diff --git a/doc/BookChapters/chapter3.do.txt b/doc/BookChapters/chapter3.do.txt index f1f70e6ee..f831f9344 100644 --- a/doc/BookChapters/chapter3.do.txt +++ b/doc/BookChapters/chapter3.do.txt @@ -1245,6 +1245,7 @@ y_pred = y_pred + y_train_mean !ec + Let us try to understand what this may imply mathematically when we subtract the mean values, also known as *zero centering*. For simplicity, we will focus on ordinary regression, as done in the above example. @@ -1281,20 +1282,6 @@ Multiplying away the constant $2/n$, we obtain \] !et - -We assume -that every column of $\bm{X}$ is centered, which we can do by subtracting the mean, -!bc pycod -X = X - np.mean(X,axis=0) -!ec - -This means that we need to rewrite $X_{ij}$ as $\tilde{X}_{ij}=X_{ij}-\mu_j$, where -!bt -\[ -\mu_j = \frac{1}{n}\sum_{i=0}^{n-1}X_{ij}. -\] -!et - Let us special first to the case where we have only two parameters $\beta_0$ and $\beta_1$. Our result for $\beta_0$ simplifies then to !bt @@ -1302,13 +1289,13 @@ Our result for $\beta_0$ simplifies then to n\beta_0 = \sum_{i=0}^{n-1}y_i - \sum_{i=0}^{n-1} X_{i1} \beta_1. \] !et -Assuming that the matrix elements $X_{i1}$ are centered, what we have is +We obtain then !bt \[ -\beta_0 = \frac{1}{n}\sum_{i=0}^{n-1}y_i - \beta_1\frac{1}{n}\sum_{i=0}^{n-1} \left(X_{i1}-\mu_{1}\right), +\beta_0 = \frac{1}{n}\sum_{i=0}^{n-1}y_i - \beta_1\frac{1}{n}\sum_{i=0}^{n-1} X_{i1}. \] !et -where +If we define !bt \[ \mu_1=\frac{1}{n}\sum_{i=0}^{n-1} (X_{i1}, @@ -1323,25 +1310,19 @@ and if we define the mean value of the outputs as we have !bt \[ -\beta_0 = \mu_y - \beta_1\frac{1}{n}\sum_{i=0}^{n-1} (X_{i1}-\mu_{1}), +\beta_0 = \mu_y - \beta_1\mu_{1}. \] !et -and it is easy to see that the last sum equals zero! This means that we have +In the general case withmore parameters than $\beta_0$ and $\beta_1$, we have !bt \[ -\beta_0 = \mu_y, +\beta_0 = \frac{1}{n}\sum_{i=0}^{n-1}y_i - \frac{1}{n}\sum_{i=0}^{n-1}\sum_{j=1}^{p-1} X_{ij}\beta_j. \] !et -if the columns of the design matrix are centered. It is straight forward to generalize this results to more values of $\beta$. -We have thus -!bt -\[ -\beta_0 = \frac{1}{n}\sum_{i=0}^{n-1} y_i = \overline{\bm{y}}, -\] -!et -the average value of $\bm{y}$. -Replacing $y_i$ with $y_i - \beta_0 = y_i - \overline{\bm{y}}$ and centering also our design matrix results in a cost function (in vector-matrix disguise) + + +Replacing $y_i$ with $y_i - y_i - \overline{\bm{y}}$ and centering also our design matrix results in a cost function (in vector-matrix disguise) !bt \[ C(\boldsymbol{\beta}) = (\boldsymbol{\tilde{y}} - \tilde{X}\boldsymbol{\beta})^T(\boldsymbol{\tilde{y}} - \tilde{X}\boldsymbol{\beta}). @@ -1349,6 +1330,7 @@ C(\boldsymbol{\beta}) = (\boldsymbol{\tilde{y}} - \tilde{X}\boldsymbol{\beta})^T !et + If we minimize with respect to $\bm{\beta}$ we have then !bt @@ -1629,8 +1611,6 @@ for i in range(nlambdas): OwnRidgeBeta = np.linalg.pinv(X_train_scaled.T @ X_train_scaled+lmb*I) @ X_train_scaled.T @ (y_train_scaled) intercept_ = y_scaler - X_train_mean@OwnRidgeBeta #The intercept can be shifted so the model can predict on uncentered data #Add intercept to prediction - ypredictOwnRidge = X_test @ OwnRidgeBeta + intercept_ - #Add intercept to prediction ypredictOwnRidge = X_test_scaled @ OwnRidgeBeta + y_scaler RegRidge = linear_model.Ridge(lmb) RegRidge.fit(X_train,y_train) @@ -2463,3 +2443,4 @@ of data presented here (either the terrain data we propose or other data sets). + diff --git a/doc/LectureNotes/_build/.doctrees/Clustering.doctree b/doc/LectureNotes/_build/.doctrees/Clustering.doctree index 59c2e4d96..104e791aa 100644 Binary files a/doc/LectureNotes/_build/.doctrees/Clustering.doctree and b/doc/LectureNotes/_build/.doctrees/Clustering.doctree differ diff --git a/doc/LectureNotes/_build/.doctrees/chapter1.doctree b/doc/LectureNotes/_build/.doctrees/chapter1.doctree index 929726e54..2836f14ae 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 75555916a..88d74cf11 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/jupyter_execute/Clustering.ipynb b/doc/LectureNotes/_build/jupyter_execute/Clustering.ipynb index 7ee16befd..13e390d92 100644 --- a/doc/LectureNotes/_build/jupyter_execute/Clustering.ipynb +++ b/doc/LectureNotes/_build/jupyter_execute/Clustering.ipynb @@ -475,7 +475,7 @@ "output_type": "stream", "text": [ "Converged at iteration 5\n", - "Runtime: 0.4849228858947754 seconds\n" + "Runtime: 0.47818493843078613 seconds\n" ] } ], @@ -604,7 +604,7 @@ "output_type": "stream", "text": [ "Converged at iteration: 5\n", - "Runtime: 0.41676807403564453 seconds\n" + "Runtime: 0.42156386375427246 seconds\n" ] } ], @@ -745,7 +745,7 @@ "output_type": "stream", "text": [ "Converged at iteration: 11\n", - "Runtime: 0.8447701930999756 seconds\n", + "Runtime: 0.8341619968414307 seconds\n", " " ] } @@ -877,7 +877,7 @@ "output_type": "stream", "text": [ "Converged at iteration: 5\n", - "Runtime: 0.004042863845825195 seconds\n" + "Runtime: 0.0038292407989501953 seconds\n" ] } ], diff --git a/doc/LectureNotes/_build/jupyter_execute/chapter1.ipynb b/doc/LectureNotes/_build/jupyter_execute/chapter1.ipynb index 56aca56d9..bbbe608f8 100644 --- a/doc/LectureNotes/_build/jupyter_execute/chapter1.ipynb +++ b/doc/LectureNotes/_build/jupyter_execute/chapter1.ipynb @@ -310,7 +310,7 @@ "outputs": [ { "data": { - "image/png": "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\n", + "image/png": "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\n", "text/plain": [ "
" ] @@ -461,7 +461,7 @@ "outputs": [ { "data": { - "image/png": "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\n", + "image/png": "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\n", "text/plain": [ "
" ] @@ -526,18 +526,18 @@ "output_type": "stream", "text": [ "The intercept alpha: \n", - " [2.07710775]\n", + " [1.95080408]\n", "Coefficient beta : \n", - " [[4.7577194]]\n", - "Mean squared error: 0.19\n", - "Variance score: 0.91\n", + " [[5.12184106]]\n", + "Mean squared error: 0.26\n", + "Variance score: 0.88\n", "Mean squared log error: 0.01\n", - "Mean absolute error: 0.34\n" + "Mean absolute error: 0.40\n" ] }, { "data": { - "image/png": "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\n", + "image/png": "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\n", "text/plain": [ "
" ] @@ -730,7 +730,7 @@ "outputs": [ { "data": { - "image/png": "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\n", + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAD4CAYAAAAXUaZHAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/Il7ecAAAACXBIWXMAAAsTAAALEwEAmpwYAAA03klEQVR4nO3deXhU5fXA8e+ZSUICBEhI2JewqoAkQkRRtCpq1WqtValb1dYWqbVqN8Vaf2KtFVsVt6ql1YpWBcG6LxURBQRRREQERUDQIISwb9nn/P6YSZgMk8y+5nyehyczd+7cOXdCzrxz7ruIqmKMMSa9OBIdgDHGmOiz5G6MMWnIkrsxxqQhS+7GGJOGLLkbY0waykh0AAAFBQVaVFSU6DCMMSalfPTRR1tVtdDfY0mR3IuKiliyZEmiwzDGmJQiIhuae8zKMsYYk4YsuRtjTBqy5G6MMWkoKWru/tTW1lJWVkZVVVWiQ0k72dnZ9OrVi8zMzESHYoyJkaRN7mVlZeTm5lJUVISIJDqctKGqbNu2jbKyMvr165focIwxMZK0ZZmqqio6d+5siT3KRITOnTvbNyJj0lzSJnfAEnuM2PtqTBRVlsPs4+HZDu6fleWJjghI8uRujDFJb8H5sHUR1O1x/1xwfqIjAiy5t2jz5s1ccMEFDBgwgJEjR3LGGWewevXqFp9TVFTE1q1bD9r+yCOP8MQTTwT92k6nk5KSksZ/69ev55hjjgFg/fr1PP3006GdjDEmNnYsA61z39Y69/0kEPCCqog8BpwJbFHVYV7bfwX8EqgHXlXV6z3bbwSu8Gy/RlX/F4vAY01VOeecc7jsssuYPn06AJ988gnl5eUMHjw45ONNmDAhpP1zcnJYtmxZk20LFy4EDiT3iy66KOQ4jDFRllfibrFrHUiG+34SCKbl/jhwmvcGETkROBsoVtWhwF2e7UOAC4Chnuc8JCLOaAYcL3PnziUzM7NJUi4uLua4447jnXfe4cwzz2zcfvXVV/P444833v/rX//K4YcfzqhRo1izZg0AkyZN4q677gJgzZo1nHzyyRQXFzNixAjWrl0bVEzt27cHYOLEicyfP5+SkhKmTJkS6akaYyIxZiYUjIaMXPfPMTMTHREQRMtdVeeJSJHP5l8Ak1W12rPPFs/2s4Hpnu1ficgaYBSwKJIgb335M1Z+uzuSQxxkSI8O3HLW0GYfX7FiBSNHjgzr2B07duTTTz/liSee4LrrruOVV15p8vjFF1/MxIkTOeecc6iqqsLlch10jMrKSkpKSgDo168fzz//fONjkydP5q677jrouMaYBMjpCqfMS3QUBwm3n/tg4DgRuR2oAn6nqh8CPYH3vfYr82w7iIiMB8YD9OnTJ8wwktOFF17Y+PPXv/51k8f27NnDxo0bOeeccwD3gCJ//JVljDEmWOEm9wwgHzgaOBJ4VkT6h3IAVZ0KTAUoLS1tcZXullrYsTJ06FBmzZrl97GMjIwmrW3fPuPeXQ2t26ExJhHC7S1TBvxX3T4AXEABsBHo7bVfL8+2lHPSSSdRXV3N1KlTG7ctX76c+fPn07dvX1auXEl1dTU7d+5kzpw5TZ47Y8aMxp+jR49u8lhubi69evXihRdeAKC6upr9+/eHFFtubi579uwJ46yMMa1FuMn9BeBEABEZDGQBW4GXgAtEpI2I9AMGAR9EIc64ExGef/553nrrLQYMGMDQoUO58cYb6datG71792bcuHEMGzaMcePGccQRRzR57o4dOxg+fDj33Xef3wueTz75JPfffz/Dhw/nmGOOYfPmzSHFNnz4cJxOJ8XFxXZB1Zhkl6BBTqLaYkUEEXkGOAF3y7wcuAV4EngMKAFqcNfc3/bsfxPwU6AOuE5VXw8URGlpqfou1rFq1SoOO+yw0M7GBM3eX2PiZPbxTbtKFoyO2gVYEflIVUv9PRZMb5kLm3nokmb2vx24PfjwjDEmjSVokJONUDXGmFjKK3G32CGug5wsuRtjTCwlaJBT0s7nbowxaSFBg5ys5W6MMWnIkrsxxqQhS+7N2LZtW+N0u926daNnz56N92tqaqLyGrW1tUycOJFBgwYxYsQIRo8ezeuvt9xz9PLLL/c7cnbJkiVcc801UYnLGOMlSRfjCMRq7s3o3Llz49wukyZNon379vzud79rfLyuro6MjMjevptvvplNmzaxYsUK2rRpQ3l5Oe+++25YxyotLaW01G93V2NMJBoW49C6A4txJOFEYb6s5R6Cyy+/nAkTJnDUUUdx/fXXN5nGF2DYsGGsX78egP/85z+MGjWKkpISrrzySurr65sca//+/fzzn//kgQceoE2bNgB07dqVcePGAQem9wWYNWsWl19+eeP9t956i9LSUgYPHtw4M6T3NMR79+7lJz/5CYcffjjDhw/nueeei/p7YUyrkaSLcQSSGi33666DaM+QWFIC994b8tPKyspYuHAhTqeTSZMm+d1n1apVzJgxg/fee4/MzEyuuuoqnnrqKS699NLGfdasWUOfPn3o0KFDyDGsX7+eDz74gLVr13LiiSc2zhnf4Lbbbmucdhjc0yEYY8KUpItxBJIayT2JnH/++TidLa8/MmfOHD766COOPPJIwD03e5cuXaIWw7hx43A4HAwaNIj+/fvz+eefN3n8rbfealw9CiAvLy9qr21MqzNmprsUs2OZO7EnyWIcgaRGcg+jhR0r7dq1a7zd3NS/qspll13GHXfc0exxBg4cyNdff83u3bv9tt69pwpuaUphf/eNMVGUpItxBGI19wgUFRWxdOlSAJYuXcpXX30FwNixY5k1axZbtrgXqNq+fTsbNmxo8ty2bdtyxRVXcO211zb2vqmoqGDmTHeroGvXrqxatQqXy9VkFSaAmTNn4nK5WLt2LevWreOQQw5p8vgpp5zC3//+98b7VpYxpvWx5B6Bc889l+3btzN06FAefPDBxoWzhwwZwp///GdOPfVUhg8fzimnnMKmTZsOev6f//xnCgsLGTJkCMOGDePMM89sbMVPnjyZM888k2OOOYbu3bs3eV6fPn0YNWoUp59+Oo888shBqzn98Y9/ZMeOHQwbNozi4mLmzp0bo3fAGJOsAk75Gw825W/82ftrTOpracpfa7kbY0waCpjcReQxEdkiIiv8PPZbEVERKfDcFxG5X0TWiMhyERkRi6CNMca0LJiW++PAab4bRaQ3cCrwtdfm03EvrTcIGA88HHmIxhhjQhUwuavqPGC7n4emANcD3kX7s4EnPAtnvw90EpHufp5rjDEmhsKquYvI2cBGVf3E56GewDde98s824wxJrml6ARhzQk5uYtIW+APwP9F8sIiMl5ElojIkoqKikgOZYwxkWuYIKxuz4EJwlJYOC33AUA/4BMRWQ/0ApaKSDdgI9Dba99enm0HUdWpqlqqqqWFhYVhhBF7TqeTkpIShg4dSnFxMXfffXeTEan+rF+/nqeffjpOERpjoiZFJwhrTsjJXVU/VdUuqlqkqkW4Sy8jVHUz8BJwqafXzNHALlU9ePROisjJyWHZsmV89tlnzJ49m9dff51bb721xedYcjcmRSVoIetYCaYr5DPAIuAQESkTkSta2P01YB2wBvgncFVUokwCXbp0YerUqTz44IOoKuvXr+e4445jxIgRjBgxgoULFwIwceJE5s+fT0lJCVOmTGl2P2NMkknQQtYxo6oJ/zdy5Ej1tXLlyoO2tWj/ZtU3j1Odkev+uX9zaM/3o127dgdt69ixo27evFn37dunlZWVqqq6evVqbTiHuXPn6ve+973G/ZvbL9FCfn+NMUkHWKLN5NXUmBUyGHFeLaW2tparr76aZcuW4XQ6Wb16dUT7GWNMNKVPco/DxZB169bhdDrp0qULt956K127duWTTz7B5XIdNHlXgylTpgS1nzHGRFP6zC0T44shFRUVTJgwgauvvhoRYdeuXXTv3h2Hw8GTTz7ZuIxebm4ue/bsaXxec/sZY1JACvd9T5/kHoOLIZWVlY1dIU8++WROPfVUbrnlFgCuuuoqpk2bRnFxMZ9//nnjIh7Dhw/H6XRSXFzMlClTmt3PGJMCUrjvu03520rZ+2tMEJ7t4E7sDTJyYdzuxMXjw6b8NcaYcKRw33dL7sYY05wU7vue1L1lVNUWf46BZCjFGZMSUnRxbEjilnt2djbbtm2zRBRlqsq2bdusS6YxaS5pW+69evWirKwMmzEy+rKzs+nVq1eiwzDGxFDSJvfMzEz69euX6DCMMemsstzdvXHHMvfF0jEz3aWYNJC0ZRljjIm5FO7HHogld2NM65Vmc7h7s+RujGm9UrgfeyCW3I0xrVcK92MPJGkvqBpjTMylcD/2QIJZiekxEdkiIiu8tv1NRD4XkeUi8ryIdPJ67EYRWSMiX4jId2MUtzHGmBYEU5Z5HDjNZ9tsYJiqDgdWAzcCiMgQ4AJgqOc5D4mIM2rRGmNMKkiCqYIDJndVnQds99n2pmrDJWbeBxpGxJwNTFfValX9CvdaqqOiGK8xxiS/JOhiGY0Lqj8FXvfc7gl84/VYmWfbQURkvIgsEZElNgrVGJNWkqCLZUTJXURuAuqAp0J9rqpOVdVSVS0tLCyMJAxjjEkuSdDFMuzkLiKXA2cCF+uB2b02Ar29duvl2WaMMekjUE09CbpYhtUVUkROA64HvqOq+70eegl4WkTuAXoAg4APIo7SGGOSSUNNXesO1NS9u1QmQRfLgMldRJ4BTgAKRKQMuAV375g2wGzPfOvvq+oEVf1MRJ4FVuIu1/xSVW1FaGNMekmCmnogAZO7ql7oZ/OjLex/O3B7JEEZY0xSyys50HJP0mkLbPoBY4wJVRLU1AOx6QeMMSZUSVBTD8Ra7sYYk4YsuRtjTBqy5G6MMWnIkrsxxqQhS+7GmPSRBLMxJgtL7saY9JEEszEmC0vuxpj0kQIjR+PFkrsxJn0kwWyMycKSuzEmfaTAyNF4sRGqxpj0kQIjR+PFWu7GmPTUynvOWHI3xqSnVt5zxpK7MSY9tfKeMwGTu4g8JiJbRGSF17Z8EZktIl96fuZ5touI3C8ia0RkuYiMiGXwxhjTrFbecyaYlvvjwGk+2yYCc1R1EDDHcx/gdNxL6w0CxgMPRydMY4wJUSg9Z9KwPh8wuavqPGC7z+azgWme29OAH3htf0Ld3gc6iUj3KMVqjGltIkm6DT1nxu12/8zp2vy+aVifD7fm3lVVN3lubwYa3rWewDde+5V5th1ERMaLyBIRWVJRURFmGMaYtBavpJuG9fmIL6iqqgIaxvOmqmqpqpYWFhZGGoYxJh3FK+mmYX0+3ORe3lBu8fzc4tm+EejttV8vzzZjjAldvJJurEe2JqCmH25yfwm4zHP7MuBFr+2XenrNHA3s8irfGGNMaOI1nUAo9flwJKCmH3D6ARF5BjgBKBCRMuAWYDLwrIhcAWwAxnl2fw04A1gD7Ad+EoOYjTGtRbpMJ9BMeamu3kWGMzbDjQImd1W9sJmHxvrZV4FfRhqUMcaklbwSd4td6xrLS/uq67joX4s5b0RPfjy6KOovaSNUjTEm1nzKSzWjZ/CLJz7gkqm3MvSzD2LykjYrpDEmNVWWu2vXO5a5W8ZjZka/Vh4tXuUll0v5/YyPOfWhP3P+8tmw23eMaHRYy90Yk5pScOCRqvLnV1fR/8G/ccmy12HiRPjNb2LyWpbcjTGpKQUHHv1j3jpc993HtQuno1dcAX/5S8xey5K7MSY1pdjAo5lLvuHLv/6dSXOmoj84B3nkERCJ2etZcjfGJE4kg3tSaEm9OavKmfuXR/jb6/fhOvlkZPozkBHbS552QdUYkzgNdXOtO1A3D7Zfe4r0gf/gq+1M/9M/eOilv6JHHYXzhRegTZuYv6613I0xiZNqdfMQv2l8WraLhyY9yoOzbkeGDcP5+mvQrl1cQrXkboxJnBSrm4fSQ+fL8j1Mvu0J/j5jEo4B/cmY/SZ06hS3UC25G2MSJ4Xq5kDQ3zS+3rafm2+fzkNP3kRWty5kznkL4jz7rdXcjTGJk8i6eTiDoPxMI+Br864qbrhjJg/8eyLtOrYn4+050NPvshYxZS13Y0zrFM4gqADfNLbtreaGyc9x79TfktfG4U7s/fvH6ARaZi13Y0zrFM7F3Ba+aeyqrOWGvz7PnX+/lvxMyHhnLhx2WNTCDZW13I0xrVMUL+burqrld399gT/ddw35TiVz7tswbFhUwgyXJXdjTOsUpYu5e6vr+P3fXuSWKVdTSC1Zc+fA8OFRDjZ0EZVlROTXwM9wr6H6Ke7FOboD04HOwEfAj1W1JsI4jTEmuqJwMXdfdR2/v/slbrrnarq6qsl8520oKYlOfBEKu+UuIj2Ba4BSVR0GOIELgDuBKao6ENgBXBGNQI0xJpnsr6nj9/e8wsS//ZLu9ZVkvv0WjBiR6LAaRVqWyQByRCQDaAtsAk4CZnkenwb8IMLXMMaYpFJZU8+Nd7/EjXdOoEfdPjJnvwmlpYkOq4mwk7uqbgTuAr7GndR34S7D7FRtuARNGeC3g6eIjBeRJSKypKKiItwwjDEmrqpq6/njXc9z/eQJdNUq98XTo45KdFgHiaQskwecDfQDegDtgKCXFFHVqapaqqqlhXEeuWWMMeGorKnn5smzuP6OKylw1JP17jtJ12JvEElZ5mTgK1WtUNVa4L/AsUAnT5kGoBewMcIYjTHJLJJpexMhzHj3Vddx8+3PcMPkCXRs46TN/HeT5uKpP5Ek96+Bo0WkrYgIMBZYCcwFzvPscxnwYmQhGmOSWqotdxdGvHuqarnl1v/wh7/9kvZt25C9YF7C+7EHEknNfTHuC6dLcXeDdABTgRuA34jIGtzdIR+NQpzGmGSVatP2hhjvrspabv2/afxxyq/I6die7EUL4NBDI48jxt94Iuoto6q3qOqhqjpMVX+sqtWquk5VR6nqQFU9X1WroxWsMSYJpdq0vSHEu3N/DXde/zCT7ruGrIJ8cha9BwMHRieOGH/jsRGqxpjItDTSMxnr8UGOTN22t5r7rrmLWx7+HfTpQ9vFi6CoKHpxxPgbj00cZoyJTEsjPSNZRi9WghiZumlXJdMm/ImbZvyV/YeX0OHtN6Fz5+jGEcT0wZGwlrsxJnZSrR4PfFW2gekXTGDi9MnsG1pIh7dmRj+xQ8wXKrGWuzEmdmLcOo22Vd/u4t1Lfs+v353JrpJ2dLxuKyy7PDbfNmK8UIm13I0xsZNCy+h9tG4rS8+5lAnvzmT3sW3p+Jt9kFmfEt82/LGWuzEmdhK5jF4IFqz4hr3jLuLiVQvYc3Y/Opz/NQgp8W2jOdZyN8a0am++t4qsM87gtFUL2PuXO8l9ZhF0OSYlvm20xFruxphWa+YLixj+8wvov+Nb9k97kvaXXuJ+IAW+bQRiyd0Y0+qoKo/942VOv/4K8uoq0ddeo+2ppyQ6rKiy5G6MaVVq611Mve0xLpl8HdK2LVnvLsB5REmiw4o6S+7GmFZjb3Ud/7nqNn7++F/Y26sPee/OQaI56jSJWHI3xrQKFburmDNuAhP+N43ykaPpOvtVyMtLdFgxY8ndGJPeKstZ+/LPWfenLVzw2WK+/cG59JjxNGRlJTqymLLkboxJa4v/czVZt37BKRtX8+25hfQYX572iR0suRtj0thrM99