimeWarning: overflow encountered in exp\n", + "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_6316/953065564.py:4: RuntimeWarning: overflow encountered in exp\n", " return 1/(1 + np.exp(-x))\n" ] }, @@ -1909,11 +1915,11 @@ "name": "stderr", "output_type": "stream", "text": [ - "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_1323/953065564.py:4: RuntimeWarning: overflow encountered in exp\n", + "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_6316/953065564.py:4: RuntimeWarning: overflow encountered in exp\n", " return 1/(1 + np.exp(-x))\n", - "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_1323/1630775253.py:43: RuntimeWarning: overflow encountered in exp\n", + "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_6316/1630775253.py:43: RuntimeWarning: overflow encountered in exp\n", " exp_term = np.exp(self.z_o)\n", - "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_1323/1630775253.py:44: RuntimeWarning: invalid value encountered in true_divide\n", + "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_6316/1630775253.py:44: RuntimeWarning: invalid value encountered in true_divide\n", " self.probabilities = exp_term / np.sum(exp_term, axis=1, keepdims=True)\n" ] }, @@ -1931,11 +1937,11 @@ "name": "stderr", "output_type": "stream", "text": [ - "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_1323/953065564.py:4: RuntimeWarning: overflow encountered in exp\n", + "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_6316/953065564.py:4: RuntimeWarning: overflow encountered in exp\n", " return 1/(1 + np.exp(-x))\n", - "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_1323/1630775253.py:43: RuntimeWarning: overflow encountered in exp\n", + "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_6316/1630775253.py:43: RuntimeWarning: overflow encountered in exp\n", " exp_term = np.exp(self.z_o)\n", - "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_1323/1630775253.py:44: RuntimeWarning: invalid value encountered in true_divide\n", + "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_6316/1630775253.py:44: RuntimeWarning: invalid value encountered in true_divide\n", " self.probabilities = exp_term / np.sum(exp_term, axis=1, keepdims=True)\n" ] }, @@ -1953,11 +1959,55 @@ "name": "stderr", "output_type": "stream", "text": [ - "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_1323/953065564.py:4: RuntimeWarning: overflow encountered in exp\n", + "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_6316/953065564.py:4: RuntimeWarning: overflow encountered in exp\n", " return 1/(1 + np.exp(-x))\n", - "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_1323/1630775253.py:43: RuntimeWarning: overflow encountered in exp\n", + "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_6316/1630775253.py:43: RuntimeWarning: overflow encountered in exp\n", " exp_term = np.exp(self.z_o)\n", - "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_1323/1630775253.py:44: RuntimeWarning: invalid value encountered in true_divide\n", + "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_6316/1630775253.py:44: RuntimeWarning: invalid value encountered in true_divide\n", + " self.probabilities = exp_term / np.sum(exp_term, axis=1, keepdims=True)\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Learning rate = 10.0\n", + "Lambda = 0.0001\n", + "Accuracy score on test set: 0.07777777777777778\n", + "\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_6316/953065564.py:4: RuntimeWarning: overflow encountered in exp\n", + " return 1/(1 + np.exp(-x))\n", + "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_6316/1630775253.py:43: RuntimeWarning: overflow encountered in exp\n", + " exp_term = np.exp(self.z_o)\n", + "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_6316/1630775253.py:44: RuntimeWarning: invalid value encountered in true_divide\n", + " self.probabilities = exp_term / np.sum(exp_term, axis=1, keepdims=True)\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Learning rate = 10.0\n", + "Lambda = 0.001\n", + "Accuracy score on test set: 0.07777777777777778\n", + "\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_6316/953065564.py:4: RuntimeWarning: overflow encountered in exp\n", + " return 1/(1 + np.exp(-x))\n", + "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_6316/1630775253.py:43: RuntimeWarning: overflow encountered in exp\n", + " exp_term = np.exp(self.z_o)\n", + "/var/folders/td/3yk470mj5p931p9dtkk0y6jw0000gn/T/ipykernel_6316/1630775253.py:44: RuntimeWarning: invalid value encountered in true_divide\n", " self.probabilities = exp_term / np.sum(exp_term, axis=1, keepdims=True)\n" ] }, diff --git a/doc/LectureNotes/_build/jupyter_execute/chapter11.ipynb b/doc/LectureNotes/_build/jupyter_execute/chapter11.ipynb index 89d4c5efa..5f7060cc9 100644 --- a/doc/LectureNotes/_build/jupyter_execute/chapter11.ipynb +++ b/doc/LectureNotes/_build/jupyter_execute/chapter11.ipynb @@ -3000,13 +3000,14 @@ "File \u001b[0;32m~/miniforge3/envs/myenv/lib/python3.9/site-packages/autograd/wrap_util.py:15\u001b[0m, in \u001b[0;36munary_to_nary..nary_operator..nary_f..unary_f\u001b[0;34m(x)\u001b[0m\n\u001b[1;32m 13\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[1;32m 14\u001b[0m subargs \u001b[38;5;241m=\u001b[39m subvals(args, \u001b[38;5;28mzip\u001b[39m(argnum, x))\n\u001b[0;32m---> 15\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mfun\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43msubargs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n", "File \u001b[0;32m~/miniforge3/envs/myenv/lib/python3.9/site-packages/autograd/wrap_util.py:15\u001b[0m, in \u001b[0;36munary_to_nary..nary_operator..nary_f..unary_f\u001b[0;34m(x)\u001b[0m\n\u001b[1;32m 13\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[1;32m 14\u001b[0m subargs \u001b[38;5;241m=\u001b[39m subvals(args, \u001b[38;5;28mzip\u001b[39m(argnum, x))\n\u001b[0;32m---> 15\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mfun\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43msubargs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n", "Cell \u001b[0;32mIn[9], line 61\u001b[0m, in \u001b[0;36mg_trial\u001b[0;34m(point, P)\u001b[0m\n\u001b[1;32m 59\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21mg_trial\u001b[39m(point,P):\n\u001b[1;32m 60\u001b[0m x,t \u001b[38;5;241m=\u001b[39m point\n\u001b[0;32m---> 61\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m (\u001b[38;5;241m1\u001b[39m\u001b[38;5;241m-\u001b[39mt)\u001b[38;5;241m*\u001b[39mu(x) \u001b[38;5;241m+\u001b[39m x\u001b[38;5;241m*\u001b[39m(\u001b[38;5;241m1\u001b[39m\u001b[38;5;241m-\u001b[39mx)\u001b[38;5;241m*\u001b[39mt\u001b[38;5;241m*\u001b[39m\u001b[43mdeep_neural_network\u001b[49m\u001b[43m(\u001b[49m\u001b[43mP\u001b[49m\u001b[43m,\u001b[49m\u001b[43mpoint\u001b[49m\u001b[43m)\u001b[49m\n", - "Cell \u001b[0;32mIn[9], line 48\u001b[0m, in \u001b[0;36mdeep_neural_network\u001b[0;34m(deep_params, x)\u001b[0m\n\u001b[1;32m 45\u001b[0m w_output \u001b[38;5;241m=\u001b[39m deep_params[\u001b[38;5;241m-\u001b[39m\u001b[38;5;241m1\u001b[39m]\n\u001b[1;32m 47\u001b[0m \u001b[38;5;66;03m# Include bias:\u001b[39;00m\n\u001b[0;32m---> 48\u001b[0m x_prev \u001b[38;5;241m=\u001b[39m \u001b[43mnp\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mconcatenate\u001b[49m\u001b[43m(\u001b[49m\u001b[43m(\u001b[49m\u001b[43mnp\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mones\u001b[49m\u001b[43m(\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m1\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43mnum_points\u001b[49m\u001b[43m)\u001b[49m\u001b[43m)\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mx_prev\u001b[49m\u001b[43m)\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43maxis\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43m \u001b[49m\u001b[38;5;241;43m0\u001b[39;49m\u001b[43m)\u001b[49m\n\u001b[1;32m 50\u001b[0m z_output \u001b[38;5;241m=\u001b[39m np\u001b[38;5;241m.