Joblib 0.12.2 (scikit-learn#11741)

* joblib 0.12.2

* Export _joblib's register_parallel_backend

* Use latest version of coverage

Author: ogrisel

doc/modules/compose.rst

@@ -342,7 +342,7 @@ and ``value`` is an estimator object::
     >>> estimators = [('linear_pca', PCA()), ('kernel_pca', KernelPCA())]
     >>> combined = FeatureUnion(estimators)
     >>> combined # doctest: +NORMALIZE_WHITESPACE, +ELLIPSIS
-    FeatureUnion(n_jobs=1,
+    FeatureUnion(n_jobs=None,
                  transformer_list=[('linear_pca', PCA(copy=True,...)),
                                    ('kernel_pca', KernelPCA(alpha=1.0,...))],
                  transformer_weights=None)
@@ -357,7 +357,7 @@ and ignored by setting to ``None``::
 
     >>> combined.set_params(kernel_pca=None)
     ... # doctest: +NORMALIZE_WHITESPACE, +ELLIPSIS
-    FeatureUnion(n_jobs=1,
+    FeatureUnion(n_jobs=None,
                  transformer_list=[('linear_pca', PCA(copy=True,...)),
                                    ('kernel_pca', None)],
                  transformer_weights=None)
@@ -423,7 +423,7 @@ By default, the remaining rating columns are ignored (``remainder='drop'``)::
   ...      remainder='drop')
 
   >>> column_trans.fit(X) # doctest: +NORMALIZE_WHITESPACE +ELLIPSIS
-  ColumnTransformer(n_jobs=1, remainder='drop', sparse_threshold=0.3,
+  ColumnTransformer(n_jobs=None, remainder='drop', sparse_threshold=0.3,
       transformer_weights=None,
       transformers=...)
 
@@ -496,7 +496,7 @@ above example would be::
   ...     ('city', CountVectorizer(analyzer=lambda x: [x])),
   ...     ('title', CountVectorizer()))
   >>> column_trans # doctest: +NORMALIZE_WHITESPACE +ELLIPSIS
-  ColumnTransformer(n_jobs=1, remainder='drop', sparse_threshold=0.3,
+  ColumnTransformer(n_jobs=None, remainder='drop', sparse_threshold=0.3,
            transformer_weights=None,
            transformers=[('countvectorizer-1', ...)
 

doc/modules/kernel_approximation.rst

@@ -64,10 +64,9 @@ a linear algorithm, for example a linear SVM::
     SGDClassifier(alpha=0.0001, average=False, class_weight=None,
            early_stopping=False, epsilon=0.1, eta0=0.0, fit_intercept=True,
            l1_ratio=0.15, learning_rate='optimal', loss='hinge', max_iter=5,
-           n_iter=None, n_iter_no_change=5, n_jobs=1, penalty='l2',
+           n_iter=None, n_iter_no_change=5, n_jobs=None, penalty='l2',
            power_t=0.5, random_state=None, shuffle=True, tol=None,
            validation_fraction=0.1, verbose=0, warm_start=False)
-
     >>> clf.score(X_features, y)
     1.0
 

doc/modules/linear_model.rst

@@ -45,7 +45,9 @@ and will store the coefficients :math:`w` of the linear model in its
     >>> from sklearn import linear_model
     >>> reg = linear_model.LinearRegression()
     >>> reg.fit ([[0, 0], [1, 1], [2, 2]], [0, 1, 2])
-    LinearRegression(copy_X=True, fit_intercept=True, n_jobs=1, normalize=False)
+    ...                                       # doctest: +NORMALIZE_WHITESPACE
+    LinearRegression(copy_X=True, fit_intercept=True, n_jobs=None,
+                     normalize=False)
     >>> reg.coef_
     array([0.5, 0.5])
 

doc/modules/sgd.rst

@@ -64,7 +64,7 @@ for the training samples::
     SGDClassifier(alpha=0.0001, average=False, class_weight=None,
                early_stopping=False, epsilon=0.1, eta0=0.0, fit_intercept=True,
                l1_ratio=0.15, learning_rate='optimal', loss='hinge', max_iter=5,
-               n_iter=None, n_iter_no_change=5, n_jobs=1, penalty='l2',
+               n_iter=None, n_iter_no_change=5, n_jobs=None, penalty='l2',
                power_t=0.5, random_state=None, shuffle=True, tol=None,
                validation_fraction=0.1, verbose=0, warm_start=False)
 

