Pushing the docs to dev/ for branch: master, commit 3379e90f6dd658881d828f1d65d3e01d2140eb1a

Author: sklearn-ci

dev/_downloads/00701bf1048deb8daeb5ad086596d260/plot_lasso_lars.ipynb

@@ -26,7 +26,7 @@
       },
       "outputs": [],
       "source": [
-        "print(__doc__)\n\n# Author: Fabian Pedregosa <[email protected]>\n#         Alexandre Gramfort <[email protected]>\n# License: BSD 3 clause\n\nimport numpy as np\nimport matplotlib.pyplot as plt\n\nfrom sklearn import linear_model\nfrom sklearn import datasets\n\ndiabetes = datasets.load_diabetes()\nX = diabetes.data\ny = diabetes.target\n\nprint(\"Computing regularization path using the LARS ...\")\n_, _, coefs = linear_model.lars_path(X, y, method='lasso', verbose=True)\n\nxx = np.sum(np.abs(coefs.T), axis=1)\nxx /= xx[-1]\n\nplt.plot(xx, coefs.T)\nymin, ymax = plt.ylim()\nplt.vlines(xx, ymin, ymax, linestyle='dashed')\nplt.xlabel('|coef| / max|coef|')\nplt.ylabel('Coefficients')\nplt.title('LASSO Path')\nplt.axis('tight')\nplt.show()"
+        "print(__doc__)\n\n# Author: Fabian Pedregosa <[email protected]>\n#         Alexandre Gramfort <[email protected]>\n# License: BSD 3 clause\n\nimport numpy as np\nimport matplotlib.pyplot as plt\n\nfrom sklearn import linear_model\nfrom sklearn import datasets\n\nX, y = datasets.load_diabetes(return_X_y=True)\n\nprint(\"Computing regularization path using the LARS ...\")\n_, _, coefs = linear_model.lars_path(X, y, method='lasso', verbose=True)\n\nxx = np.sum(np.abs(coefs.T), axis=1)\nxx /= xx[-1]\n\nplt.plot(xx, coefs.T)\nymin, ymax = plt.ylim()\nplt.vlines(xx, ymin, ymax, linestyle='dashed')\nplt.xlabel('|coef| / max|coef|')\nplt.ylabel('Coefficients')\nplt.title('LASSO Path')\nplt.axis('tight')\nplt.show()"
       ]
     }
   ],