\u001b[39mmatmul(w_output, x_prev)\n\u001b[1;32m 51\u001b[0m x_output \u001b[38;5;241m=\u001b[39m z_output\n", - "File \u001b[0;32m~/miniforge3/envs/myenv/lib/python3.9/site-packages/autograd/numpy/numpy_wrapper.py:38\u001b[0m, in \u001b[0;36m\u001b[0;34m(arr_list, axis)\u001b[0m\n\u001b[1;32m 35\u001b[0m \u001b[38;5;129m@primitive\u001b[39m\n\u001b[1;32m 36\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21mconcatenate_args\u001b[39m(axis, \u001b[38;5;241m*\u001b[39margs):\n\u001b[1;32m 37\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m _np\u001b[38;5;241m.\u001b[39mconcatenate(args, axis)\u001b[38;5;241m.\u001b[39mview(ndarray)\n\u001b[0;32m---> 38\u001b[0m concatenate \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;01mlambda\u001b[39;00m arr_list, axis\u001b[38;5;241m=\u001b[39m\u001b[38;5;241m0\u001b[39m : \u001b[43mconcatenate_args\u001b[49m\u001b[43m(\u001b[49m\u001b[43maxis\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43marr_list\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m 39\u001b[0m vstack \u001b[38;5;241m=\u001b[39m row_stack \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;01mlambda\u001b[39;00m tup: concatenate([atleast_2d(_m) \u001b[38;5;28;01mfor\u001b[39;00m _m \u001b[38;5;129;01min\u001b[39;00m tup], axis\u001b[38;5;241m=\u001b[39m\u001b[38;5;241m0\u001b[39m)\n\u001b[1;32m 40\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21mhstack\u001b[39m(tup):\n", + "Cell \u001b[0;32mIn[9], line 37\u001b[0m, in \u001b[0;36mdeep_neural_network\u001b[0;34m(deep_params, x)\u001b[0m\n\u001b[1;32m 34\u001b[0m x_prev \u001b[38;5;241m=\u001b[39m np\u001b[38;5;241m.\u001b[39mconcatenate((np\u001b[38;5;241m.\u001b[39mones((\u001b[38;5;241m1\u001b[39m,num_points)), x_prev ), axis \u001b[38;5;241m=\u001b[39m \u001b[38;5;241m0\u001b[39m)\n\u001b[1;32m 36\u001b[0m z_hidden \u001b[38;5;241m=\u001b[39m np\u001b[38;5;241m.\u001b[39mmatmul(w_hidden, x_prev)\n\u001b[0;32m---> 37\u001b[0m x_hidden \u001b[38;5;241m=\u001b[39m \u001b[43msigmoid\u001b[49m\u001b[43m(\u001b[49m\u001b[43mz_hidden\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m 39\u001b[0m \u001b[38;5;66;03m# Update x_prev such that next layer can use the output from this layer\u001b[39;00m\n\u001b[1;32m 40\u001b[0m x_prev \u001b[38;5;241m=\u001b[39m x_hidden\n", + "Cell \u001b[0;32mIn[9], line 11\u001b[0m, in \u001b[0;36msigmoid\u001b[0;34m(z)\u001b[0m\n\u001b[1;32m 10\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21msigmoid\u001b[39m(z):\n\u001b[0;32m---> 11\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;241;43m1\u001b[39;49m\u001b[38;5;241;43m/\u001b[39;49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m1\u001b[39;49m\u001b[43m \u001b[49m\u001b[38;5;241;43m+\u001b[39;49m\u001b[43m \u001b[49m\u001b[43mnp\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mexp\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m-\u001b[39;49m\u001b[43mz\u001b[49m\u001b[43m)\u001b[49m\u001b[43m)\u001b[49m\n", + "File \u001b[0;32m~/miniforge3/envs/myenv/lib/python3.9/site-packages/autograd/numpy/numpy_boxes.py:39\u001b[0m, in \u001b[0;36mArrayBox.