m2PXz6bp3O9uu6kCPYytgZ1Wiw4qLiAYxiUgnEZklIp+LyCoRGS0i+SIyW0S+9PxM36KWMSY2Ipwq2OVSnv3TPxhzyVl0rNtP3U2ZdD52d0qPOA1VpCNU7wPeUNVDgWJgFTARmKOqg4A5nvvGGBO8CBayqKqp48WLr+W8W37B3p59aPfBHNodOzLyEae+Hzg7ViTfXPVeRFXDe6JIR2AZ0F+9DiIiXwAnqOomEekOvKOqh7R0rNLSUl2yZElYcRhj0tCzHdyJvUFGLozbfeB+Zbk74e9Y5m6Jj5kJOV0pL9/BqrMu4IQP32TdiWfQ76UZSPv20Ylp9vFNZ7h0toX6/QfuF4yO+8hWEflIVUv9PRZJy70fUAH8W0Q+FpF/iUg7oKuqbvLssxno2kxQ40VkiYgsqaioiCAMY0zaCbQUnp+W/YrFK9g28mhO+PBNvvzVDfSf80r0EjscPDd93e6knqs+kuSeAYwAHlbVI4B9+JRgPC16v18NVHWqqpaqamlhYWEEYRhj0o73VMH5I8FV07T84ZNo575ZR9exx1NU8Q1lj09n0P2TQSS6Mfl+4GR0SOq1YyNJ7mVAmaou9tyfhTvZl3vKMXh+boksRGNMq9MwVfC43eDIgu0fNa2/exJtnUt4fdaxHHvPh9S3bUvtgoX0uuxHsYnJd276U94Lfa76OK4pG3ZXSFXdLCLfiMghqvoFMBZY6fl3GTDZ8/PFqERqjGmd/C3Vd9aX7Hr9QlbctoPTl73H6hHH0u+NF8gsLIhdHP7mpg+1xh7HNWUj7ef+K+ApEckC1gE/wf1t4FkRuQLYAIyL8DWMMYnSzIXLuPKzVN/apRuRX65n9Ob1LB//G4Y//DdwpMDyFHFcUzai5K6qywB/V2rHRnJcY0yS8G1pzjvbXSbxTfax/BAYM7PJsT/4+DwO+7/jqHdm8OW0mQy/9NzovE48xHFN2bC7QkaTdYU0Jkn5dknEAeI40PoUJxQc477guf2jmHYLrKmq4YOLf8GY/z7Gmr6H0emV5ykY1mIv6+QT5Q/BlrpC2vQDxpjm+bY00QOJHUDr3Y+j7tsQk3JDxZfrKf/eDxnz5ccsOf1HFM/6N5ltc6L6GnERxzVlU6BIZYxJGN8eIvmlB7r/NdA6UI1Zt8BV02bhOOIIBny1ko//fB+lr01PzcQeZ9ZyN8YczLd8cNaXTWvrFe8BLve+kuHui+5bi4+Qq6qaZZdfzYgZ/+KrrkXseeUNjjjhqIiP21pYcjfGHKy5LnsNZYUY96LZtnwVO88+jxHrV7LgpHMpnvkoufkdo3b81sCSuzGtnb9EHajLXgxrx6unTKXHjb+mQBzMv+NhxtxwJRLt0aatgNXcjUlnwYyI9DcDY6C5XWKgfvceVpx2LoN/cyXru/ejYt4ijps4wRJ7mCy5G5POgpk6118r3fdCahRq6C3Z+s4CNg8expD/Pc/sc35GvxVLGHjksJi+Zrqzsowx6SyYEZH+BtbEq8tebS2rr5lI/3/cy9b2ecx/+BlOmRCjuWFaGWu5G5POgimvxLmV3mDvsuV8c2gxgx+5hwWlp1CzdBnfscQeNdZyNyad+Qzd95u44ziwBgCXiw23TKbb5Ftpl5nNKzffx2m3XE2G09qa0WTJ3Zh0Fu/EHUDN2q/Y/MML6Lv8AxYedjTtn/g3Z5YeGvsXToYJ0OLMPiqNSSdxnC88pBhUKb/3IeqGDiP/80+Z9YtbKF46j+HxSOwQ+pqsyfA+RsiSuzHpJIKFpWMVQ/2zZ1F25HF0/fUvWdVtAEtfeofzHppEu+zM+MUU6lS74byPSfaBYMndmHQSx/nCA8bggooX21P780/o9OlSnr70enp/8j7Hf3dU/GMKtd9+OO9jMnyweok4uYuI07NA9iue+/1EZLGIrBGRGZ6FPIwx4QqlRZiAwUf+Yqj9OpPym/IpnLGTD/sOY9FL73Lh45Pp0rFt/OOB0HsEhfM+JsMHq5dotNyvBVZ53b8TmKKqA4EdwBVReA1jWq9QWoQJ6tbYqLqaipdLkJtcZFTU8djFFzFk4Quc8t0jEzvS1HtN1oY5cloSzvuYDB+sXiLqLSMivYDvAbcDvxH3b+8k4CLPLtOAScDDkbyOMa1aKC3CBPaOqX1nHrsv+ymFX6/lteKxtHngXn56XIqOMg3nfQym22kcRdoV8l7geiDXc78zsFO1cTb/MqBnhK9hTOvU0H2vbt+BbUnQIjxIeTkVv7iWwudnUNmhkH/c8ADjbh5PXrtWVpH1/UBoKKclqPtl2GUZETkT2KKqH4X5/PEiskREllRUVIQbhjHpq6Ec0zhvujMxpZbm1Nez7+57qew/kI4vPscTJ1zI2rmLuXLy1a0vsfuT4AuskbTcjwW+LyJnANlAB+A+oJOIZHha772Ajf6erKpTgangXkM1gjiMSU/e5RgAZ9ukGZDkWriIXT8dT94XK3ivqJhVf/gLF1/2XXLqtya0tZpUEnyBNeyWu6reqKq9VLUIuAB4W1UvBuYC53l2uwx4MeIojWmNkuwCHQBbt7Lr4stwHHsM1Ru/5Z6f/YnCRfP42c/PICfLmfDWalJJ8O8vFv3cb8B9cXUN7hr8ozF4DWPSX6J7vnirraX6nilU9htIu2f+wxPHnMvCVxZw3T/+yOBuHQ7sl2TdARMqwb+/qMwto6rvAO94bq8DEjBKwZg0E4+eL4HmXFHF9dLL7Lvm1+R+vY4FfYt5f9LN/HTCWeT7q6v7mz64tUrwvD42QtWYZBTtoezNHa+lMsry5ew57kQcPzibLXtruO3KO2n77tv87rfn+k/sEFlrNdxzTrJh/8lCVBN/LbO0tFSXLFmS6DCMSR6zj2/aAi4YHVkrsLnjPdvBndgbZOTC8avZf8MfyH5yGrvatOPRsZfS/+bf8oMji3A4YjgQKdxzjvZ7lUJE5CNVLfX3mE35a0wyCrV2Hai80tzxvMso1U5q/1eAXj6QzOoqppV+nz2/v4EJ3y+lfZs4pIpw6/VW5/fLyjLGJKNQe1oE6qXS3PHGzIQOR+GanU3VdVlkPvEV7/Q6nNvueJaxrz/FNecfHZ/E3lKMsXpemrPkbkwyCrV2Haj16u949fXo9NeoHL8Bx+NVLOs4gN9f9xDtX3uZP13/Q/p0jvMkX+HW632fV/qQ1eCxmrsx6SGUuvP+zXDnSVT+8ytyNlWxvNtApp35c068+mLOOLxHbOrq8VwJqRXV4K3mbky6C2bSKlV44w2qrrmI7DU72Zjfi6nn/pDDz1AmX/Y7MmO5hmlD2UjrDpSNYpVwrQYPWHI3Jj1496n2bSUf+yy89QHVt9xKm2VL2dahkIe+9wu6nbCNW7pOpV2bTHDeH/2YvOOo3w9a794e64Rrfe0BS+7GJI9olS4aWsn1dfDae9SMP4Ss9bvZ3KkbU793LTmn5PKb/Ifo7NzuSX5HRvtMmsbhPT8OxD7hJnLq3SRaiNuSuzHJIlqli20fw8I6ap7PIGtjHWX57XnkrJ/T4YpLufakwXTJ2AUL3o198vOd+AwHZLSLfcJN5MjQeJafArDkbky8hNsXPVhVVfDkk9ROUjK/hfWdu/PI988l/+Rcfnf5H+iSm+3ZMTs+Cce3PJLGFzYbJVG935K7MfHi3aqrWAgvDwb0QKIPt1a8dSs8/DB19z9AxtYKPu86gH+e80MKj9nDxEM/p8vYxyEnO+BhQhbowyrJViaKiySq91tXSGPixXeof4OGVq2/ZNhSvXbtWvSee3A99m+cVZW83b+U/xx7PoMuOJMrxvSnS4cYJHRv0exymES16ojE+TysK6QxycC7Veet4et7MLViVVi0CNfddyPPP0+dI4P/DjmB/54wjhN+eCL3Ht2HDtmZgWOJRhKKZgkiiWrVEUnwTJDeLLkbE6lgEmVlObhqQF0cGBjesHxeEF/f9++HZ56h/sEHcS5bxr6c9jxx1HnMPeVHnHfmkTwxoidtMpzBxxyNZBrNEkQwHxTp0rqPE0vuxjQn2GQSTKJccD5s/4jGhO7N2bb5evS6dfDww9T/61GcO3ewtksRj3/3l6w79WwuP3UYE4Z0wxnOiNJotLqjWVMP5oMiXVr3cRJ2cheR3sATQFdAgamqep+I5AMzgCJgPTBOVXdEHqoxcRZsMgkmUR7ULdCbNv3QcLngzTfRBx6E11/DJQ5eHzSap888iy5nnMxlx/bjiD55oZ+P94cV4k6ikbS6o1mCCOaDIol6oqSCSFrudcBvVXWpiOQCH4nIbOByYI6qThaRicBE3EvvGZNagk0mwbQ6m6u34wTEfbG19lBY9R1cT87E8fUGdrTP48nRP+KNY8/mu6eO4N6j+nh1ZwxDk0FFTnefc+/eOokUzAdFEvVESQVhJ3dV3QRs8tzeIyKrgJ7A2cAJnt2m4V5+z5K7ST2BkklDS3j7UndpBRfkHeE/UXq3TDsOcW/btRJqgff2onNdsOJDRD9kUb8Spp81jk1jz+DH3xnEi8O6k5XRwrwvwZaPmnx7qAcUxu0OfJxkqXW3xq6VEYhKV0gRKQLmAcOAr1W1k2e7ADsa7vs8ZzwwHqBPnz4jN2zYEHEcxkRVoKQWSVfAFSvg0Udx/et+HHtdbOmQxzPDTuOV4jGUnnwcF47qzfBenYI7VrBxBNqvucdb0SyLqSamXSFFpD3wHHCdqu5253M3VVUR8fvpoapTgang7uceaRzGRJVvYi996OBEH2oN+Jtv4Jln0KeeQpYvp86ZweyBRzN9+HfZOySHHxXM4YUBz9Hu9KtCizXYOAK1fJs7TrRq3cnyDaCViCi5i0gm7sT+lKr+17O5XES6q+omEekObIk0SGPCEmoy8b3gWL//wMXU2cc2vb/g/OBqwDt3wqxZ8NRT6LvvIqp81vtQZp58Je8ecRInHN2PG+UODq19K3CpobnzCbYWHaiu3dxxolXrtt4ucRV2WcZTcpkGbFfV67y2/w3Y5nVBNV9Vr2/pWDZC1cREqOUE7/0DyciFs770n2z37oXXXoPp09FXX0Vqaigr7M2zhxzPq0NPoN/oYn5wRE9OPqwr2Zkh9E1v7nyi1SKOdc3d32Lc3jV/E7KWyjKRJPcxwHzgUw503v0DsBh4FugDbMDdFXJ7S8ey5G5i4qDh/j6zEvomqJamB3C2PdBy9/dBsWsXvPIKzJqFvvEGUlXFzg75/HfwGF4YcgLOUUdyzohenDm8B/ntsqJzPqmWHK12H3Uxqbmr6gKgudETY8M9rjFRk1finqALzyIRuJouIO2bWJp0V/TpKlj6ECy5qmnrdft2ePFFeO45dPZspKaG7Z0KeWXoybw6+Fi2FJfy/RF9uP+InhQVtIvO+aRyV8Bo9Haxun3QbOIwk74qy90zL9b5a9064JxvmyaGYBLHF1+4W+ivvILOn4/U11OR340XBxzNa4OPZcfhR3D60E6cUXMHQ2vfRPKbOU645xOPxJbMCdRa/03YxGGmdcppGDztj+tA6903mZ315YFkVlMDc16Gf12LLvwW2ew+3voeA3jtyHN4ffCx7Du8mO8N78Hth3fn0G65yFvfgT0xuHAYr0mpEn3hs6UPFxulGjRL7ia5RLvV2OzIUA4kBt9kNvMsqPoZzJ6Nvvkmsns3tRlOlvQdyqunHMfcAUfSeeggTjq0C3cN687gru3x7gKc8gko0fG39OGS6qWpOLLkbpJLoFZjqMnfu87r3b3ROzF88zEsr4MVoJ/WIVs+BD5ke34X3h5wDG/0O5KPig5hZOfPGZv7Ib/Ke5Wul7Qw6C7VE1Ci42/pw6W5un0yl5ISxGruJrkE6hESSc21IQFs/Bi+7Qs7xsL8xeiHixEXVGdl8mHfIczuezQLikqo6j+I7xzahZOr7uYY10yyZX9wr5nqiSbR8YfzO26ltXiruZvUEajVGE7JYPt2mD8f5s2DeZXo0v2I6zPqnV/wZZ9D+d/RFzK/Xwlre/amtHAzxx15MpcP7U9R57buckvlnbBg3YF5Yer2wdNOEIH8Ujj+xabJL1kWbAg3Sccr/ubiC6dXTaJLSUnIkruJjXASi/eCFuKE/JEH/2EHSv4uF6xaBYsXu/8tWgSffgpAXWYWq/oOYe7R43i/9zA+7XUYgwd0Y8zAAiYOKqC4dycynX4m6PJOdrOPh22L3bcV9+14XnAM5X1N9IXRQJqLL5wPl0SXkpKQJXcTfb5dECvmw/PdPAnbT0u3QZMFLTLAkXVgv+ZmYBz8ELz8sjuRL5yHLl6I7Hf3a69sl8tnvQ5l7nE/5oPeQ1nV+zAO79+FUf3yubpfPkf0ySMnK4QRotD8XO3xEkrCjnZrNtrlmmjGZzNGHsSSu4m+Bef771uu9S23dFv6Y59/HqxZBF/Vw9cO2JQPZWXw1eEAuBxONnTvweJDxvJBj6Es634I27vkUTygiFH98rmhXz6H9+oYeCm6QAksr8T9YeUtnq3EUBJitFuz0f4mEM34kqUUFow4XdOw5G5C592KFifggo5DwVXrSTZ+lpLzFmjRi7o6KHfC1i4wqzd8sQldX480fl642NxZWdGtF4tPOJGPex7C590H0i93I0fkrGRM22X8qu0MitruQX4U4vD8QAlszEyYdzZs+/BAzT2arcRgPlyCTYjRbs1Ge3bIYObBT0dxKpdZcjf+tZRkmqzo49FQhw5GQ0KqrobVq+GTRfC/22HNRtio6CaQunpgLXVOBxu6dGdZ0WA+6TKYlV37s6awN907VzJk8EiG9ujAqb06cnjPjmS/c2LTxJc/OvTzDpTAcrrCd98/+Hn+3i8IvYUWzIdLsAk72q3ZWMwO2Yp6tjSK08Vf6wppDqgsP9Aq9W59+/4BNjfBlj95o2DLflj9mXvy513doLIYXb0O1q5F6t31cZfAlrx8Vnfuy8r8/qwu6MNnXQewuSCfwe2+YUjOOobmfMWQ9psY1L0zbY6ffnCi9PeNoqFFGOw0A+F2qfP3PAj9WMFODpaI7oo2O2R0RLHbZkxmhYwmS+5xFCix+daTGzjbQf6IgwcDKbAf2Ob5t8XzbzPothwor0dqahoPU+d08m1+AV907seqvL6sKejDlwW9qeicR6/ccga1+YaBbbcwMHsjAzNW0zurHEfDei+FxzX/R9DcXOyBVhwKZoKwYBKYv4QFoSexaK2qlMxSOfZoiOIHsyX3dOP9n8N7Pc7m/qN4t2jrq2icJbHhD6vhq76/xF4N7HTCjjZQUQVbXY2J3LVNYBs4qpr+H6rKymRjXjfWdOrNVx27sSGvBxs6daMsvysZ+fX0zN5C3zabGNCmjEHZGxnYtoKujk0IPq3eJjM6OiEjx39LHFqei903qbY0tW+4icb39cXp/kCs23fgHDI6wFmrg18wpKU//FRu/SZykFSiB2hFmSX3dNBcy7QJTytU65uWJVw17i6GDS3tKmAvsAvYCez2/NwFugvqdzpw7XLg3OXCWX3wxdEd7XLZlFvA1x26sSm3gI0dCtmUW8imDgXs7tyeNv2G0KdzO/p0bkvf/Hb07dyWPvlt6T67Bxn1u/yfX0YHGlvPvvXq5lri3u9J/X73eftqruXe3DeUcJNkQywV73GgpNXQM8fnwzQardTW3voNV5q9bwkZoSoipwH34f4f/i9VnRyr10qo5lrR/lrUEHilG3/H2LniQItbgVrcCboKd0mk4XZlPezbje6F+r0O6vY4ce1djO51IHudOPcJmfvqcLj8f6DvymlHRds8trTLZ2vnTmzt3YmK9u77mzvksy8vB1e+UJizi26ZW+mRVUG3zO0ckrWUHhnldMvaRXaXkc3/seQPb751XbfbXXbxbkl51/j9XYBacL5P676BT6nF96LjmJnwQs+DPwwiuUjYcPGySYva5/jNXTwLpzXp3WsHdX+AV5andCs0LlrRSNaYtNxFxAmsBk4ByoAPgQtVdaW//cNquVeWs/Cl33LX6hFkSj1ZVJMhdWQ6lEypI0vqyaCaTKklywmZWkWm1JHhhCzP7Uyv7ZlOIZNKMqUOcWTgcNXgEBdORwai1Thw4ZAMHK4qnPX1OCULZ10djrpqnK4697Z6RV0K9YLWg9aL+++7XqBWoQYctQo1itSCw/NTal1QA1oLUgNSq0iN4qhVMqrqyayuI7OqjqzqWrKqa8moD9DVENiXmc3O7Fx25uSyM6c9O7Nz2ZXtvr0juwN7c3Koys2iukMWdR0c0MFBh+y9FGTspqBTJwpH/pbOX0ykYN8iCjK2kufcg1N8Xlcy3KNIHVnBJSa/rVuf4/lrSfm2thpe01/rW5xQcEzgBOmvjBLM8wLxjTXQCk7+nhPJRdwUboXGRZq9Z4louY8C1qjqOk8A04GzAb/JPSwLzidvYRkPPP0/QNwtWs/nlKg2uY+CoIgCqohnH0EbH3OoNm4Tzz547me4XDhd9WS5glhbMwwuhKrMLKqdWVRlZFGVmUWVsw1VmVnsy8phb/scKvPbUJmdTVWbLKqzsqjNzqA+20l9tgPNFshRHDmg7RRne6VNVg1tHdW0c+yno3MvhRk7GeT8ho7OSjq2qSPbUX/wQCPfBLf6Rcj21ysmwHJ1zWlo3TbXK6e5lpRv9z9XjfsP1B9n2+D+WP11KYxGq9f3uP4u0PoKtzXZilqhUdOKRrLGKrn3BL7xul8GHOW9g4iMB8YD9OnTJ/RX2LGMw/L3wICGA/r8w+e2Z8oQBVQEF0K9uz2OSxzu++K+rzhQkcZ91SG4HILL6cTl9Nx2OHA5HdQ7HajDQb3T/XwyBBwKTnEXpJyKOEEzQbMEMoEs0ExBshQyQTKULGc9WVJLptTSTirJl91kOerIlFoyfFvMkWioKXsn2OYmwPI3F3o0WjvefcV9W1L+yiK+/bW9yzTeQimrxGpEo7/jBnqdcPuP23wqoUulkawRStggJlWdCkwFd1km5APklcDgRTAotNa0O98rDpSMQCMpk4YDMtrTOBIU3DX5JhdWm+nS510C8U4AzQ3G8dbQyvHXbzxawmlJtbTWaSq2xMJtTbaiVqgJXaxq7qOBSar6Xc/9GwFU9Q5/+4dbc29MPA1JDiCjrXtWQRwHtjmzPRcko3i74XUkg4OSru9F0YZh+SLQqRgcme6LpA1xe29v7mKsv5JBMBfi0qzrF5Ce52RMGOLeFVJEMnBfUB0LbMR9QfUiVf3M3/7WFdIYY0IX9wuqqlonIlcD/8NdeX6sucRujDEm+mJWc1fV14DXYnV8Y4wxzfOz7IwxxphUZ8ndGGPSkCV3Y4xJQ5bcjTEmDSXFrJAiUgFsCLBbAbA1DuEko9Z87tC6z781nzu07vMP5tz7qmqhvweSIrkHQ0SWNNefM9215nOH1n3+rfncoXWff6TnbmUZY4xJQ5bcjTEmDaVScp+a6AASqDWfO7Tu82/N5w6t+/wjOveUqbkbY4wJXiq13I0xxgTJkrsxxqShpEvuInKaiHwhImtEZKKfx9uIyAzP44tFpCgBYcZEEOf+GxFZKSLLRWSOiPRNRJyxEuj8vfY7V0RURNKmi1ww5y4i4zy//89E5Ol4xxgrQfy/7yMic0XkY8///TMSEWcsiMhjIrJFRFY087iIyP2e92a5iIwI+uCqmjT/cE8PvBboD2QBnwBDfPa5CnjEc/sCYEai447juZ8ItPXc/kW6nHuw5+/ZLxeYB7wPlCY67jj+7gcBHwN5nvtdEh13HM99KvALz+0hwPpExx3F8z8eGAGsaObxM4DXcS8idzSwONhjJ1vLvXFhbVWtARoW1vZ2NjDNc3sWMFZEhNQX8NxVda6qepaX4n2gV5xjjKVgfvcAtwF3AlXxDC7Ggjn3nwN/V9UdAKq6Jc4xxkow565AB8/tjsC3cYwvplR1HrC9hV3OBp5Qt/eBTiLSPZhjJ1ty97ewds/m9lHVOmAX0Dku0cVWMOfu7Qrcn+jpIuD5e76S9lbVV+MZWBwE87sfDAwWkfdE5H0ROS1u0cVWMOc+CbhERMpwrxHxq/iElhRCzQuNErZAtgmfiFwClALfSXQs8SIiDuAe4PIEh5IoGbhLMyfg/sY2T0QOV9WdiQwqTi4EHlfVuz3rMz8pIsNUNVVWuE+IZGu5bwR6e93v5dnmdx/PWq0dgW1xiS62gjl3RORk4Cbg+6paHafY4iHQ+ecCw4B3RGQ97vrjS2lyUTWY330Z8JKq1qrqV7jXKB4Up/hiKZhzvwJ4FkBVFwHZuCfVag2Cygv+JFty/xAYJCL9RCQL9wXTl3z2eQm4zHP7POBt9Vx5SHEBz11EjgD+gTuxp0vNtUGL56+qu1S1QFWLVLUI9zWH76tqOqysHsz/+xdwt9oRkQLcZZp1cYwxVoI596+BsQAichju5F4R1ygT5yXgUk+vmaOBXaq6KahnJvpqcTNXh1fjvoJ+k2fbn3D/IYP7FzsTWAN8APRPdMxxPPe3gHJgmeffS4mOOZ7n77PvO6RJb5kgf/eCuyy1EvgUuCDRMcfx3IcA7+HuSbMMODXRMUfx3J8BNgG1uL+dXQFMACZ4/d7/7nlvPg3l/7xNP2CMMWko2coyxhhjosCSuzHGpCFL7sYYk4YsuRtjTBqy5G6MMWnIkrsxxqQhS+7GGJOG/h+rqyK36UZw/QAAAABJRU5ErkJggg==\n", "text/plain": [ "
" ] @@ -747,7 +747,7 @@ "name": "stdout", "output_type": "stream", "text": [ - "0.004999999999999996\n" + "0.004999999999999991\n" ] } ], @@ -1322,7 +1322,7 @@ "270 3344 160 110 270 Ds 7.253775 7.253775\n", "\n", "[267 rows x 6 columns]\n", - "0.009883615646716182\n" + "0.009883615646716186\n" ] } ], @@ -1402,6 +1402,48 @@ "name": "stderr", "output_type": "stream", "text": [ + "/Users/mhjensen/opt/anaconda3/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (100) reached and the optimization hasn't converged yet.\n", + " warnings.warn(\n", + "/Users/mhjensen/opt/anaconda3/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (100) reached and the optimization hasn't converged yet.\n", + " warnings.warn(\n", + "/Users/mhjensen/opt/anaconda3/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (100) reached and the optimization hasn't converged yet.\n", + " warnings.warn(\n", + "/Users/mhjensen/opt/anaconda3/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (100) reached and the optimization hasn't converged yet.\n", + " warnings.warn(\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/Users/mhjensen/opt/anaconda3/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (100) reached and the optimization hasn't converged yet.\n", + " warnings.warn(\n", + "/Users/mhjensen/opt/anaconda3/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (100) reached and the optimization hasn't converged yet.\n", + " warnings.warn(\n", + "/Users/mhjensen/opt/anaconda3/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (100) reached and the optimization hasn't converged yet.\n", + " warnings.warn(\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/Users/mhjensen/opt/anaconda3/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (100) reached and the optimization hasn't converged yet.\n", + " warnings.warn(\n", + "/Users/mhjensen/opt/anaconda3/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (100) reached and the optimization hasn't converged yet.\n", + " warnings.warn(\n", + "/Users/mhjensen/opt/anaconda3/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (100) reached and the optimization hasn't converged yet.\n", + " warnings.warn(\n", + "/Users/mhjensen/opt/anaconda3/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (100) reached and the optimization hasn't converged yet.\n", + " warnings.warn(\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/Users/mhjensen/opt/anaconda3/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (100) reached and the optimization hasn't converged yet.\n", + " warnings.warn(\n", "/Users/mhjensen/opt/anaconda3/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (100) reached and the optimization hasn't converged yet.\n", " warnings.warn(\n", "/Users/mhjensen/opt/anaconda3/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (100) reached and the optimization hasn't converged yet.\n", @@ -1452,8 +1494,6 @@ "name": "stderr", "output_type": "stream", "text": [ - "/Users/mhjensen/opt/anaconda3/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (100) reached and the optimization hasn't converged yet.\n", - " warnings.warn(\n", "/Users/mhjensen/opt/anaconda3/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (100) reached and the optimization hasn't converged yet.\n", " warnings.warn(\n", "/Users/mhjensen/opt/anaconda3/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (100) reached and the optimization hasn't converged yet.