Pushing the docs to dev/ for branch: master, commit e33e84dbf7281c4e090df176b7d82e7152a1078f

Author: sklearn-ci

dev/_downloads/auto_examples_jupyter.zip

@@ -26,7 +26,7 @@
       },
       "outputs": [],
       "source": [
-        "import time\nimport matplotlib.pyplot as plt\nimport numpy as np\n\nfrom sklearn.datasets import fetch_mldata\nfrom sklearn.linear_model import LogisticRegression\nfrom sklearn.model_selection import train_test_split\nfrom sklearn.preprocessing import StandardScaler\nfrom sklearn.utils import check_random_state\n\nprint(__doc__)\n\n# Author: Arthur Mensch <[email protected]>\n# License: BSD 3 clause\n\n# Turn down for faster convergence\nt0 = time.time()\ntrain_samples = 5000\n\nmnist = fetch_mldata('MNIST original')\nX = mnist.data.astype('float64')\ny = mnist.target\nrandom_state = check_random_state(0)\npermutation = random_state.permutation(X.shape[0])\nX = X[permutation]\ny = y[permutation]\nX = X.reshape((X.shape[0], -1))\n\nX_train, X_test, y_train, y_test = train_test_split(\n    X, y, train_size=train_samples, test_size=10000)\n\nscaler = StandardScaler()\nX_train = scaler.fit_transform(X_train)\nX_test = scaler.transform(X_test)\n\n# Turn up tolerance for faster convergence\nclf = LogisticRegression(C=50 / train_samples,\n                         multi_class='multinomial',\n                         penalty='l1', solver='saga', tol=0.1)\nclf.fit(X_train, y_train)\nsparsity = np.mean(clf.coef_ == 0) * 100\nscore = clf.score(X_test, y_test)\n# print('Best C % .4f' % clf.C_)\nprint(\"Sparsity with L1 penalty: %.2f%%\" % sparsity)\nprint(\"Test score with L1 penalty: %.4f\" % score)\n\ncoef = clf.coef_.copy()\nplt.figure(figsize=(10, 5))\nscale = np.abs(coef).max()\nfor i in range(10):\n    l1_plot = plt.subplot(2, 5, i + 1)\n    l1_plot.imshow(coef[i].reshape(28, 28), interpolation='nearest',\n                   cmap=plt.cm.RdBu, vmin=-scale, vmax=scale)\n    l1_plot.set_xticks(())\n    l1_plot.set_yticks(())\n    l1_plot.set_xlabel('Class %i' % i)\nplt.suptitle('Classification vector for...')\n\nrun_time = time.time() - t0\nprint('Example run in %.3f s' % run_time)\nplt.show()"
+        "import time\nimport matplotlib.pyplot as plt\nimport numpy as np\n\nfrom sklearn.datasets import fetch_mldata\nfrom sklearn.linear_model import LogisticRegression\nfrom sklearn.model_selection import train_test_split\nfrom sklearn.preprocessing import StandardScaler\nfrom sklearn.utils import check_random_state\n\nprint(__doc__)\n\n# Author: Arthur Mensch <[email protected]>\n# License: BSD 3 clause\n\n# Turn down for faster convergence\nt0 = time.time()\ntrain_samples = 5000\n\nmnist = fetch_mldata('MNIST original')\nX = mnist.data.astype('float64')\ny = mnist.target\nrandom_state = check_random_state(0)\npermutation = random_state.permutation(X.shape[0])\nX = X[permutation]\ny = y[permutation]\nX = X.reshape((X.shape[0], -1))\n\nX_train, X_test, y_train, y_test = train_test_split(\n    X, y, train_size=train_samples, test_size=10000)\n\nscaler = StandardScaler()\nX_train = scaler.fit_transform(X_train)\nX_test = scaler.transform(X_test)\n\n# Turn up tolerance for faster convergence\nclf = LogisticRegression(C=50. / train_samples,\n                         multi_class='multinomial',\n                         penalty='l1', solver='saga', tol=0.1)\nclf.fit(X_train, y_train)\nsparsity = np.mean(clf.coef_ == 0) * 100\nscore = clf.score(X_test, y_test)\n# print('Best C % .4f' % clf.C_)\nprint(\"Sparsity with L1 penalty: %.2f%%\" % sparsity)\nprint(\"Test score with L1 penalty: %.4f\" % score)\n\ncoef = clf.coef_.copy()\nplt.figure(figsize=(10, 5))\nscale = np.abs(coef).max()\nfor i in range(10):\n    l1_plot = plt.subplot(2, 5, i + 1)\n    l1_plot.imshow(coef[i].reshape(28, 28), interpolation='nearest',\n                   cmap=plt.cm.RdBu, vmin=-scale, vmax=scale)\n    l1_plot.set_xticks(())\n    l1_plot.set_yticks(())\n    l1_plot.set_xlabel('Class %i' % i)\nplt.suptitle('Classification vector for...')\n\nrun_time = time.time() - t0\nprint('Example run in %.3f s' % run_time)\nplt.show()"
       ]
     }
   ],