doc/tutorial/statistical_inference/model_selection.rst

@@ -269,9 +269,9 @@ parameter automatically by cross-validation::
     >>> y_diabetes = diabetes.target
     >>> lasso.fit(X_diabetes, y_diabetes)
     LassoCV(alphas=None, copy_X=True, cv=3, eps=0.001, fit_intercept=True,
-        max_iter=1000, n_alphas=100, n_jobs=1, normalize=False, positive=False,
-        precompute='auto', random_state=None, selection='cyclic', tol=0.0001,
-        verbose=False)
+        max_iter=1000, n_alphas=100, n_jobs=None, normalize=False,
+        positive=False, precompute='auto', random_state=None,
+        selection='cyclic', tol=0.0001, verbose=False)
     >>> # The estimator chose automatically its lambda:
     >>> lasso.alpha_ # doctest: +ELLIPSIS
     0.01229...

doc/tutorial/statistical_inference/supervised_learning.rst

@@ -95,7 +95,7 @@ Scikit-learn documentation for more information about this type of classifier.)
     >>> knn = KNeighborsClassifier()
     >>> knn.fit(iris_X_train, iris_y_train) # doctest: +NORMALIZE_WHITESPACE
     KNeighborsClassifier(algorithm='auto', leaf_size=30, metric='minkowski',
-               metric_params=None, n_jobs=1, n_neighbors=5, p=2,
+               metric_params=None, n_jobs=None, n_neighbors=5, p=2,
                weights='uniform')
     >>> knn.predict(iris_X_test)
     array([1, 2, 1, 0, 0, 0, 2, 1, 2, 0])
@@ -176,13 +176,16 @@ Linear models: :math:`y = X\beta + \epsilon`
     >>> from sklearn import linear_model
     >>> regr = linear_model.LinearRegression()
     >>> regr.fit(diabetes_X_train, diabetes_y_train)
-    LinearRegression(copy_X=True, fit_intercept=True, n_jobs=1, normalize=False)
+    ...                                       # doctest: +NORMALIZE_WHITESPACE
+    LinearRegression(copy_X=True, fit_intercept=True, n_jobs=None,
+                     normalize=False)
     >>> print(regr.coef_)
     [   0.30349955 -237.63931533  510.53060544  327.73698041 -814.13170937
       492.81458798  102.84845219  184.60648906  743.51961675   76.09517222]
 
     >>> # The mean square error
-    >>> np.mean((regr.predict(diabetes_X_test)-diabetes_y_test)**2)# doctest: +ELLIPSIS
+    >>> np.mean((regr.predict(diabetes_X_test)-diabetes_y_test)**2)
+    ...                                                   # doctest: +ELLIPSIS
     2004.56760268...
 
     >>> # Explained variance score: 1 is perfect prediction
@@ -257,8 +260,11 @@ diabetes dataset rather than our synthetic data::
     >>> from __future__ import print_function
     >>> print([regr.set_params(alpha=alpha
     ...             ).fit(diabetes_X_train, diabetes_y_train,
-    ...             ).score(diabetes_X_test, diabetes_y_test) for alpha in alphas]) # doctest: +ELLIPSIS
-    [0.5851110683883..., 0.5852073015444..., 0.5854677540698..., 0.5855512036503..., 0.5830717085554..., 0.57058999437...]
+    ...             ).score(diabetes_X_test, diabetes_y_test)
+    ...        for alpha in alphas])
+    ...                            # doctest: +ELLIPSIS, +NORMALIZE_WHITESPACE
+    [0.5851110683883..., 0.5852073015444..., 0.5854677540698...,
+     0.5855512036503..., 0.5830717085554..., 0.57058999437...]
 
