Udpate unit tests

ClemDoum · ClemDoum · commit fe23a0cffc45 · 2019-03-11T16:20:20.000+01:00
diff --git a/sklearn/kernel_approximation.py b/sklearn/kernel_approximation.py
@@ -291,7 +291,7 @@ class AdditiveChi2Sampler(BaseEstimator, TransformerMixin):
            power_t=0.5, random_state=0, shuffle=True, tol=0.001,
            validation_fraction=0.1, verbose=0, warm_start=False)
     >>> clf.score(X_transformed, y) # doctest: +ELLIPSIS
-    0.9543...
+    0.9499...
 
     Notes
     -----
diff --git a/sklearn/linear_model/passive_aggressive.py b/sklearn/linear_model/passive_aggressive.py
@@ -148,9 +148,9 @@ class PassiveAggressiveClassifier(BaseSGDClassifier):
                   random_state=0, shuffle=True, tol=0.001,
                   validation_fraction=0.1, verbose=0, warm_start=False)
     >>> print(clf.coef_)
-    [[-0.6543424   1.54603022  1.35361642  0.22199435]]
+    [[0.26642044 0.45070924 0.67251877 0.64185414]]
     >>> print(clf.intercept_)
-    [0.63310933]
+    [1.84127814]
     >>> print(clf.predict([[0, 0, 0, 0]]))
     [1]
 
diff --git a/sklearn/linear_model/perceptron.py b/sklearn/linear_model/perceptron.py
@@ -137,7 +137,7 @@ class Perceptron(BaseSGDClassifier):
           n_jobs=None, penalty=None, random_state=0, shuffle=True, tol=0.001,
           validation_fraction=0.1, verbose=0, warm_start=False)
     >>> clf.score(X, y) # doctest: +ELLIPSIS
-    0.946...
+    0.939...
 
     See also
     --------
diff --git a/sklearn/linear_model/tests/test_logistic.py b/sklearn/linear_model/tests/test_logistic.py
@@ -1671,14 +1671,14 @@ def test_elastic_net_versus_sgd(C, l1_ratio):
     n_samples = 500
     X, y = make_classification(n_samples=n_samples, n_classes=2, n_features=5,
                                n_informative=5, n_redundant=0, n_repeated=0,
-                               random_state=0)
+                               random_state=1)
     X = scale(X)
 
     sgd = SGDClassifier(
-        penalty='elasticnet', random_state=0, fit_intercept=False, tol=-np.inf,
+        penalty='elasticnet', random_state=1, fit_intercept=False, tol=-np.inf,
         max_iter=2000, l1_ratio=l1_ratio, alpha=1. / C / n_samples, loss='log')
     log = LogisticRegression(
-        penalty='elasticnet', random_state=0, fit_intercept=False, tol=1e-5,
+        penalty='elasticnet', random_state=1, fit_intercept=False, tol=1e-5,
         max_iter=1000, l1_ratio=l1_ratio, C=C, solver='saga')
 
     sgd.fit(X, y)
diff --git a/sklearn/linear_model/tests/test_passive_aggressive.py b/sklearn/linear_model/tests/test_passive_aggressive.py
@@ -76,7 +76,7 @@ def test_classifier_accuracy():
             for average in (False, True):
                 clf = PassiveAggressiveClassifier(
                     C=1.0, max_iter=30, fit_intercept=fit_intercept,
-                    random_state=0, average=average, tol=None)
+                    random_state=1, average=average, tol=None)
                 clf.fit(data, y)
                 score = clf.score(data, y)
                 assert_greater(score, 0.79)
diff --git a/sklearn/linear_model/tests/test_sgd.py b/sklearn/linear_model/tests/test_sgd.py
@@ -1032,22 +1032,25 @@ def test_partial_fit_equal_fit_classif(klass, lr):
 
 @pytest.mark.parametrize('klass', [SGDClassifier, SparseSGDClassifier])
 def test_regression_losses(klass):
+    random_state = np.random.RandomState(1)
     clf = klass(alpha=0.01, learning_rate="constant",
-                eta0=0.1, loss="epsilon_insensitive")
+                eta0=0.1, loss="epsilon_insensitive",
+                random_state=random_state)
     clf.fit(X, Y)
     assert_equal(1.0, np.mean(clf.predict(X) == Y))
 
     clf = klass(alpha=0.01, learning_rate="constant",
-                eta0=0.1, loss="squared_epsilon_insensitive")
+                eta0=0.1, loss="squared_epsilon_insensitive",
+                random_state=random_state)
     clf.fit(X, Y)
     assert_equal(1.0, np.mean(clf.predict(X) == Y))
 
-    clf = klass(alpha=0.01, loss="huber")
+    clf = klass(alpha=0.01, loss="huber", random_state=random_state)
     clf.fit(X, Y)
     assert_equal(1.0, np.mean(clf.predict(X) == Y))
 
     clf = klass(alpha=0.01, learning_rate="constant", eta0=0.01,
-                loss="squared_loss")
+                loss="squared_loss", random_state=random_state)
     clf.fit(X, Y)
     assert_equal(1.0, np.mean(clf.predict(X) == Y))
 
diff --git a/sklearn/tests/test_multioutput.py b/sklearn/tests/test_multioutput.py
@@ -332,14 +332,14 @@ def test_multi_output_classification_partial_fit_sample_weights():
     Xw = [[1, 2, 3], [4, 5, 6], [1.5, 2.5, 3.5]]
     yw = [[3, 2], [2, 3], [3, 2]]
     w = np.asarray([2., 1., 1.])
-    sgd_linear_clf = SGDClassifier(random_state=1, max_iter=5)
+    sgd_linear_clf = SGDClassifier(random_state=1, max_iter=20)
     clf_w = MultiOutputClassifier(sgd_linear_clf)
     clf_w.fit(Xw, yw, w)
 
     # unweighted, but with repeated samples
     X = [[1, 2, 3], [1, 2, 3], [4, 5, 6], [1.5, 2.5, 3.5]]
     y = [[3, 2], [3, 2], [2, 3], [3, 2]]
-    sgd_linear_clf = SGDClassifier(random_state=1, max_iter=5)
+    sgd_linear_clf = SGDClassifier(random_state=1, max_iter=20)
     clf = MultiOutputClassifier(sgd_linear_clf)
     clf.fit(X, y)
     X_test = [[1.5, 2.5, 3.5]]
