scikit-learn · jnothman · Jun 20, 2018 · Jun 20, 2018 · Jun 20, 2018 · Jun 20, 2018
diff --git a/sklearn/base.py b/sklearn/base.py
@@ -301,6 +301,29 @@ def __setstate__(self, state):
         except AttributeError:
             self.__dict__.update(state)
 
+    @classmethod
+    def _generate_test_params(cls):
+        """Generates estimator initialization parameters for check_estimator
+
+        To be overridden where parameters are required or specific settings are
+        helpful for testing.
+
+        Yields
+        ------
+        param_dict : dict
+            Should yield at least one dict that can be passed to cls's
+            constructor as kwargs
+        """
+        try:
+            yield {}
+        except TypeError as exc:
+            raise TypeError('TypeError %s raised when constructing %r.'
+                            'Scikit-learn compatible estimators should '
+                            'usually be constructible without any required '
+                            'parameters.  If this is not possible, '
+                            'classmethod _generate_test_params should be '
+                            'overridden.' % (exc, cls))
+
 
 ###############################################################################
 class ClassifierMixin(object):

diff --git a/sklearn/cluster/affinity_propagation_.py b/sklearn/cluster/affinity_propagation_.py
@@ -328,6 +328,10 @@ def __init__(self, damping=.5, max_iter=200, convergence_iter=15,
         self.preference = preference
         self.affinity = affinity
 
+    @classmethod
+    def _generate_test_params(cls):
+        yield dict(max_iter=5)
+
     @property
     def _pairwise(self):
         return self.affinity == "precomputed"

diff --git a/sklearn/cluster/bicluster.py b/sklearn/cluster/bicluster.py
@@ -103,6 +103,10 @@ def __init__(self, n_clusters=3, svd_method="randomized",
         self.n_jobs = n_jobs
         self.random_state = random_state
 
+    @classmethod
+    def _generate_test_params(cls):
+        yield dict(n_init=2)
+
     def _check_parameters(self):
         legal_svd_methods = ('randomized', 'arpack')
         if self.svd_method not in legal_svd_methods:
@@ -278,6 +282,10 @@ def __init__(self, n_clusters=3, svd_method='randomized',
                                                    n_jobs,
                                                    random_state)
 
+    @classmethod
+    def _generate_test_params(cls):
+        yield dict(n_init=2)
+
     def _fit(self, X):
         normalized_data, row_diag, col_diag = _scale_normalize(X)
         n_sv = 1 + int(np.ceil(np.log2(self.n_clusters)))
@@ -413,6 +421,10 @@ def __init__(self, n_clusters=3, method='bistochastic',
         self.n_components = n_components
         self.n_best = n_best
 
+    @classmethod
+    def _generate_test_params(cls):
+        yield dict(n_init=2)
+
     def _check_parameters(self):
         super(SpectralBiclustering, self)._check_parameters()
         legal_methods = ('bistochastic', 'scale', 'log')

diff --git a/sklearn/cluster/k_means_.py b/sklearn/cluster/k_means_.py
@@ -918,6 +918,10 @@ def __init__(self, n_clusters=8, init='k-means++', n_init=10,
         self.n_jobs = n_jobs
         self.algorithm = algorithm
 
+    @classmethod
+    def _generate_test_params(cls):
+        yield dict(max_iter=5, n_init=2)
+
     def _check_test_data(self, X):
         X = check_array(X, accept_sparse='csr', dtype=FLOAT_DTYPES)
         n_samples, n_features = X.shape
@@ -1424,6 +1428,10 @@ def __init__(self, n_clusters=8, init='k-means++', max_iter=100,
         self.init_size = init_size
         self.reassignment_ratio = reassignment_ratio
 
+    @classmethod
+    def _generate_test_params(cls):
+        yield dict(max_iter=5, n_init=2)
+
     def fit(self, X, y=None, sample_weight=None):
         """Compute the centroids on X by chunking it into mini-batches.
 