\n", @@ -1466,72 +1506,20 @@ "text": [ "/Users/mhjensen/opt/anaconda3/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (100) reached and the optimization hasn't converged yet.\n", " warnings.warn(\n", - "/Users/mhjensen/opt/anaconda3/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (100) reached and the optimization hasn't converged yet.\n", - " warnings.warn(\n", - "/Users/mhjensen/opt/anaconda3/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (100) reached and the optimization hasn't converged yet.\n", - " warnings.warn(\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "/Users/mhjensen/opt/anaconda3/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (100) reached and the optimization hasn't converged yet.\n", - " warnings.warn(\n", - "/Users/mhjensen/opt/anaconda3/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (100) reached and the optimization hasn't converged yet.\n", - " warnings.warn(\n", - "/Users/mhjensen/opt/anaconda3/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (100) reached and the optimization hasn't converged yet.\n", - " warnings.warn(\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "/Users/mhjensen/opt/anaconda3/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (100) reached and the optimization hasn't converged yet.\n", - " warnings.warn(\n", - "/Users/mhjensen/opt/anaconda3/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (100) reached and the optimization hasn't converged yet.\n", - " warnings.warn(\n", - "/Users/mhjensen/opt/anaconda3/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (100) reached and the optimization hasn't converged yet.\n", - " warnings.warn(\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "/Users/mhjensen/opt/anaconda3/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (100) reached and the optimization hasn't converged yet.\n", - " warnings.warn(\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "/Users/mhjensen/opt/anaconda3/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (100) reached and the optimization hasn't converged yet.\n", - " warnings.warn(\n", - "/Users/mhjensen/opt/anaconda3/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (100) reached and the optimization hasn't converged yet.\n", - " warnings.warn(\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ "/Users/mhjensen/opt/anaconda3/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (100) reached and the optimization hasn't converged yet.\n", " warnings.warn(\n" ] }, { "data": { - "image/png": "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\n", + "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "filenames": { - "image/png": "/Users/mhjensen/Teaching/MachineLearning/doc/LectureNotes/_build/jupyter_execute/chapter1_61_12.png" + "image/png": "/Users/mhjensen/Teaching/MachineLearning/doc/LectureNotes/_build/jupyter_execute/chapter1_61_10.png" } }, "output_type": "display_data" @@ -3278,15 +3266,15 @@ "name": "stdout", "output_type": "stream", "text": [ - "[ 2.02891525 -0.17480446 5.16545955]\n", + "[1.99543707 0.0761498 4.92956436]\n", "Training R2\n", - "0.9956684541910987\n", + "0.9944058125579283\n", "Training MSE\n", - "0.009129474532468922\n", + "0.010574954922481266\n", "Test R2\n", - "0.9944170482746688\n", + "0.9967910596862095\n", "Test MSE\n", - "0.009641081654050466\n" + "0.006369143720284525\n" ] } ], @@ -3390,13 +3378,13 @@ "output_type": "stream", "text": [ "Training R2\n", - "0.9999863415880615\n", + "0.9999864830619026\n", "Training MSE\n", - "7.039580368603554\n", + "6.55983997813135\n", "Test R2\n", - "0.9999670230644252\n", + "0.9999701967624061\n", "Test MSE\n", - "3.8118616620903856\n" + "6.146706341928796\n" ] } ], diff --git a/doc/LectureNotes/_build/jupyter_execute/chapter10.ipynb b/doc/LectureNotes/_build/jupyter_execute/chapter10.ipynb index e0377854a..398d4d5cf 100644 --- a/doc/LectureNotes/_build/jupyter_execute/chapter10.ipynb +++ b/doc/LectureNotes/_build/jupyter_execute/chapter10.ipynb @@ -2284,15 +2284,15 @@ "Learning rate = 0.1\n", "Lambda = 1e-05\n", "Accuracy score on test set: 0.9111111111111111\n", - "\n" + "\n", + "Learning rate = 0.1\n", + "Lambda = 0.0001\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ - "Learning rate = 0.1\n", - "Lambda = 0.0001\n", "Accuracy score on test set: 0.9222222222222223\n", "\n", "Learning rate = 0.1\n", diff --git a/doc/LectureNotes/_build/jupyter_execute/chapter1_11_0.png b/doc/LectureNotes/_build/jupyter_execute/chapter1_11_0.png index c5d5c6d46..0176a67d5 100644 Binary files a/doc/LectureNotes/_build/jupyter_execute/chapter1_11_0.png and b/doc/LectureNotes/_build/jupyter_execute/chapter1_11_0.png differ diff --git a/doc/LectureNotes/_build/jupyter_execute/chapter1_13_1.png