__rtruediv__\u001b[0;34m(self, other)\u001b[0m\n\u001b[0;32m---> 39\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21m__rtruediv__\u001b[39m(\u001b[38;5;28mself\u001b[39m, other): \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43manp\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mtrue_divide\u001b[49m\u001b[43m(\u001b[49m\u001b[43mother\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[43m)\u001b[49m\n", "File \u001b[0;32m~/miniforge3/envs/myenv/lib/python3.9/site-packages/autograd/tracer.py:45\u001b[0m, in \u001b[0;36mprimitive..f_wrapped\u001b[0;34m(*args, **kwargs)\u001b[0m\n\u001b[1;32m 43\u001b[0m argnums \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mtuple\u001b[39m(argnum \u001b[38;5;28;01mfor\u001b[39;00m argnum, _ \u001b[38;5;129;01min\u001b[39;00m boxed_args)\n\u001b[1;32m 44\u001b[0m ans \u001b[38;5;241m=\u001b[39m f_wrapped(\u001b[38;5;241m*\u001b[39margvals, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mkwargs)\n\u001b[0;32m---> 45\u001b[0m node \u001b[38;5;241m=\u001b[39m \u001b[43mnode_constructor\u001b[49m\u001b[43m(\u001b[49m\u001b[43mans\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mf_wrapped\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43margvals\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mkwargs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43margnums\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mparents\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m 46\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m new_box(ans, trace, node)\n\u001b[1;32m 47\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n", "File \u001b[0;32m~/miniforge3/envs/myenv/lib/python3.9/site-packages/autograd/core.py:36\u001b[0m, in \u001b[0;36mVJPNode.__init__\u001b[0;34m(self, value, fun, args, kwargs, parent_argnums, parents)\u001b[0m\n\u001b[1;32m 33\u001b[0m fun_name \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mgetattr\u001b[39m(fun, \u001b[38;5;124m'\u001b[39m\u001b[38;5;124m__name__\u001b[39m\u001b[38;5;124m'\u001b[39m, fun)\n\u001b[1;32m 34\u001b[0m \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mNotImplementedError\u001b[39;00m(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mVJP of \u001b[39m\u001b[38;5;132;01m{}\u001b[39;00m\u001b[38;5;124m wrt argnums \u001b[39m\u001b[38;5;132;01m{}\u001b[39;00m\u001b[38;5;124m not defined\u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[1;32m 35\u001b[0m \u001b[38;5;241m.\u001b[39mformat(fun_name, parent_argnums))\n\u001b[0;32m---> 36\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mvjp \u001b[38;5;241m=\u001b[39m \u001b[43mvjpmaker\u001b[49m\u001b[43m(\u001b[49m\u001b[43mparent_argnums\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mvalue\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n", - "File \u001b[0;32m~/miniforge3/envs/myenv/lib/python3.9/site-packages/autograd/core.py:48\u001b[0m, in \u001b[0;36mdefvjp_argnum..vjp_argnums\u001b[0;34m(argnums, *args)\u001b[0m\n\u001b[1;32m 47\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21mvjp_argnums\u001b[39m(argnums, \u001b[38;5;241m*\u001b[39margs):\n\u001b[0;32m---> 48\u001b[0m vjps \u001b[38;5;241m=\u001b[39m [vjpmaker(argnum, \u001b[38;5;241m*\u001b[39margs) \u001b[38;5;28;01mfor\u001b[39;00m argnum \u001b[38;5;129;01min\u001b[39;00m argnums]\n\u001b[1;32m 49\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;01mlambda\u001b[39;00m g: (vjp(g) \u001b[38;5;28;01mfor\u001b[39;00m vjp \u001b[38;5;129;01min\u001b[39;00m vjps)\n", - "File \u001b[0;32m~/miniforge3/envs/myenv/lib/python3.9/site-packages/autograd/core.py:48\u001b[0m, in \u001b[0;36m\u001b[0;34m(.0)\u001b[0m\n\u001b[1;32m 47\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21mvjp_argnums\u001b[39m(argnums, \u001b[38;5;241m*\u001b[39margs):\n\u001b[0;32m---> 48\u001b[0m vjps \u001b[38;5;241m=\u001b[39m [\u001b[43mvjpmaker\u001b[49m\u001b[43m(\u001b[49m\u001b[43margnum\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m)\u001b[49m \u001b[38;5;28;01mfor\u001b[39;00m argnum \u001b[38;5;129;01min\u001b[39;00m argnums]\n\u001b[1;32m 49\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;01mlambda\u001b[39;00m g: (vjp(g) \u001b[38;5;28;01mfor\u001b[39;00m vjp \u001b[38;5;129;01min\u001b[39;00m vjps)\n", - "File \u001b[0;32m~/miniforge3/envs/myenv/lib/python3.9/site-packages/autograd/numpy/numpy_vjps.py:537\u001b[0m, in \u001b[0;36mgrad_concatenate_args\u001b[0;34m(argnum, ans, axis_args, kwargs)\u001b[0m\n\u001b[1;32m 532\u001b[0m defvjp(tensordot_adjoint_1, \u001b[38;5;28;01mlambda\u001b[39;00m ans, A, G, axes, An, Bn: \u001b[38;5;28;01mlambda\u001b[39;00m B: match_complex(A, tensordot_adjoint_0(B, G, axes, An, Bn)),\n\u001b[1;32m 533\u001b[0m \u001b[38;5;28;01mlambda\u001b[39;00m ans, A, G, axes, An, Bn: \u001b[38;5;28;01mlambda\u001b[39;00m B: match_complex(G, anp\u001b[38;5;241m.\u001b[39mtensordot(A, B, axes)))\n\u001b[1;32m 534\u001b[0m defvjp(anp\u001b[38;5;241m.\u001b[39mouter, \u001b[38;5;28;01mlambda\u001b[39;00m ans, a, b : \u001b[38;5;28;01mlambda\u001b[39;00m g: match_complex(a, anp\u001b[38;5;241m.\u001b[39mdot(g, b\u001b[38;5;241m.\u001b[39mT)),\n\u001b[1;32m 535\u001b[0m \u001b[38;5;28;01mlambda\u001b[39;00m ans, a, b : \u001b[38;5;28;01mlambda\u001b[39;00m g: match_complex(b, anp\u001b[38;5;241m.\u001b[39mdot(a\u001b[38;5;241m.\u001b[39mT, g)))\n\u001b[0;32m--> 537\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21mgrad_concatenate_args\u001b[39m(argnum, ans, axis_args, kwargs):\n\u001b[1;32m 538\u001b[0m axis, args \u001b[38;5;241m=\u001b[39m axis_args[\u001b[38;5;241m0\u001b[39m], axis_args[\u001b[38;5;241m1\u001b[39m:]\n\u001b[1;32m 539\u001b[0m sizes \u001b[38;5;241m=\u001b[39m [anp\u001b[38;5;241m.\u001b[39mshape(a)[axis] \u001b[38;5;28;01mfor\u001b[39;00m a \u001b[38;5;129;01min\u001b[39;00m args[:argnum]]\n", + "File \u001b[0;32m~/miniforge3/envs/myenv/lib/python3.9/site-packages/autograd/core.py:66\u001b[0m, in \u001b[0;36mdefvjp..vjp_argnums\u001b[0;34m(argnums, ans, args, kwargs)\u001b[0m\n\u001b[1;32m 63\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mKeyError\u001b[39;00m:\n\u001b[1;32m 64\u001b[0m \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mNotImplementedError\u001b[39;00m(\n\u001b[1;32m 65\u001b[0m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mVJP of \u001b[39m\u001b[38;5;132;01m{}\u001b[39;00m\u001b[38;5;124m wrt argnum 0 not defined\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;241m.\u001b[39mformat(fun\u001b[38;5;241m.\u001b[39m\u001b[38;5;18m__name__\u001b[39m))\n\u001b[0;32m---> 66\u001b[0m vjp \u001b[38;5;241m=\u001b[39m \u001b[43mvjpfun\u001b[49m\u001b[43m(\u001b[49m\u001b[43mans\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m 67\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;01mlambda\u001b[39;00m g: (vjp(g),)\n\u001b[1;32m 68\u001b[0m \u001b[38;5;28;01melif\u001b[39;00m L \u001b[38;5;241m==\u001b[39m \u001b[38;5;241m2\u001b[39m:\n", + "File \u001b[0;32m~/miniforge3/envs/myenv/lib/python3.9/site-packages/autograd/numpy/numpy_vjps.py:53\u001b[0m, in \u001b[0;36m\u001b[0;34m(ans, x, y)\u001b[0m\n\u001b[1;32m 48\u001b[0m defvjp(anp\u001b[38;5;241m.\u001b[39mlogaddexp, \u001b[38;5;28;01mlambda\u001b[39;00m ans, x, y : unbroadcast_f(x, \u001b[38;5;28;01mlambda\u001b[39;00m g: g \u001b[38;5;241m*\u001b[39m anp\u001b[38;5;241m.\u001b[39mexp(x\u001b[38;5;241m-\u001b[39mans)),\n\u001b[1;32m 49\u001b[0m \u001b[38;5;28;01mlambda\u001b[39;00m ans, x, y : unbroadcast_f(y, \u001b[38;5;28;01mlambda\u001b[39;00m g: g \u001b[38;5;241m*\u001b[39m anp\u001b[38;5;241m.