diff --git a/sklearn/cluster/spectral.py b/sklearn/cluster/spectral.py
@@ -424,6 +424,10 @@ def __init__(self, n_clusters=8, eigen_solver=None, random_state=None,
         self.kernel_params = kernel_params
         self.n_jobs = n_jobs
 
+    @classmethod
+    def _generate_test_params(cls):
+        yield dict(n_init=2)
+
     def fit(self, X, y=None):
         """Creates an affinity matrix for X using the selected affinity,
         then applies spectral clustering to this affinity matrix.

diff --git a/sklearn/covariance/graph_lasso_.py b/sklearn/covariance/graph_lasso_.py
@@ -342,6 +342,10 @@ def __init__(self, alpha=.01, mode='cd', tol=1e-4, enet_tol=1e-4,
         self.max_iter = max_iter
         self.verbose = verbose
 
+    @classmethod
+    def _generate_test_params(cls):
+        yield dict(max_iter=5)
+
     def fit(self, X, y=None):
         """Fits the GraphicalLasso model to X.
 
@@ -574,6 +578,10 @@ def __init__(self, alphas=4, n_refinements=4, cv=None, tol=1e-4,
         self.cv = cv
         self.n_jobs = n_jobs
 
+    @classmethod
+    def _generate_test_params(cls):
+        yield dict(max_iter=5)
+
     @property
     @deprecated("Attribute grid_scores was deprecated in version 0.19 and "
                 "will be removed in 0.21. Use ``grid_scores_`` instead")
@@ -861,6 +869,10 @@ class GraphLasso(GraphicalLasso):
     graph_lasso, GraphLassoCV
     """
 
+    @classmethod
+    def _generate_test_params(cls):
+        yield dict(max_iter=5)
+
 
 @deprecated("The 'GraphLassoCV' was renamed to 'GraphicalLassoCV' "
             "in version 0.20 and will be removed in 0.22.")
@@ -965,3 +977,7 @@ class GraphLassoCV(GraphicalLassoCV):
     values of alpha then come out as missing values, but the optimum may
     be close to these missing values.
     """
+
+    @classmethod
+    def _generate_test_params(cls):
+        yield dict(max_iter=5)
diff --git a/sklearn/cross_decomposition/cca_.py b/sklearn/cross_decomposition/cca_.py
@@ -105,3 +105,7 @@ def __init__(self, n_components=2, scale=True,
                                   deflation_mode="canonical", mode="B",
                                   norm_y_weights=True, algorithm="nipals",
                                   max_iter=max_iter, tol=tol, copy=copy)
+
+    @classmethod
+    def _generate_test_params(cls):
+        yield dict(max_iter=5)
diff --git a/sklearn/cross_decomposition/pls_.py b/sklearn/cross_decomposition/pls_.py
@@ -595,6 +595,10 @@ def __init__(self, n_components=2, scale=True,
             norm_y_weights=False, max_iter=max_iter, tol=tol,
             copy=copy)
 
+    @classmethod
+    def _generate_test_params(cls):
+        yield dict(max_iter=5)
+
 
 class PLSCanonical(_PLS):
     """ PLSCanonical implements the 2 blocks canonical PLS of the original Wold
@@ -741,6 +745,10 @@ def __init__(self, n_components=2, scale=True, algorithm="nipals",
             norm_y_weights=True, algorithm=algorithm,
             max_iter=max_iter, tol=tol, copy=copy)
 
+    @classmethod
+    def _generate_test_params(cls):
+        yield dict(max_iter=5)
+
 
 class PLSSVD(BaseEstimator, TransformerMixin):
     """Partial Least Square SVD

diff --git a/sklearn/decomposition/dict_learning.py b/sklearn/decomposition/dict_learning.py
@@ -1071,6 +1071,10 @@ def __init__(self, n_components=None, alpha=1, max_iter=1000, tol=1e-8,
         self.verbose = verbose
         self.random_state = random_state
 
+    @classmethod
+    def _generate_test_params(cls):
+        yield dict(max_iter=5)
+
     def fit(self, X, y=None):
         """Fit the model from data in X.
 