dev/_downloads/206285a20b3b1231a7792b027515ea8f/plot_lasso_model_selection.ipynb

@@ -26,7 +26,7 @@
       },
       "outputs": [],
       "source": [
-        "print(__doc__)\n\n# Author: Olivier Grisel, Gael Varoquaux, Alexandre Gramfort\n# License: BSD 3 clause\n\nimport time\n\nimport numpy as np\nimport matplotlib.pyplot as plt\n\nfrom sklearn.linear_model import LassoCV, LassoLarsCV, LassoLarsIC\nfrom sklearn import datasets\n\n# This is to avoid division by zero while doing np.log10\nEPSILON = 1e-4\n\ndiabetes = datasets.load_diabetes()\nX = diabetes.data\ny = diabetes.target\n\nrng = np.random.RandomState(42)\nX = np.c_[X, rng.randn(X.shape[0], 14)]  # add some bad features\n\n# normalize data as done by Lars to allow for comparison\nX /= np.sqrt(np.sum(X ** 2, axis=0))\n\n# #############################################################################\n# LassoLarsIC: least angle regression with BIC/AIC criterion\n\nmodel_bic = LassoLarsIC(criterion='bic')\nt1 = time.time()\nmodel_bic.fit(X, y)\nt_bic = time.time() - t1\nalpha_bic_ = model_bic.alpha_\n\nmodel_aic = LassoLarsIC(criterion='aic')\nmodel_aic.fit(X, y)\nalpha_aic_ = model_aic.alpha_\n\n\ndef plot_ic_criterion(model, name, color):\n    alpha_ = model.alpha_ + EPSILON\n    alphas_ = model.alphas_ + EPSILON\n    criterion_ = model.criterion_\n    plt.plot(-np.log10(alphas_), criterion_, '--', color=color,\n             linewidth=3, label='%s criterion' % name)\n    plt.axvline(-np.log10(alpha_), color=color, linewidth=3,\n                label='alpha: %s estimate' % name)\n    plt.xlabel('-log(alpha)')\n    plt.ylabel('criterion')\n\nplt.figure()\nplot_ic_criterion(model_aic, 'AIC', 'b')\nplot_ic_criterion(model_bic, 'BIC', 'r')\nplt.legend()\nplt.title('Information-criterion for model selection (training time %.3fs)'\n          % t_bic)\n\n# #############################################################################\n# LassoCV: coordinate descent\n\n# Compute paths\nprint(\"Computing regularization path using the coordinate descent lasso...\")\nt1 = time.time()\nmodel = LassoCV(cv=20).fit(X, y)\nt_lasso_cv = time.time() - t1\n\n# Display results\nm_log_alphas = -np.log10(model.alphas_ + EPSILON)\n\nplt.figure()\nymin, ymax = 2300, 3800\nplt.plot(m_log_alphas, model.mse_path_, ':')\nplt.plot(m_log_alphas, model.mse_path_.mean(axis=-1), 'k',\n         label='Average across the folds', linewidth=2)\nplt.axvline(-np.log10(model.alpha_ + EPSILON), linestyle='--', color='k',\n            label='alpha: CV estimate')\n\nplt.legend()\n\nplt.xlabel('-log(alpha)')\nplt.ylabel('Mean square error')\nplt.title('Mean square error on each fold: coordinate descent '\n          '(train time: %.2fs)' % t_lasso_cv)\nplt.axis('tight')\nplt.ylim(ymin, ymax)\n\n# #############################################################################\n# LassoLarsCV: least angle regression\n\n# Compute paths\nprint(\"Computing regularization path using the Lars lasso...\")\nt1 = time.time()\nmodel = LassoLarsCV(cv=20).fit(X, y)\nt_lasso_lars_cv = time.time() - t1\n\n# Display results\nm_log_alphas = -np.log10(model.cv_alphas_ + EPSILON)\n\nplt.figure()\nplt.plot(m_log_alphas, model.mse_path_, ':')\nplt.plot(m_log_alphas, model.mse_path_.mean(axis=-1), 'k',\n         label='Average across the folds', linewidth=2)\nplt.axvline(-np.log10(model.alpha_), linestyle='--', color='k',\n            label='alpha CV')\nplt.legend()\n\nplt.xlabel('-log(alpha)')\nplt.ylabel('Mean square error')\nplt.title('Mean square error on each fold: Lars (train time: %.2fs)'\n          % t_lasso_lars_cv)\nplt.axis('tight')\nplt.ylim(ymin, ymax)\n\nplt.show()"
