[MRG] Fix appveyor error link to RuntimeWarning (#165)

* Change the unique checking

* Change type of warning

* Change runtimewarning to userwarning

* Set the warnings filters

* Add some debugging

* Downgrade python

* add some dbg

* avoid capturing the warning in logger

* Make crippy test

* Create the warning 2 subclasses

* solve an error

* check if we remove something that rise somethign before

* Try something else

* try to force the raising

* add the check estimator again

* reset appveyor

Author: glemaitre

imblearn/base.py

@@ -76,16 +76,17 @@ def fit(self, X, y):
 
         self.logger.info('Compute classes statistics ...')
 
-        # Get all the unique elements in the target array
-        uniques = np.unique(y)
-
         # # Raise an error if there is only one class
         # if uniques.size == 1:
         #     raise RuntimeError("Only one class detected, aborting...")
         # Raise a warning for the moment to be compatible with BaseEstimator
-        if uniques.size == 1:
-            warnings.warn('Only one class detected, something will get wrong',
-                          RuntimeWarning)
+        self.logger.debug('The number of classes is %s', np.unique(y).size)
+        self.logger.debug('Shall we raise a warning: %s',
+                          np.unique(y).size == 1)
+        if np.unique(y).size == 1:
+            warnings.simplefilter('always', UserWarning)
+            warnings.warn('Only one class detected, something will get wrong')
+            self.logger.debug('The warning should has been raised.')
 
         # Store the size of X to check at sampling time if we have the
         # same data
@@ -98,7 +99,7 @@ def fit(self, X, y):
         self.min_c_ = min(self.stats_c_, key=self.stats_c_.get)
         self.maj_c_ = max(self.stats_c_, key=self.stats_c_.get)
 
-        self.logger.info('%s classes detected: %s', uniques.size,
+        self.logger.info('%s classes detected: %s', np.unique(y).size,
                          self.stats_c_)
 
         # Check if the ratio provided at initialisation make sense
@@ -254,6 +255,7 @@ def fit(self, X, y):
 
         # Check that the target type is binary
         if not type_of_target(y) == 'binary':
+            warnings.simplefilter('always', UserWarning)
             warnings.warn('The target type should be binary.')
 
         return self
@@ -290,6 +292,7 @@ def fit(self, X, y):
         # Check that the target type is either binary or multiclass
         if not (type_of_target(y) == 'binary' or
                 type_of_target(y) == 'multiclass'):
+            warnings.simplefilter('always', UserWarning)
             warnings.warn('The target type should be binary or multiclass.')
 
         return self

imblearn/combine/tests/test_smote_enn.py

@@ -74,7 +74,7 @@ def test_smote_fit_single_class():
     # Resample the data
     # Create a wrong y
     y_single_class = np.zeros((X.shape[0], ))
-    assert_warns(RuntimeWarning, smote.fit, X, y_single_class)
+    assert_warns(UserWarning, smote.fit, X, y_single_class)
 
 
 def test_smote_fit():

imblearn/combine/tests/test_smote_tomek.py

@@ -74,7 +74,7 @@ def test_smote_fit_single_class():
     # Resample the data
     # Create a wrong y
     y_single_class = np.zeros((X.shape[0], ))
-    assert_warns(RuntimeWarning, smote.fit, X, y_single_class)
+    assert_warns(UserWarning, smote.fit, X, y_single_class)
 
 
 def test_smote_fit():

imblearn/ensemble/tests/test_balance_cascade.py

@@ -90,7 +90,7 @@ def test_bc_fit_single_class():
     # Resample the data
     # Create a wrong y
     y_single_class = np.zeros((X.shape[0], ))
-    assert_warns(RuntimeWarning, bc.fit, X, y_single_class)
+    assert_warns(UserWarning, bc.fit, X, y_single_class)
 
 
 def test_bc_fit_invalid_ratio():

imblearn/ensemble/tests/test_easy_ensemble.py

@@ -81,7 +81,7 @@ def test_ee_fit_single_class():
     # Resample the data
     # Create a wrong y
     y_single_class = np.zeros((X.shape[0], ))
-    assert_warns(RuntimeWarning, ee.fit, X, y_single_class)
+    assert_warns(UserWarning, ee.fit, X, y_single_class)
 
 
 def test_ee_fit_invalid_ratio():

imblearn/over_sampling/tests/test_adasyn.py

@@ -84,7 +84,7 @@ def test_ada_fit_single_class():
     # Resample the data
     # Create a wrong y
     y_single_class = np.zeros((X.shape[0], ))
-    assert_warns(RuntimeWarning, ada.fit, X, y_single_class)
+    assert_warns(UserWarning, ada.fit, X, y_single_class)
 
 
 def test_ada_fit_invalid_ratio():