@@ -1243,6 +1247,10 @@ def __init__(self, n_components=None, alpha=1, n_iter=1000,
         self.split_sign = split_sign
         self.random_state = random_state
 
+    @classmethod
+    def _generate_test_params(cls):
+        yield dict(n_iter=5)
+
     def fit(self, X, y=None):
         """Fit the model from data in X.
 

diff --git a/sklearn/decomposition/factor_analysis.py b/sklearn/decomposition/factor_analysis.py
@@ -141,6 +141,10 @@ def __init__(self, n_components=None, tol=1e-2, copy=True, max_iter=1000,
         self.iterated_power = iterated_power
         self.random_state = random_state
 
+    @classmethod
+    def _generate_test_params(cls):
+        yield dict(max_iter=5)
+
     def fit(self, X, y=None):
         """Fit the FactorAnalysis model to X using EM
 

diff --git a/sklearn/decomposition/fastica_.py b/sklearn/decomposition/fastica_.py
@@ -468,6 +468,10 @@ def __init__(self, n_components=None, algorithm='parallel', whiten=True,
         self.w_init = w_init
         self.random_state = random_state
 
+    @classmethod
+    def _generate_test_params(cls):
+        yield dict(max_iter=5)
+
     def _fit(self, X, compute_sources=False):
         """Fit the model
 

diff --git a/sklearn/decomposition/nmf.py b/sklearn/decomposition/nmf.py
@@ -1203,6 +1203,10 @@ def __init__(self, n_components=None, init=None, solver='cd',
         self.verbose = verbose
         self.shuffle = shuffle
 
+    @classmethod
+    def _generate_test_params(cls):
+        yield dict(max_iter=100)
+
     def fit_transform(self, X, y=None, W=None, H=None):
         """Learn a NMF model for the data X and returns the transformed data.
 

diff --git a/sklearn/decomposition/online_lda.py b/sklearn/decomposition/online_lda.py
@@ -285,6 +285,10 @@ def __init__(self, n_components=10, doc_topic_prior=None,
         self.random_state = random_state
         self.n_topics = n_topics
 
+    @classmethod
+    def _generate_test_params(cls):
+        yield dict(max_iter=5)
+
     def _check_params(self):
         """Check model parameters."""
         if self.n_topics is not None:

diff --git a/sklearn/decomposition/sparse_pca.py b/sklearn/decomposition/sparse_pca.py
@@ -98,6 +98,10 @@ def __init__(self, n_components=None, alpha=1, ridge_alpha=0.01,
         self.verbose = verbose
         self.random_state = random_state
 
+    @classmethod
+    def _generate_test_params(cls):
+        yield dict(max_iter=5)
+
     def fit(self, X, y=None):
         """Fit the model from data in X.
 
@@ -265,6 +269,10 @@ def __init__(self, n_components=None, alpha=1, ridge_alpha=0.01,
         self.batch_size = batch_size
         self.shuffle = shuffle
 
+    @classmethod
+    def _generate_test_params(cls):
+        yield dict(n_iter=5)
+
     def fit(self, X, y=None):
         """Fit the model from data in X.
 

diff --git a/sklearn/ensemble/bagging.py b/sklearn/ensemble/bagging.py
@@ -574,6 +574,10 @@ def __init__(self,
             random_state=random_state,
             verbose=verbose)
 
+    @classmethod
+    def _generate_test_params(cls):
+        yield dict(n_estimators=5)
+
     def _validate_estimator(self):
         """Check the estimator and set the base_estimator_ attribute."""
         super(BaggingClassifier, self)._validate_estimator(
@@ -941,6 +945,10 @@ def __init__(self,
             random_state=random_state,
             verbose=verbose)
 
+    @classmethod
+    def _generate_test_params(cls):
+        yield dict(n_estimators=5)
+
     def predict(self, X):
         """Predict regression target for X.
 

diff --git a/sklearn/ensemble/forest.py b/sklearn/ensemble/forest.py
@@ -1014,6 +1014,10 @@ def __init__(self,
         self.min_impurity_decrease = min_impurity_decrease
         self.min_impurity_split = min_impurity_split
 
