Pushing the docs to dev/ for branch: master, commit be6bf6902c199b27f4f001ce3597e521c3a0d0aa

Author: sklearn-ci

dev/_downloads/1d4bce7e02c88f07c85f455ef1193191/plot_caching_nearest_neighbors.ipynb

@@ -26,7 +26,7 @@
       },
       "outputs": [],
       "source": [
-        "# Author: Tom Dupre la Tour\n#\n# License: BSD 3 clause\nimport matplotlib.pyplot as plt\n\nfrom sklearn.neighbors import KNeighborsTransformer, KNeighborsClassifier\nfrom sklearn.model_selection import GridSearchCV\nfrom sklearn.datasets import load_digits\nfrom sklearn.pipeline import Pipeline\n\nprint(__doc__)\n\nX, y = load_digits(return_X_y=True)\nn_neighbors_list = [1, 2, 3, 4, 5, 6, 7, 8, 9]\n\n# The transformer computes the nearest neighbors graph using the maximum number\n# of neighbors necessary in the grid search. The classifier model filters the\n# nearest neighbors graph as required by its own n_neighbors parameter.\ngraph_model = KNeighborsTransformer(n_neighbors=max(n_neighbors_list),\n                                    mode='distance')\nclassifier_model = KNeighborsClassifier(metric='precomputed')\n\n# Note that we give `memory` a directory to cache the graph computation.\nfull_model = Pipeline(\n    steps=[('graph', graph_model), ('classifier', classifier_model)],\n    memory='./cache')\n\nparam_grid = {'classifier__n_neighbors': n_neighbors_list}\ngrid_model = GridSearchCV(full_model, param_grid)\ngrid_model.fit(X, y)\n\n# Plot the results of the grid search.\nfig, axes = plt.subplots(1, 2, figsize=(8, 4))\naxes[0].errorbar(x=n_neighbors_list,\n                 y=grid_model.cv_results_['mean_test_score'],\n                 yerr=grid_model.cv_results_['std_test_score'])\naxes[0].set(xlabel='n_neighbors', title='Classification accuracy')\naxes[1].errorbar(x=n_neighbors_list, y=grid_model.cv_results_['mean_fit_time'],\n                 yerr=grid_model.cv_results_['std_fit_time'], color='r')\naxes[1].set(xlabel='n_neighbors', title='Fit time (with caching)')\nfig.tight_layout()\nplt.show()"
+        "# Author: Tom Dupre la Tour\n#\n# License: BSD 3 clause\nfrom tempfile import TemporaryDirectory\nimport matplotlib.pyplot as plt\n\nfrom sklearn.neighbors import KNeighborsTransformer, KNeighborsClassifier\nfrom sklearn.model_selection import GridSearchCV\nfrom sklearn.datasets import load_digits\nfrom sklearn.pipeline import Pipeline\n\nprint(__doc__)\n\nX, y = load_digits(return_X_y=True)\nn_neighbors_list = [1, 2, 3, 4, 5, 6, 7, 8, 9]\n\n# The transformer computes the nearest neighbors graph using the maximum number\n# of neighbors necessary in the grid search. The classifier model filters the\n# nearest neighbors graph as required by its own n_neighbors parameter.\ngraph_model = KNeighborsTransformer(n_neighbors=max(n_neighbors_list),\n                                    mode='distance')\nclassifier_model = KNeighborsClassifier(metric='precomputed')\n\n# Note that we give `memory` a directory to cache the graph computation\n# that will be used several times when tuning the hyperparameters of the\n# classifier.\nwith TemporaryDirectory(prefix=\"sklearn_graph_cache_\") as tmpdir:\n    full_model = Pipeline(\n        steps=[('graph', graph_model), ('classifier', classifier_model)],\n        memory=tmpdir)\n\n    param_grid = {'classifier__n_neighbors': n_neighbors_list}\n    grid_model = GridSearchCV(full_model, param_grid)\n    grid_model.fit(X, y)\n\n# Plot the results of the grid search.\nfig, axes = plt.subplots(1, 2, figsize=(8, 4))\naxes[0].errorbar(x=n_neighbors_list,\n                 y=grid_model.cv_results_['mean_test_score'],\n                 yerr=grid_model.cv_results_['std_test_score'])\naxes[0].set(xlabel='n_neighbors', title='Classification accuracy')\naxes[1].errorbar(x=n_neighbors_list, y=grid_model.cv_results_['mean_fit_time'],\n                 yerr=grid_model.cv_results_['std_fit_time'], color='r')\naxes[1].set(xlabel='n_neighbors', title='Fit time (with caching)')\nfig.tight_layout()\nplt.show()"
       ]
     }
   ],

dev/_downloads/3409d9766d352cc9f9b169d4a799a87a/auto_examples_python.zip

@@ -19,6 +19,7 @@
 # Author: Tom Dupre la Tour
 #
 # License: BSD 3 clause
+from tempfile import TemporaryDirectory
 import matplotlib.pyplot as plt
 
 from sklearn.neighbors import KNeighborsTransformer, KNeighborsClassifier
@@ -38,14 +39,17 @@
                                     mode='distance')
 classifier_model = KNeighborsClassifier(metric='precomputed')
 