b/doc/LectureNotes/_build/jupyter_execute/chapter1_13_1.png index 6b2662505..b09cce127 100644 Binary files a/doc/LectureNotes/_build/jupyter_execute/chapter1_13_1.png and b/doc/LectureNotes/_build/jupyter_execute/chapter1_13_1.png differ diff --git a/doc/LectureNotes/_build/jupyter_execute/chapter1_27_0.png b/doc/LectureNotes/_build/jupyter_execute/chapter1_27_0.png index c14dab1e4..1e8daf4fa 100644 Binary files a/doc/LectureNotes/_build/jupyter_execute/chapter1_27_0.png and b/doc/LectureNotes/_build/jupyter_execute/chapter1_27_0.png differ diff --git a/doc/LectureNotes/_build/jupyter_execute/chapter1_3_0.png b/doc/LectureNotes/_build/jupyter_execute/chapter1_3_0.png index fc2b30307..705b45f02 100644 Binary files a/doc/LectureNotes/_build/jupyter_execute/chapter1_3_0.png and b/doc/LectureNotes/_build/jupyter_execute/chapter1_3_0.png differ diff --git a/doc/LectureNotes/_build/jupyter_execute/chapter1_61_10.png b/doc/LectureNotes/_build/jupyter_execute/chapter1_61_10.png index febd56e9a..90390bb17 100644 Binary files a/doc/LectureNotes/_build/jupyter_execute/chapter1_61_10.png and b/doc/LectureNotes/_build/jupyter_execute/chapter1_61_10.png differ diff --git a/doc/LectureNotes/chapter3.ipynb b/doc/LectureNotes/chapter3.ipynb index ad19ad998..4f608ee8b 100644 --- a/doc/LectureNotes/chapter3.ipynb +++ b/doc/LectureNotes/chapter3.ipynb @@ -1708,42 +1708,6 @@ "$$" ] }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "We assume \n", - "that every column of $\\boldsymbol{X}$ is centered, which we can do by subtracting the mean," - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": { - "collapsed": false, - "editable": true - }, - "outputs": [], - "source": [ - "X = X - np.mean(X,axis=0)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "This means that we need to rewrite $X_{ij}$ as $\\tilde{X}_{ij}=X_{ij}-\\mu_j$, where" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "$$\n", - "\\mu_j = \\frac{1}{n}\\sum_{i=0}^{n-1}X_{ij}.\n", - "$$" - ] - }, { "cell_type": "markdown", "metadata": {}, @@ -1765,7 +1729,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "Assuming that the matrix elements $X_{i1}$ are centered, what we have is" + "We obtain then" ] }, { @@ -1773,7 +1737,7 @@ "metadata": {}, "source": [ "$$\n", - "\\beta_0 = \\frac{1}{n}\\sum_{i=0}^{n-1}y_i - \\beta_1\\frac{1}{n}\\sum_{i=0}^{n-1} \\left(X_{i1}-\\mu_{1}\\right),\n", + "\\beta_0 = \\frac{1}{n}\\sum_{i=0}^{n-1}y_i - \\beta_1\\frac{1}{n}\\sum_{i=0}^{n-1} X_{i1}.\n", "$$" ] }, @@ -1781,7 +1745,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "where" + "If we define" ] }, { @@ -1821,7 +1785,7 @@ "metadata": {}, "source": [ "$$\n", - "\\beta_0 = \\mu_y - \\beta_1\\frac{1}{n}\\sum_{i=0}^{n-1} (X_{i1}-\\mu_{1}),\n", + "\\beta_0 = \\mu_y - \\beta_1\\mu_{1}.\n", "$$" ] }, @@ -1829,7 +1793,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "and it is easy to see that the last sum equals zero! This means that we have" + "In the general case withmore parameters than $\\beta_0$ and $\\beta_1$, we have" ] }, { @@ -1837,7 +1801,7 @@ "metadata": {}, "source": [ "$$\n", - "\\beta_0 = \\mu_y,\n", + "\\beta_0 = \\frac{1}{n}\\sum_{i=0}^{n-1}y_i - \\frac{1}{n}\\sum_{i=0}^{n-1}\\sum_{j=1}^{p-1} X_{ij}\\beta_j.\n", "$$" ] }, @@ -1845,26 +1809,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "if the columns of the design matrix are centered. It is straight forward to generalize this results to more values of $\\beta$.\n", - "We have thus" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "$$\n", - "\\beta_0 = \\frac{1}{n}\\sum_{i=0}^{n-1} y_i = \\overline{\\boldsymbol{y}},\n", - "$$" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "the average value of $\\boldsymbol{y}$.\n", - "\n", - "Replacing $y_i$ with $y_i - \\beta_0 = y_i - \\overline{\\boldsymbol{y}}$ and centering also our design matrix results in a cost function (in vector-matrix disguise)" + "Replacing $y_i$ with $y_i - y_i - \\overline{\\boldsymbol{y}}$ and centering also our design matrix results in a cost function (in vector-matrix disguise)" ] }, { @@ -2238,8 +2183,6 @@ " OwnRidgeBeta = np.linalg.pinv(X_train_scaled.T @ X_train_scaled+lmb*I) @ X_train_scaled.T @ (y_train_scaled)\n", " intercept_ = y_scaler - X_train_mean@OwnRidgeBeta #The intercept can be shifted so the model can predict on uncentered data\n", " #Add intercept to prediction\n", - " ypredictOwnRidge = X_test @ OwnRidgeBeta + intercept_ \n", - " #Add intercept to prediction\n", " ypredictOwnRidge = X_test_scaled @ OwnRidgeBeta + y_scaler \n", " RegRidge = linear_model.Ridge(lmb)\n", " RegRidge.fit(X_train,y_train)\n", @@ -2925,8 +2868,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "6\n", - "0\n", + "5\n", + "8\n", " \n", "<\n", "<\n",