+    @classmethod
+    def _generate_test_params(cls):
+        yield dict(n_estimators=5)
+
 
 class RandomForestRegressor(ForestRegressor):
     """A random forest regressor.
@@ -1252,6 +1256,10 @@ def __init__(self,
         self.min_impurity_decrease = min_impurity_decrease
         self.min_impurity_split = min_impurity_split
 
+    @classmethod
+    def _generate_test_params(cls):
+        yield dict(n_estimators=5)
+
 
 class ExtraTreesClassifier(ForestClassifier):
     """An extra-trees classifier.
@@ -1497,6 +1505,10 @@ def __init__(self,
         self.min_impurity_decrease = min_impurity_decrease
         self.min_impurity_split = min_impurity_split
 
+    @classmethod
+    def _generate_test_params(cls):
+        yield dict(n_estimators=5)
+
 
 class ExtraTreesRegressor(ForestRegressor):
     """An extra-trees regressor.
@@ -1707,6 +1719,10 @@ def __init__(self,
         self.min_impurity_decrease = min_impurity_decrease
         self.min_impurity_split = min_impurity_split
 
+    @classmethod
+    def _generate_test_params(cls):
+        yield dict(n_estimators=5)
+
 
 class RandomTreesEmbedding(BaseForest):
     """An ensemble of totally random trees.
@@ -1872,6 +1888,10 @@ def __init__(self,
         self.min_impurity_split = min_impurity_split
         self.sparse_output = sparse_output
 
+    @classmethod
+    def _generate_test_params(cls):
+        yield dict(n_estimators=5)
+
     def _set_oob_score(self, X, y):
         raise NotImplementedError("OOB score not supported by tree embedding")
 

diff --git a/sklearn/ensemble/gradient_boosting.py b/sklearn/ensemble/gradient_boosting.py
@@ -1552,6 +1552,10 @@ def __init__(self, loss='deviance', learning_rate=0.1, n_estimators=100,
             validation_fraction=validation_fraction,
             n_iter_no_change=n_iter_no_change, tol=tol)
 
+    @classmethod
+    def _generate_test_params(cls):
+        yield dict(n_estimators=5)
+
     def _validate_y(self, y, sample_weight):
         check_classification_targets(y)
         self.classes_, y = np.unique(y, return_inverse=True)
@@ -2000,6 +2004,10 @@ def __init__(self, loss='ls', learning_rate=0.1, n_estimators=100,
             presort=presort, validation_fraction=validation_fraction,
             n_iter_no_change=n_iter_no_change, tol=tol)
 
+    @classmethod
+    def _generate_test_params(cls):
+        yield dict(n_estimators=5)
+
     def predict(self, X):
         """Predict regression target for X.
 

diff --git a/sklearn/ensemble/iforest.py b/sklearn/ensemble/iforest.py
@@ -158,6 +158,10 @@ def __init__(self,
                           DeprecationWarning)
         self.contamination = contamination
 
+    @classmethod
+    def _generate_test_params(cls):
+        yield dict(n_estimators=5)
+
     def _set_oob_score(self, X, y):
         raise NotImplementedError("OOB score not supported by iforest")
 

diff --git a/sklearn/ensemble/weight_boosting.py b/sklearn/ensemble/weight_boosting.py
@@ -384,6 +384,10 @@ def __init__(self,
 
         self.algorithm = algorithm
 
+    @classmethod
+    def _generate_test_params(cls):
+        yield dict(n_estimators=5)
+
     def fit(self, X, y, sample_weight=None):
         """Build a boosted classifier from the training set (X, y).
 
@@ -931,6 +935,10 @@ def __init__(self,
         self.loss = loss
         self.random_state = random_state
 
+    @classmethod
+    def _generate_test_params(cls):
+        yield dict(n_estimators=5)
+
     def fit(self, X, y, sample_weight=None):
         """Build a boosted regressor from the training set (X, y).