 
 .. note::
@@ -372,7 +378,7 @@ function or **logistic** function:
     >>> logistic.fit(iris_X_train, iris_y_train)
     LogisticRegression(C=100000.0, class_weight=None, dual=False,
               fit_intercept=True, intercept_scaling=1, max_iter=100,
-              multi_class='ovr', n_jobs=1, penalty='l2', random_state=None,
+              multi_class='ovr', n_jobs=None, penalty='l2', random_state=None,
               solver='liblinear', tol=0.0001, verbose=0, warm_start=False)
 
 This is known as :class:`LogisticRegression`.

sklearn/cluster/bicluster.py

@@ -93,7 +93,7 @@ class BaseSpectral(six.with_metaclass(ABCMeta, BaseEstimator,
     @abstractmethod
     def __init__(self, n_clusters=3, svd_method="randomized",
                  n_svd_vecs=None, mini_batch=False, init="k-means++",
-                 n_init=10, n_jobs=1, random_state=None):
+                 n_init=10, n_jobs=None, random_state=None):
         self.n_clusters = n_clusters
         self.svd_method = svd_method
         self.n_svd_vecs = n_svd_vecs
@@ -271,7 +271,7 @@ class SpectralCoclustering(BaseSpectral):
     array([0, 0], dtype=int32)
     >>> clustering # doctest: +NORMALIZE_WHITESPACE
     SpectralCoclustering(init='k-means++', mini_batch=False, n_clusters=2,
-               n_init=10, n_jobs=1, n_svd_vecs=None, random_state=0,
+               n_init=10, n_jobs=None, n_svd_vecs=None, random_state=0,
                svd_method='randomized')
 
     References
@@ -284,7 +284,7 @@ class SpectralCoclustering(BaseSpectral):
     """
     def __init__(self, n_clusters=3, svd_method='randomized',
                  n_svd_vecs=None, mini_batch=False, init='k-means++',
-                 n_init=10, n_jobs=1, random_state=None):
+                 n_init=10, n_jobs=None, random_state=None):
         super(SpectralCoclustering, self).__init__(n_clusters,
                                                    svd_method,
                                                    n_svd_vecs,
@@ -419,7 +419,7 @@ class SpectralBiclustering(BaseSpectral):
     >>> clustering # doctest: +NORMALIZE_WHITESPACE
     SpectralBiclustering(init='k-means++', method='bistochastic',
                mini_batch=False, n_best=3, n_clusters=2, n_components=6,
-               n_init=10, n_jobs=1, n_svd_vecs=None, random_state=0,
+               n_init=10, n_jobs=None, n_svd_vecs=None, random_state=0,
                svd_method='randomized')
 
     References
@@ -433,7 +433,7 @@ class SpectralBiclustering(BaseSpectral):
     def __init__(self, n_clusters=3, method='bistochastic',
                  n_components=6, n_best=3, svd_method='randomized',
                  n_svd_vecs=None, mini_batch=False, init='k-means++',
-                 n_init=10, n_jobs=1, random_state=None):
+                 n_init=10, n_jobs=None, random_state=None):
         super(SpectralBiclustering, self).__init__(n_clusters,
                                                    svd_method,
                                                    n_svd_vecs,

sklearn/cluster/dbscan_.py

@@ -20,7 +20,8 @@
 
 
 def dbscan(X, eps=0.5, min_samples=5, metric='minkowski', metric_params=None,
-           algorithm='auto', leaf_size=30, p=2, sample_weight=None, n_jobs=1):
+           algorithm='auto', leaf_size=30, p=2, sample_weight=None,
+           n_jobs=None):
     """Perform DBSCAN clustering from vector array or distance matrix.
 
     Read more in the :ref:`User Guide <dbscan>`.
@@ -255,7 +256,7 @@ class DBSCAN(BaseEstimator, ClusterMixin):
     array([ 0,  0,  0,  1,  1, -1])
     >>> clustering # doctest: +NORMALIZE_WHITESPACE
     DBSCAN(algorithm='auto', eps=3, leaf_size=30, metric='euclidean',
-        metric_params=None, min_samples=2, n_jobs=1, p=None)
+        metric_params=None, min_samples=2, n_jobs=None, p=None)
 