+        "print(__doc__)\n\n# Author: Olivier Grisel, Gael Varoquaux, Alexandre Gramfort\n# License: BSD 3 clause\n\nimport time\n\nimport numpy as np\nimport matplotlib.pyplot as plt\n\nfrom sklearn.linear_model import LassoCV, LassoLarsCV, LassoLarsIC\nfrom sklearn import datasets\n\n# This is to avoid division by zero while doing np.log10\nEPSILON = 1e-4\n\nX, y = datasets.load_diabetes(return_X_y=True)\n\nrng = np.random.RandomState(42)\nX = np.c_[X, rng.randn(X.shape[0], 14)]  # add some bad features\n\n# normalize data as done by Lars to allow for comparison\nX /= np.sqrt(np.sum(X ** 2, axis=0))\n\n# #############################################################################\n# LassoLarsIC: least angle regression with BIC/AIC criterion\n\nmodel_bic = LassoLarsIC(criterion='bic')\nt1 = time.time()\nmodel_bic.fit(X, y)\nt_bic = time.time() - t1\nalpha_bic_ = model_bic.alpha_\n\nmodel_aic = LassoLarsIC(criterion='aic')\nmodel_aic.fit(X, y)\nalpha_aic_ = model_aic.alpha_\n\n\ndef plot_ic_criterion(model, name, color):\n    alpha_ = model.alpha_ + EPSILON\n    alphas_ = model.alphas_ + EPSILON\n    criterion_ = model.criterion_\n    plt.plot(-np.log10(alphas_), criterion_, '--', color=color,\n             linewidth=3, label='%s criterion' % name)\n    plt.axvline(-np.log10(alpha_), color=color, linewidth=3,\n                label='alpha: %s estimate' % name)\n    plt.xlabel('-log(alpha)')\n    plt.ylabel('criterion')\n\n\nplt.figure()\nplot_ic_criterion(model_aic, 'AIC', 'b')\nplot_ic_criterion(model_bic, 'BIC', 'r')\nplt.legend()\nplt.title('Information-criterion for model selection (training time %.3fs)'\n          % t_bic)\n\n# #############################################################################\n# LassoCV: coordinate descent\n\n# Compute paths\nprint(\"Computing regularization path using the coordinate descent lasso...\")\nt1 = time.time()\nmodel = LassoCV(cv=20).fit(X, y)\nt_lasso_cv = time.time() - t1\n\n# Display results\nm_log_alphas = -np.log10(model.alphas_ + EPSILON)\n\nplt.figure()\nymin, ymax = 2300, 3800\nplt.plot(m_log_alphas, model.mse_path_, ':')\nplt.plot(m_log_alphas, model.mse_path_.mean(axis=-1), 'k',\n         label='Average across the folds', linewidth=2)\nplt.axvline(-np.log10(model.alpha_ + EPSILON), linestyle='--', color='k',\n            label='alpha: CV estimate')\n\nplt.legend()\n\nplt.xlabel('-log(alpha)')\nplt.ylabel('Mean square error')\nplt.title('Mean square error on each fold: coordinate descent '\n          '(train time: %.2fs)' % t_lasso_cv)\nplt.axis('tight')\nplt.ylim(ymin, ymax)\n\n# #############################################################################\n# LassoLarsCV: least angle regression\n\n# Compute paths\nprint(\"Computing regularization path using the Lars lasso...\")\nt1 = time.time()\nmodel = LassoLarsCV(cv=20).fit(X, y)\nt_lasso_lars_cv = time.time() - t1\n\n# Display results\nm_log_alphas = -np.log10(model.cv_alphas_ + EPSILON)\n\nplt.figure()\nplt.plot(m_log_alphas, model.mse_path_, ':')\nplt.plot(m_log_alphas, model.mse_path_.mean(axis=-1), 'k',\n         label='Average across the folds', linewidth=2)\nplt.axvline(-np.log10(model.alpha_), linestyle='--', color='k',\n            label='alpha CV')\nplt.legend()\n\nplt.xlabel('-log(alpha)')\nplt.ylabel('Mean square error')\nplt.title('Mean square error on each fold: Lars (train time: %.2fs)'\n          % t_lasso_lars_cv)\nplt.axis('tight')\nplt.ylim(ymin, ymax)\n\nplt.show()"
       ]
     }
   ],