\u001b[39mexp(y\u001b[38;5;241m-\u001b[39mans)))\n\u001b[1;32m 50\u001b[0m defvjp(anp\u001b[38;5;241m.\u001b[39mlogaddexp2, \u001b[38;5;28;01mlambda\u001b[39;00m ans, x, y : unbroadcast_f(x, \u001b[38;5;28;01mlambda\u001b[39;00m g: g \u001b[38;5;241m*\u001b[39m \u001b[38;5;241m2\u001b[39m\u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39m(x\u001b[38;5;241m-\u001b[39mans)),\n\u001b[1;32m 51\u001b[0m \u001b[38;5;28;01mlambda\u001b[39;00m ans, x, y : unbroadcast_f(y, \u001b[38;5;28;01mlambda\u001b[39;00m g: g \u001b[38;5;241m*\u001b[39m \u001b[38;5;241m2\u001b[39m\u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39m(y\u001b[38;5;241m-\u001b[39mans)))\n\u001b[1;32m 52\u001b[0m defvjp(anp\u001b[38;5;241m.\u001b[39mtrue_divide, \u001b[38;5;28;01mlambda\u001b[39;00m ans, x, y : unbroadcast_f(x, \u001b[38;5;28;01mlambda\u001b[39;00m g: g \u001b[38;5;241m/\u001b[39m y),\n\u001b[0;32m---> 53\u001b[0m \u001b[38;5;28;01mlambda\u001b[39;00m ans, x, y : \u001b[43munbroadcast_f\u001b[49m\u001b[43m(\u001b[49m\u001b[43my\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;28;43;01mlambda\u001b[39;49;00m\u001b[43m \u001b[49m\u001b[43mg\u001b[49m\u001b[43m:\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m-\u001b[39;49m\u001b[43m \u001b[49m\u001b[43mg\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43m \u001b[49m\u001b[43mx\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m/\u001b[39;49m\u001b[43m \u001b[49m\u001b[43my\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m2\u001b[39;49m\u001b[43m)\u001b[49m)\n\u001b[1;32m 54\u001b[0m defvjp(anp\u001b[38;5;241m.\u001b[39mmod, \u001b[38;5;28;01mlambda\u001b[39;00m ans, x, y : unbroadcast_f(x, \u001b[38;5;28;01mlambda\u001b[39;00m g: g),\n\u001b[1;32m 55\u001b[0m \u001b[38;5;28;01mlambda\u001b[39;00m ans, x, y : unbroadcast_f(y, \u001b[38;5;28;01mlambda\u001b[39;00m g: \u001b[38;5;241m-\u001b[39mg \u001b[38;5;241m*\u001b[39m anp\u001b[38;5;241m.\u001b[39mfloor(x\u001b[38;5;241m/\u001b[39my)))\n\u001b[1;32m 56\u001b[0m defvjp(anp\u001b[38;5;241m.\u001b[39mremainder, \u001b[38;5;28;01mlambda\u001b[39;00m ans, x, y : unbroadcast_f(x, \u001b[38;5;28;01mlambda\u001b[39;00m g: g),\n\u001b[1;32m 57\u001b[0m \u001b[38;5;28;01mlambda\u001b[39;00m ans, x, y : unbroadcast_f(y, \u001b[38;5;28;01mlambda\u001b[39;00m g: \u001b[38;5;241m-\u001b[39mg \u001b[38;5;241m*\u001b[39m anp\u001b[38;5;241m.\u001b[39mfloor(x\u001b[38;5;241m/\u001b[39my)))\n", + "File \u001b[0;32m~/miniforge3/envs/myenv/lib/python3.9/site-packages/autograd/numpy/numpy_vjps.py:658\u001b[0m, in \u001b[0;36munbroadcast_f\u001b[0;34m(target, f)\u001b[0m\n\u001b[1;32m 655\u001b[0m x \u001b[38;5;241m=\u001b[39m anp\u001b[38;5;241m.\u001b[39mreal(x)\n\u001b[1;32m 656\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m x\n\u001b[0;32m--> 658\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21munbroadcast_f\u001b[39m(target, f):\n\u001b[1;32m 659\u001b[0m target_meta \u001b[38;5;241m=\u001b[39m anp\u001b[38;5;241m.\u001b[39mmetadata(target)\n\u001b[1;32m 660\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;01mlambda\u001b[39;00m g: unbroadcast(f(g), target_meta)\n", "\u001b[0;31mKeyboardInterrupt\u001b[0m: " ] } diff --git a/doc/LectureNotes/_build/jupyter_execute/chapter1_17_0.png b/doc/LectureNotes/_build/jupyter_execute/chapter1_17_0.png index da9f24db5..ec1d7e7fa 100644 Binary files a/doc/LectureNotes/_build/jupyter_execute/chapter1_17_0.png and b/doc/LectureNotes/_build/jupyter_execute/chapter1_17_0.png differ diff --git a/doc/LectureNotes/_build/jupyter_execute/chapter1_19_1.png b/doc/LectureNotes/_build/jupyter_execute/chapter1_19_1.png index e9a927d85..ef4481ba4 100644 Binary files a/doc/LectureNotes/_build/jupyter_execute/chapter1_19_1.png