-# Note that we give `memory` a directory to cache the graph computation.
-full_model = Pipeline(
-    steps=[('graph', graph_model), ('classifier', classifier_model)],
-    memory='./cache')
+# Note that we give `memory` a directory to cache the graph computation
+# that will be used several times when tuning the hyperparameters of the
+# classifier.
+with TemporaryDirectory(prefix="sklearn_graph_cache_") as tmpdir:
+    full_model = Pipeline(
+        steps=[('graph', graph_model), ('classifier', classifier_model)],
+        memory=tmpdir)
 
-param_grid = {'classifier__n_neighbors': n_neighbors_list}
-grid_model = GridSearchCV(full_model, param_grid)
-grid_model.fit(X, y)
+    param_grid = {'classifier__n_neighbors': n_neighbors_list}
+    grid_model = GridSearchCV(full_model, param_grid)
+    grid_model.fit(X, y)
 
 # Plot the results of the grid search.
 fig, axes = plt.subplots(1, 2, figsize=(8, 4))

dev/_downloads/4d6dfc2952a4797fcae4a1abee99951f/plot_caching_nearest_neighbors.py

@@ -133,16 +133,19 @@
 # implementations, such as approximate nearest neighbors methods.
 # See more details in the :ref:`User Guide <neighbors_transformer>`.
 
+from tempfile import TemporaryDirectory
 from sklearn.neighbors import KNeighborsTransformer
 from sklearn.manifold import Isomap
 from sklearn.pipeline import make_pipeline
 
-estimator = make_pipeline(
-    KNeighborsTransformer(n_neighbors=10, mode='distance'),
-    Isomap(n_neighbors=10, metric='precomputed'),
-    memory='.')
-estimator.fit(X)
-
-# We can decrease the number of neighbors and the graph will not be recomputed.
-estimator.set_params(isomap__n_neighbors=5)
-estimator.fit(X)
+with TemporaryDirectory(prefix="sklearn_cache_") as tmpdir:
+    estimator = make_pipeline(
+        KNeighborsTransformer(n_neighbors=10, mode='distance'),
+        Isomap(n_neighbors=10, metric='precomputed'),
+        memory=tmpdir)
+    estimator.fit(X)
+
+    # We can decrease the number of neighbors and the graph will not be
+    # recomputed.
+    estimator.set_params(isomap__n_neighbors=5)
+    estimator.fit(X)

dev/_downloads/7ee55c12f8d3eb1dd8d2005d9dd7b6f1/plot_release_highlights_0_22_0.py

@@ -123,7 +123,7 @@
       },
       "outputs": [],
       "source": [
-        "from sklearn.neighbors import KNeighborsTransformer\nfrom sklearn.manifold import Isomap\nfrom sklearn.pipeline import make_pipeline\n\nestimator = make_pipeline(\n    KNeighborsTransformer(n_neighbors=10, mode='distance'),\n    Isomap(n_neighbors=10, metric='precomputed'),\n    memory='.')\nestimator.fit(X)\n\n# We can decrease the number of neighbors and the graph will not be recomputed.\nestimator.set_params(isomap__n_neighbors=5)\nestimator.fit(X)"
+        "from tempfile import TemporaryDirectory\nfrom sklearn.neighbors import KNeighborsTransformer\nfrom sklearn.manifold import Isomap\nfrom sklearn.pipeline import make_pipeline\n\nwith TemporaryDirectory(prefix=\"sklearn_cache_\") as tmpdir:\n    estimator = make_pipeline(\n        KNeighborsTransformer(n_neighbors=10, mode='distance'),\n        Isomap(n_neighbors=10, metric='precomputed'),\n        memory=tmpdir)\n    estimator.fit(X)\n\n    # We can decrease the number of neighbors and the graph will not be\n    # recomputed.\n    estimator.set_params(isomap__n_neighbors=5)\n    estimator.fit(X)"
       ]
     }
   ],