dev/_downloads/3409d9766d352cc9f9b169d4a799a87a/auto_examples_python.zip

@@ -26,7 +26,7 @@
       },
       "outputs": [],
       "source": [
-        "print(__doc__)\n\n\n# Code source: Jaques Grobler\n# License: BSD 3 clause\n\n\nimport matplotlib.pyplot as plt\nimport numpy as np\nfrom sklearn import datasets, linear_model\nfrom sklearn.metrics import mean_squared_error, r2_score\n\n# Load the diabetes dataset\ndiabetes = datasets.load_diabetes()\n\n\n# Use only one feature\ndiabetes_X = diabetes.data[:, np.newaxis, 2]\n\n# Split the data into training/testing sets\ndiabetes_X_train = diabetes_X[:-20]\ndiabetes_X_test = diabetes_X[-20:]\n\n# Split the targets into training/testing sets\ndiabetes_y_train = diabetes.target[:-20]\ndiabetes_y_test = diabetes.target[-20:]\n\n# Create linear regression object\nregr = linear_model.LinearRegression()\n\n# Train the model using the training sets\nregr.fit(diabetes_X_train, diabetes_y_train)\n\n# Make predictions using the testing set\ndiabetes_y_pred = regr.predict(diabetes_X_test)\n\n# The coefficients\nprint('Coefficients: \\n', regr.coef_)\n# The mean squared error\nprint('Mean squared error: %.2f'\n      % mean_squared_error(diabetes_y_test, diabetes_y_pred))\n# The coefficient of determination: 1 is perfect prediction\nprint('Coefficient of determination: %.2f'\n      % r2_score(diabetes_y_test, diabetes_y_pred))\n\n# Plot outputs\nplt.scatter(diabetes_X_test, diabetes_y_test,  color='black')\nplt.plot(diabetes_X_test, diabetes_y_pred, color='blue', linewidth=3)\n\nplt.xticks(())\nplt.yticks(())\n\nplt.show()"
+        "print(__doc__)\n\n\n# Code source: Jaques Grobler\n# License: BSD 3 clause\n\n\nimport matplotlib.pyplot as plt\nimport numpy as np\nfrom sklearn import datasets, linear_model\nfrom sklearn.metrics import mean_squared_error, r2_score\n\n# Load the diabetes dataset\ndiabetes_X, diabetes_y = datasets.load_diabetes(return_X_y=True)\n\n# Use only one feature\ndiabetes_X = diabetes_X[:, np.newaxis, 2]\n\n# Split the data into training/testing sets\ndiabetes_X_train = diabetes_X[:-20]\ndiabetes_X_test = diabetes_X[-20:]\n\n# Split the targets into training/testing sets\ndiabetes_y_train = diabetes_y[:-20]\ndiabetes_y_test = diabetes_y[-20:]\n\n# Create linear regression object\nregr = linear_model.LinearRegression()\n\n# Train the model using the training sets\nregr.fit(diabetes_X_train, diabetes_y_train)\n\n# Make predictions using the testing set\ndiabetes_y_pred = regr.predict(diabetes_X_test)\n\n# The coefficients\nprint('Coefficients: \\n', regr.coef_)\n# The mean squared error\nprint('Mean squared error: %.2f'\n      % mean_squared_error(diabetes_y_test, diabetes_y_pred))\n# The coefficient of determination: 1 is perfect prediction\nprint('Coefficient of determination: %.2f'\n      % r2_score(diabetes_y_test, diabetes_y_pred))\n\n# Plot outputs\nplt.scatter(diabetes_X_test, diabetes_y_test,  color='black')\nplt.plot(diabetes_X_test, diabetes_y_pred, color='blue', linewidth=3)\n\nplt.xticks(())\nplt.yticks(())\n\nplt.show()"
       ]
     }
   ],

dev/_downloads/3bc388ac00a41366cf48b6e294838489/plot_ols.ipynb

@@ -29,19 +29,18 @@
 from sklearn.metrics import mean_squared_error, r2_score
 
 # Load the diabetes dataset
-diabetes = datasets.load_diabetes()
-
+diabetes_X, diabetes_y = datasets.load_diabetes(return_X_y=True)
 
 # Use only one feature
-diabetes_X = diabetes.data[:, np.newaxis, 2]
+diabetes_X = diabetes_X[:, np.newaxis, 2]
 
 # Split the data into training/testing sets
 diabetes_X_train = diabetes_X[:-20]
 diabetes_X_test = diabetes_X[-20:]
 
 # Split the targets into training/testing sets
-diabetes_y_train = diabetes.target[:-20]
-diabetes_y_test = diabetes.target[-20:]
+diabetes_y_train = diabetes_y[:-20]
+diabetes_y_test = diabetes_y[-20:]
 