and b/doc/LectureNotes/_build/jupyter_execute/chapter1_19_1.png differ diff --git a/doc/LectureNotes/_build/jupyter_execute/chapter1_33_0.png b/doc/LectureNotes/_build/jupyter_execute/chapter1_33_0.png index a5ffb3dee..0362aa22e 100644 Binary files a/doc/LectureNotes/_build/jupyter_execute/chapter1_33_0.png and b/doc/LectureNotes/_build/jupyter_execute/chapter1_33_0.png differ diff --git a/doc/LectureNotes/_build/jupyter_execute/chapter1_9_0.png b/doc/LectureNotes/_build/jupyter_execute/chapter1_9_0.png index 88665d1b8..4831fbb1f 100644 Binary files a/doc/LectureNotes/_build/jupyter_execute/chapter1_9_0.png and b/doc/LectureNotes/_build/jupyter_execute/chapter1_9_0.png differ diff --git a/doc/LectureNotes/_build/jupyter_execute/chapter2.ipynb b/doc/LectureNotes/_build/jupyter_execute/chapter2.ipynb index d7c28f34d..3144e9d56 100644 --- a/doc/LectureNotes/_build/jupyter_execute/chapter2.ipynb +++ b/doc/LectureNotes/_build/jupyter_execute/chapter2.ipynb @@ -1798,10 +1798,10 @@ "name": "stdout", "output_type": "stream", "text": [ - "0.05005634426334421\n", - "4.366375489616074\n", - "[[ 0.93787605 2.95563211]\n", - " [ 2.95563211 10.33025801]]\n" + "-0.0369544130635358\n", + "3.662836197064178\n", + "[[1.058997 3.11439407]\n", + " [3.11439407 9.99498272]]\n" ] } ], @@ -1845,10 +1845,10 @@ "name": "stdout", "output_type": "stream", "text": [ - "0.07808426989543932\n", - "1.4121966338442804\n", - "[[1. 0.70362677]\n", - " [0.70362677 1. ]]\n" + "0.08826182458028335\n", + "1.7026722043092946\n", + "[[1. 0.61113781]\n", + " [0.61113781 1. ]]\n" ] } ], @@ -1905,30 +1905,30 @@ "name": "stdout", "output_type": "stream", "text": [ - "[[-0.34661376 -1.6809195 ]\n", - " [ 0.05792927 0.30915293]\n", - " [ 0.65183066 3.00564344]\n", - " [ 1.75018686 4.35667342]\n", - " [-0.75682834 -1.67875366]\n", - " [ 1.16654048 3.9065894 ]\n", - " [-1.86267497 -5.53585173]\n", - " [ 0.29803738 2.45731144]\n", - " [-0.63031478 -2.76157429]\n", - " [-0.3280928 -2.37827145]]\n", + "[[-0.7252563 -2.26264849]\n", + " [ 1.19052935 3.11261935]\n", + " [-0.62158409 -2.99662602]\n", + " [-0.06216141 -0.18120973]\n", + " [ 1.32065614 3.50269821]\n", + " [ 0.83995705 2.80855691]\n", + " [ 0.1571284 0.96919021]\n", + " [-0.03404758 1.01551815]\n", + " [-0.23596934 0.77449804]\n", + " [-1.82925222 -6.74259663]]\n", " 0 1\n", - "0 -0.346614 -1.680920\n", - "1 0.057929 0.309153\n", - "2 0.651831 3.005643\n", - "3 1.750187 4.356673\n", - "4 -0.756828 -1.678754\n", - "5 1.166540 3.906589\n", - "6 -1.862675 -5.535852\n", - "7 0.298037 2.457311\n", - "8 -0.630315 -2.761574\n", - "9 -0.328093 -2.378271\n", + "0 -0.725256 -2.262648\n", + "1 1.190529 3.112619\n", + "2 -0.621584 -2.996626\n", + "3 -0.062161 -0.181210\n", + "4 1.320656 3.502698\n", + "5 0.839957 2.808557\n", + "6 0.157128 0.969190\n", + "7 -0.034048 1.015518\n", + "8 -0.235969 0.774498\n", + "9 -1.829252 -6.742597\n", " 0 1\n", - "0 1.000000 0.959076\n", - "1 0.959076 1.000000\n" + "0 1.000000 0.963187\n", + "1 0.963187 1.000000\n" ] } ], @@ -1974,37 +1974,37 @@ "text": [ " 0 1 2 3 4 5 6 7 \\\n", "0 0.0 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 \n", - "1 0.0 0.084996 0.084434 0.084712 0.085455 0.086212 0.075771 0.076638 \n", - "2 0.0 0.084434 0.084514 0.084042 0.085120 0.086230 0.075176 0.076259 \n", - "3 0.0 0.084712 0.084042 0.089718 0.090439 0.091159 0.083479 0.084358 \n", - "4 0.0 0.085455 0.085120 0.090439 0.091388 0.092347 0.084091 0.085139 \n", - "5 0.0 0.086212 0.086230 0.091159 0.092347 0.093554 0.084695 0.085918 \n", - "6 0.0 0.075771 0.075176 0.083479 0.084091 0.084695 0.079903 0.080657 \n", - "7 0.0 0.076638 0.076259 0.084358 0.085139 0.085918 0.080657 0.081544 \n", - "8 0.0 0.077567 0.077411 0.085294 0.086250 0.087210 