     See also
     --------
@@ -296,7 +297,7 @@ class DBSCAN(BaseEstimator, ClusterMixin):
 
     def __init__(self, eps=0.5, min_samples=5, metric='euclidean',
                  metric_params=None, algorithm='auto', leaf_size=30, p=None,
-                 n_jobs=1):
+                 n_jobs=None):
         self.eps = eps
         self.min_samples = min_samples
         self.metric = metric

sklearn/cluster/k_means_.py

@@ -25,12 +25,13 @@
 from ..utils.sparsefuncs import mean_variance_axis
 from ..utils.validation import _num_samples
 from ..utils import check_array
-from ..utils import check_random_state
 from ..utils import gen_batches
+from ..utils import check_random_state
 from ..utils.validation import check_is_fitted
 from ..utils.validation import FLOAT_DTYPES
 from ..utils import Parallel
 from ..utils import delayed
+from ..utils import effective_n_jobs
 from ..externals.six import string_types
 from ..exceptions import ConvergenceWarning
 from . import _k_means
@@ -184,8 +185,8 @@ def _check_sample_weight(X, sample_weight):
 
 def k_means(X, n_clusters, sample_weight=None, init='k-means++',
             precompute_distances='auto', n_init=10, max_iter=300,
-            verbose=False, tol=1e-4, random_state=None, copy_x=True, n_jobs=1,
-            algorithm="auto", return_n_iter=False):
+            verbose=False, tol=1e-4, random_state=None, copy_x=True,
+            n_jobs=None, algorithm="auto", return_n_iter=False):
     """K-means clustering algorithm.
 
     Read more in the :ref:`User Guide <k_means>`.
@@ -368,7 +369,7 @@ def k_means(X, n_clusters, sample_weight=None, init='k-means++',
     else:
         raise ValueError("Algorithm must be 'auto', 'full' or 'elkan', got"
                          " %s" % str(algorithm))
-    if n_jobs == 1:
+    if effective_n_jobs(n_jobs):
         # For a single thread, less memory is needed if we just store one set
         # of the best results (as opposed to one set per run per thread).
         for it in range(n_init):
@@ -913,7 +914,7 @@ class KMeans(BaseEstimator, ClusterMixin, TransformerMixin):
     def __init__(self, n_clusters=8, init='k-means++', n_init=10,
                  max_iter=300, tol=1e-4, precompute_distances='auto',
                  verbose=0, random_state=None, copy_x=True,
-                 n_jobs=1, algorithm='auto'):
+                 n_jobs=None, algorithm='auto'):
 
         self.n_clusters = n_clusters
         self.init = init

sklearn/cluster/mean_shift_.py

@@ -29,7 +29,7 @@
 
 
 def estimate_bandwidth(X, quantile=0.3, n_samples=None, random_state=0,
-                       n_jobs=1):
+                       n_jobs=None):
     """Estimate the bandwidth to use with the mean-shift algorithm.
 
     That this function takes time at least quadratic in n_samples. For large
@@ -107,7 +107,7 @@ def _mean_shift_single_seed(my_mean, X, nbrs, max_iter):
 
 def mean_shift(X, bandwidth=None, seeds=None, bin_seeding=False,
                min_bin_freq=1, cluster_all=True, max_iter=300,
-               n_jobs=1):
+               n_jobs=None):
     """Perform mean shift clustering of data using a flat kernel.
 
     Read more in the :ref:`User Guide <mean_shift>`.
@@ -364,7 +364,7 @@ class MeanShift(BaseEstimator, ClusterMixin):
     array([0, 1])
     >>> clustering # doctest: +NORMALIZE_WHITESPACE
     MeanShift(bandwidth=2, bin_seeding=False, cluster_all=True, min_bin_freq=1,
-         n_jobs=1, seeds=None)
+         n_jobs=None, seeds=None)
 
     Notes
     -----
@@ -392,7 +392,7 @@ class MeanShift(BaseEstimator, ClusterMixin):
 
     """
     def __init__(self, bandwidth=None, seeds=None, bin_seeding=False,
-                 min_bin_freq=1, cluster_all=True, n_jobs=1):
+                 min_bin_freq=1, cluster_all=True, n_jobs=None):
         self.bandwidth = bandwidth
         self.seeds = seeds
         self.bin_seeding = bin_seeding