dev/_downloads/auto_examples_python.zip

@@ -52,7 +52,7 @@
 X_test = scaler.transform(X_test)
 
 # Turn up tolerance for faster convergence
-clf = LogisticRegression(C=50 / train_samples,
+clf = LogisticRegression(C=50. / train_samples,
                          multi_class='multinomial',
                          penalty='l1', solver='saga', tol=0.1)
 clf.fit(X_train, y_train)

dev/_downloads/plot_sparse_logistic_regression_mnist.ipynb

@@ -26,7 +26,7 @@
       },
       "outputs": [],
       "source": [
-        "print(__doc__)\n\n# Author: Emmanuelle Gouillart <[email protected]>\n# License: BSD 3 clause\n\nimport numpy as np\nfrom scipy import sparse\nfrom scipy import ndimage\nfrom sklearn.linear_model import Lasso\nfrom sklearn.linear_model import Ridge\nimport matplotlib.pyplot as plt\n\n\ndef _weights(x, dx=1, orig=0):\n    x = np.ravel(x)\n    floor_x = np.floor((x - orig) / dx)\n    alpha = (x - orig - floor_x * dx) / dx\n    return np.hstack((floor_x, floor_x + 1)), np.hstack((1 - alpha, alpha))\n\n\ndef _generate_center_coordinates(l_x):\n    X, Y = np.mgrid[:l_x, :l_x].astype(np.float64)\n    center = l_x / 2.\n    X += 0.5 - center\n    Y += 0.5 - center\n    return X, Y\n\n\ndef build_projection_operator(l_x, n_dir):\n    \"\"\" Compute the tomography design matrix.\n\n    Parameters\n    ----------\n\n    l_x : int\n        linear size of image array\n\n    n_dir : int\n        number of angles at which projections are acquired.\n\n    Returns\n    -------\n    p : sparse matrix of shape (n_dir l_x, l_x**2)\n    \"\"\"\n    X, Y = _generate_center_coordinates(l_x)\n    angles = np.linspace(0, np.pi, n_dir, endpoint=False)\n    data_inds, weights, camera_inds = [], [], []\n    data_unravel_indices = np.arange(l_x ** 2)\n    data_unravel_indices = np.hstack((data_unravel_indices,\n                                      data_unravel_indices))\n    for i, angle in enumerate(angles):\n        Xrot = np.cos(angle) * X - np.sin(angle) * Y\n        inds, w = _weights(Xrot, dx=1, orig=X.min())\n        mask = np.logical_and(inds >= 0, inds < l_x)\n        weights += list(w[mask])\n        camera_inds += list(inds[mask] + i * l_x)\n        data_inds += list(data_unravel_indices[mask])\n    proj_operator = sparse.coo_matrix((weights, (camera_inds, data_inds)))\n    return proj_operator\n\n\ndef generate_synthetic_data():\n    \"\"\" Synthetic binary data \"\"\"\n    rs = np.random.RandomState(0)\n    n_pts = 36\n    x, y = np.ogrid[0:l, 0:l]\n    mask_outer = (x - l / 2) ** 2 + (y - l / 2) ** 2 < (l / 2) ** 2\n    mask = np.zeros((l, l))\n    points = l * rs.rand(2, n_pts)\n    mask[(points[0]).astype(np.int), (points[1]).astype(np.int)] = 1\n    mask = ndimage.gaussian_filter(mask, sigma=l / n_pts)\n    res = np.logical_and(mask > mask.mean(), mask_outer)\n    return np.logical_xor(res, ndimage.binary_erosion(res))\n\n\n# Generate synthetic images, and projections\nl = 128\nproj_operator = build_projection_operator(l, l / 7.)