0.081457 0.082482 \n", - "9 0.0 0.078557 0.078636 0.086286 0.087424 0.088574 0.082303 0.083471 \n", - "10 0.0 0.066997 0.066458 0.075909 0.076389 0.076857 0.074221 0.074824 \n", - "11 0.0 0.067764 0.067380 0.076693 0.077304 0.077906 0.074892 0.075602 \n", - "12 0.0 0.068592 0.068369 0.077539 0.078284 0.079027 0.075615 0.076436 \n", - "13 0.0 0.069484 0.069427 0.078447 0.079334 0.080222 0.076393 0.077329 \n", - "14 0.0 0.070441 0.070558 0.079420 0.080453 0.081494 0.077226 0.078282 \n", + "1 0.0 0.077382 0.076459 0.074114 0.076542 0.079074 0.064978 0.067074 \n", + "2 0.0 0.076459 0.078118 0.070150 0.073866 0.078123 0.059624 0.062356 \n", + "3 0.0 0.074114 0.070150 0.076618 0.077248 0.077529 0.070432 0.071572 \n", + "4 0.0 0.076542 0.073866 0.077248 0.078731 0.080082 0.069802 0.071449 \n", + "5 0.0 0.079074 0.078123 0.077529 0.080082 0.082803 0.068607 0.070859 \n", + "6 0.0 0.064978 0.059624 0.070432 0.069802 0.068607 0.066865 0.067168 \n", + "7 0.0 0.067074 0.062356 0.071572 0.071449 0.070859 0.067168 0.067809 \n", + "8 0.0 0.069439 0.065556 0.072752 0.073253 0.073422 0.067367 0.068408 \n", + "9 0.0 0.072092 0.069317 0.073923 0.075198 0.076333 0.067380 0.068897 \n", + "10 0.0 0.056856 0.051021 0.063637 0.062282 0.060288 0.061848 0.061588 \n", + "11 0.0 0.058361 0.052856 0.064604 0.063560 0.061921 0.062263 0.062231 \n", + "12 0.0 0.060085 0.055000 0.065679 0.065007 0.063802 0.062700 0.062932 \n", + "13 0.0 0.062053 0.057514 0.066855 0.066633 0.065963 0.063133 0.063673 \n", + "14 0.0 0.064294 0.060470 0.068116 0.068446 0.068446 0.063525 0.064427 \n", "\n", " 8 9 10 11 12 13 14 \n", "0 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 \n", - "1 0.077567 0.078557 0.066997 0.067764 0.068592 0.069484 0.070441 \n", - "2 0.077411 0.078636 0.066458 0.067380 0.068369 0.069427 0.070558 \n", - "3 0.085294 0.086286 0.075909 0.076693 0.077539 0.078447 0.079420 \n", - "4 0.086250 0.087424 0.076389 0.077304 0.078284 0.079334 0.080453 \n", - "5 0.087210 0.088574 0.076857 0.077906 0.079027 0.080222 0.081494 \n", - "6 0.081457 0.082303 0.074221 0.074892 0.075615 0.076393 0.077226 \n", - "7 0.082482 0.083471 0.074824 0.075602 0.076436 0.077329 0.078282 \n", - "8 0.083564 0.084702 0.075463 0.076351 0.077301 0.078313 0.079391 \n", - "9 0.084702 0.085996 0.076138 0.077141 0.078210 0.079347 0.080555 \n", - "10 0.075463 0.076138 0.070095 0.070630 0.071208 0.071831 0.072500 \n", - "11 0.076351 0.077141 0.070630 0.071252 0.071921 0.072638 0.073405 \n", - "12 0.077301 0.078210 0.071208 0.071921 0.072684 0.073499 0.074368 \n", - "13 0.078313 0.079347 0.071831 0.072638 0.073499 0.074417 0.075392 \n", - "14 0.079391 0.080555 0.072500 0.073405 0.074368 0.075392 0.076479 \n" + "1 0.069439 0.072092 0.056856 0.058361 0.060085 0.062053 0.064294 \n", + "2 0.065556 0.069317 0.051021 0.052856 0.055000 0.057514 0.060470 \n", + "3 0.072752 0.073923 0.063637 0.064604 0.065679 0.066855 0.068116 \n", + "4 0.073253 0.075198 0.062282 0.063560 0.065007 0.066633 0.068446 \n", + "5 0.073422 0.076333 0.060288 0.061921 0.063802 0.065963 0.068446 \n", + "6 0.067367 0.067380 0.061848 0.062263 0.062700 0.063133 0.063525 \n", + "7 0.068408 0.068897 0.061588 0.062231 0.062932 0.063673 0.064427 \n", + "8 0.069486 0.070556 0.061154 0.062058 0.063064 0.064166 0.065352 \n", + "9 0.070556 0.072344 0.060453 0.061656 0.063017 0.064547 0.066256 \n", + "10 0.061154 0.060453 0.058245 0.058254 0.058234 0.058152 0.057961 \n", + "11 0.062058 0.061656 0.058254 0.058427 0.058593 0.058721 0.058770 \n", + "12 0.063064 0.063017 0.058234 0.058593 0.058970 0.059340 0.059669 \n", + "13 0.064166 0.064547 0.058152 0.058721 0.059340 0.059992 0.060652 \n", + "14 0.065352 0.066256 0.057961 0.058770 0.059669 0.060652 0.061706 \n" ] } ],