 # Create linear regression object
 regr = linear_model.LinearRegression()

dev/_downloads/525570147c5d8b2eba26233cf65c8da7/plot_ols.py

@@ -57,9 +57,7 @@
 # This is to avoid division by zero while doing np.log10
 EPSILON = 1e-4
 
-diabetes = datasets.load_diabetes()
-X = diabetes.data
-y = diabetes.target
+X, y = datasets.load_diabetes(return_X_y=True)
 
 rng = np.random.RandomState(42)
 X = np.c_[X, rng.randn(X.shape[0], 14)]  # add some bad features
@@ -92,6 +90,7 @@ def plot_ic_criterion(model, name, color):
     plt.xlabel('-log(alpha)')
     plt.ylabel('criterion')
 
+
 plt.figure()
 plot_ic_criterion(model_aic, 'AIC', 'b')
 plot_ic_criterion(model_bic, 'BIC', 'r')

dev/_downloads/64cb62fe8f8f5f93b1b63fb7d0d85095/plot_lasso_model_selection.py

@@ -20,9 +20,9 @@
 from sklearn.model_selection import KFold
 from sklearn.model_selection import GridSearchCV
 
-diabetes = datasets.load_diabetes()
-X = diabetes.data[:150]
-y = diabetes.target[:150]
+X, y = datasets.load_diabetes(return_X_y=True)
+X = X[:150]
+y = y[:150]
 
 lasso = Lasso(random_state=0, max_iter=10000)
 alphas = np.logspace(-4, -0.5, 30)

dev/_downloads/703c77d8ad4ea9fea3ccbab3a691d9d8/plot_cv_diabetes.py

@@ -22,9 +22,7 @@
 from sklearn import linear_model
 from sklearn import datasets
 
-diabetes = datasets.load_diabetes()
-X = diabetes.data
-y = diabetes.target
+X, y = datasets.load_diabetes(return_X_y=True)
 
 print("Computing regularization path using the LARS ...")
 _, _, coefs = linear_model.lars_path(X, y, method='lasso', verbose=True)

dev/_downloads/79e4690100f4bd29f64fb06c544129d6/plot_lasso_lars.py

@@ -25,13 +25,13 @@
 
 from sklearn import datasets, linear_model
 
-diabetes = datasets.load_diabetes()
+X, y = datasets.load_diabetes(return_X_y=True)
 indices = (0, 1)
 
-X_train = diabetes.data[:-20, indices]
-X_test = diabetes.data[-20:, indices]
-y_train = diabetes.target[:-20]
-y_test = diabetes.target[-20:]
+X_train = X[:-20, indices]
+X_test = X[-20:, indices]
+y_train = y[:-20]
+y_test = y[-20:]
 
 ols = linear_model.LinearRegression()
 ols.fit(X_train, y_train)
@@ -58,7 +58,8 @@ def plot_figs(fig_num, elev, azim, X_train, clf):
     ax.w_yaxis.set_ticklabels([])
     ax.w_zaxis.set_ticklabels([])
 
-#Generate the three different figures from different views
+
+# Generate the three different figures from different views
 elev = 43.5
 azim = -110
 plot_figs(1, elev, azim, X_train, ols)

dev/_downloads/915460448434dcfa808fb3b4305bc4fa/plot_ols_3d.py

@@ -20,9 +20,9 @@
 from sklearn.linear_model import lasso_path, enet_path
 from sklearn import datasets
 
-diabetes = datasets.load_diabetes()
-X = diabetes.data
-y = diabetes.target
+
+X, y = datasets.load_diabetes(return_X_y=True)
+
 
 X /= X.std(axis=0)  # Standardize data (easier to set the l1_ratio parameter)