\ndata = generate_synthetic_data()\nproj = proj_operator * data.ravel()[:, np.newaxis]\nproj += 0.15 * np.random.randn(*proj.shape)\n\n# Reconstruction with L2 (Ridge) penalization\nrgr_ridge = Ridge(alpha=0.2)\nrgr_ridge.fit(proj_operator, proj.ravel())\nrec_l2 = rgr_ridge.coef_.reshape(l, l)\n\n# Reconstruction with L1 (Lasso) penalization\n# the best value of alpha was determined using cross validation\n# with LassoCV\nrgr_lasso = Lasso(alpha=0.001)\nrgr_lasso.fit(proj_operator, proj.ravel())\nrec_l1 = rgr_lasso.coef_.reshape(l, l)\n\nplt.figure(figsize=(8, 3.3))\nplt.subplot(131)\nplt.imshow(data, cmap=plt.cm.gray, interpolation='nearest')\nplt.axis('off')\nplt.title('original image')\nplt.subplot(132)\nplt.imshow(rec_l2, cmap=plt.cm.gray, interpolation='nearest')\nplt.title('L2 penalization')\nplt.axis('off')\nplt.subplot(133)\nplt.imshow(rec_l1, cmap=plt.cm.gray, interpolation='nearest')\nplt.title('L1 penalization')\nplt.axis('off')\n\nplt.subplots_adjust(hspace=0.01, wspace=0.01, top=1, bottom=0, left=0,\n                    right=1)\n\nplt.show()"
+        "print(__doc__)\n\n# Author: Emmanuelle Gouillart <[email protected]>\n# License: BSD 3 clause\n\nimport numpy as np\nfrom scipy import sparse\nfrom scipy import ndimage\nfrom sklearn.linear_model import Lasso\nfrom sklearn.linear_model import Ridge\nimport matplotlib.pyplot as plt\n\n\ndef _weights(x, dx=1, orig=0):\n    x = np.ravel(x)\n    floor_x = np.floor((x - orig) / dx)\n    alpha = (x - orig - floor_x * dx) / dx\n    return np.hstack((floor_x, floor_x + 1)), np.hstack((1 - alpha, alpha))\n\n\ndef _generate_center_coordinates(l_x):\n    X, Y = np.mgrid[:l_x, :l_x].astype(np.float64)\n    center = l_x / 2.\n    X += 0.5 - center\n    Y += 0.5 - center\n    return X, Y\n\n\ndef build_projection_operator(l_x, n_dir):\n    \"\"\" Compute the tomography design matrix.\n\n    Parameters\n    ----------\n\n    l_x : int\n        linear size of image array\n\n    n_dir : int\n        number of angles at which projections are acquired.\n\n    Returns\n    -------\n    p : sparse matrix of shape (n_dir l_x, l_x**2)\n    \"\"\"\n    X, Y = _generate_center_coordinates(l_x)\n    angles = np.linspace(0, np.pi, n_dir, endpoint=False)\n    data_inds, weights, camera_inds = [], [], []\n    data_unravel_indices = np.arange(l_x ** 2)\n    data_unravel_indices = np.hstack((data_unravel_indices,\n                                      data_unravel_indices))\n    for i, angle in enumerate(angles):\n        Xrot = np.cos(angle) * X - np.sin(angle) * Y\n        inds, w = _weights(Xrot, dx=1, orig=X.min())\n        mask = np.logical_and(inds >= 0, inds < l_x)\n        weights += list(w[mask])\n        camera_inds += list(inds[mask] + i * l_x)\n        data_inds += list(data_unravel_indices[mask])\n    proj_operator = sparse.coo_matrix((weights, (camera_inds, data_inds)))\n    return proj_operator\n\n\ndef generate_synthetic_data():\n    \"\"\" Synthetic binary data \"\"\"\n    rs = np.random.RandomState(0)\n    n_pts = 36\n    x, y = np.ogrid[0:l, 0:l]\n    mask_outer = (x - l / 2.) ** 2 + (y - l / 2.) ** 2 < (l / 2.) ** 2\n    mask = np.zeros((l, l))\n    points = l * rs.rand(2, n_pts)\n    mask[(points[0]).astype(np.int), (points[1]).astype(np.int)] = 1\n    mask = ndimage.gaussian_filter(mask, sigma=l / n_pts)\n    res = np.logical_and(mask > mask.mean(), mask_outer)\n    return np.logical_xor(res, ndimage.binary_erosion(res))\n\n\n# Generate synthetic images, and projections\nl = 128\nproj_operator = build_projection_operator(l, l / 7.)\ndata = generate_synthetic_data()\nproj = proj_operator * data.ravel()[:, np.newaxis]\nproj += 0.15 * np.random.randn(*proj.shape)\n\n# Reconstruction with L2 (Ridge) penalization\nrgr_ridge = Ridge(alpha=0.2)\nrgr_ridge.fit(proj_operator, proj.ravel())\nrec_l2 = rgr_ridge.coef_.reshape(l, l)\n\n# Reconstruction with L1 (Lasso) penalization\n# the best value of alpha was determined using cross validation\n# with LassoCV\nrgr_lasso = Lasso(alpha=0.001)\nrgr_lasso.fit(proj_operator, proj.ravel())\nrec_l1 = rgr_lasso.coef_.reshape(l, l)\n\nplt.figure(figsize=(8, 3.3))\nplt.subplot(131)\nplt.imshow(data, cmap=plt.cm.gray, interpolation='nearest')\nplt.axis('off')\nplt.title('original image')\nplt.subplot(132)\nplt.imshow(rec_l2, cmap=plt.cm.gray, interpolation='nearest')\nplt.title('L2 penalization')\nplt.axis('off')\nplt.subplot(133)\nplt.imshow(rec_l1, cmap=plt.cm.gray, interpolation='nearest')\nplt.title('L1 penalization')\nplt.axis('off')\n\nplt.subplots_adjust(hspace=0.01, wspace=0.01, top=1, bottom=0, left=0,\n                    right=1)\n\nplt.show()"
       ]
     }
   ],

dev/_downloads/plot_sparse_logistic_regression_mnist.py

@@ -101,7 +101,7 @@ def generate_synthetic_data():
     rs = np.random.RandomState(0)
     n_pts = 36
     x, y = np.ogrid[0:l, 0:l]
-    mask_outer = (x - l / 2) ** 2 + (y - l / 2) ** 2 < (l / 2) ** 2
+    mask_outer = (x - l / 2.) ** 2 + (y - l / 2.) ** 2 < (l / 2.) ** 2
     mask = np.zeros((l, l))
     points = l * rs.rand(2, n_pts)
     mask[(points[0]).astype(np.int), (points[1]).astype(np.int)] = 1