imblearn/over_sampling/tests/test_random_over_sampler.py

@@ -75,7 +75,7 @@ def test_ros_fit_single_class():
     # Resample the data
     # Create a wrong y
     y_single_class = np.zeros((X.shape[0], ))
-    assert_warns(RuntimeWarning, ros.fit, X, y_single_class)
+    assert_warns(UserWarning, ros.fit, X, y_single_class)
 
 
 def test_ros_fit_invalid_ratio():

imblearn/over_sampling/tests/test_smote.py

@@ -83,7 +83,7 @@ def test_smote_fit_single_class():
     # Resample the data
     # Create a wrong y
     y_single_class = np.zeros((X.shape[0], ))
-    assert_warns(RuntimeWarning, smote.fit, X, y_single_class)
+    assert_warns(UserWarning, smote.fit, X, y_single_class)
 
 
 def test_smote_fit():

imblearn/under_sampling/tests/test_allknn.py

@@ -83,7 +83,7 @@ def test_allknn_fit_single_class():
     # Resample the data
     # Create a wrong y
     y_single_class = np.zeros((X.shape[0], ))
-    assert_warns(RuntimeWarning, allknn.fit, X, y_single_class)
+    assert_warns(UserWarning, allknn.fit, X, y_single_class)
 
 
 def test_allknn_fit():

imblearn/under_sampling/tests/test_cluster_centroids.py

@@ -78,7 +78,7 @@ def test_cc_fit_single_class():
     # Resample the data
     # Create a wrong y
     y_single_class = np.zeros((X.shape[0], ))
-    assert_warns(RuntimeWarning, cc.fit, X, y_single_class)
+    assert_warns(UserWarning, cc.fit, X, y_single_class)
 
 
 def test_cc_fit_invalid_ratio():

imblearn/under_sampling/tests/test_condensed_nearest_neighbour.py

@@ -60,7 +60,7 @@ def test_cnn_fit_single_class():
     # Resample the data
     # Create a wrong y
     y_single_class = np.zeros((X.shape[0], ))
-    assert_warns(RuntimeWarning, cnn.fit, X, y_single_class)
+    assert_warns(UserWarning, cnn.fit, X, y_single_class)
 
 
 def test_cnn_fit():

imblearn/under_sampling/tests/test_edited_nearest_neighbours.py

@@ -61,7 +61,7 @@ def test_enn_fit_single_class():
     # Resample the data
     # Create a wrong y
     y_single_class = np.zeros((X.shape[0], ))
-    assert_warns(RuntimeWarning, enn.fit, X, y_single_class)
+    assert_warns(UserWarning, enn.fit, X, y_single_class)
 
 
 def test_enn_fit():

imblearn/under_sampling/tests/test_instance_hardness_threshold.py

@@ -93,7 +93,7 @@ def test_iht_fit_single_class():
     # Resample the data
     # Create a wrong y
     y_single_class = np.zeros((X.shape[0], ))
-    assert_warns(RuntimeWarning, iht.fit, X, y_single_class)
+    assert_warns(UserWarning, iht.fit, X, y_single_class)
 
 
 def test_iht_fit_invalid_ratio():

imblearn/under_sampling/tests/test_nearmiss_1.py

@@ -97,7 +97,7 @@ def test_nearmiss_fit_single_class():
     # Resample the data
     # Create a wrong y
     y_single_class = np.zeros((X.shape[0], ))
-    assert_warns(RuntimeWarning, nm1.fit, X, y_single_class)
+    assert_warns(UserWarning, nm1.fit, X, y_single_class)
 
 
 def test_nm_fit_invalid_ratio():

imblearn/under_sampling/tests/test_nearmiss_2.py

@@ -97,7 +97,7 @@ def test_nearmiss_fit_single_class():
     # Resample the data
     # Create a wrong y
     y_single_class = np.zeros((X.shape[0], ))
-    assert_warns(RuntimeWarning, nm2.fit, X, y_single_class)
+    assert_warns(UserWarning, nm2.fit, X, y_single_class)
 
 
 def test_nm_fit_invalid_ratio():

imblearn/under_sampling/tests/test_nearmiss_3.py

@@ -97,7 +97,7 @@ def test_nearmiss_fit_single_class():
     # Resample the data
     # Create a wrong y
     y_single_class = np.zeros((X.shape[0], ))
-    assert_warns(RuntimeWarning, nm3.fit, X, y_single_class)
+    assert_warns(UserWarning, nm3.fit, X, y_single_class)
 
 
 def test_nm_fit_invalid_ratio():

imblearn/under_sampling/tests/test_neighbourhood_cleaning_rule.py

@@ -55,7 +55,7 @@ def test_ncr_fit_single_class():
     # Resample the data
     # Create a wrong y
     y_single_class = np.zeros((X.shape[0], ))
-    assert_warns(RuntimeWarning, ncr.fit, X, y_single_class)
+    assert_warns(UserWarning, ncr.fit, X, y_single_class)
 
 
 def test_ncr_fit():

imblearn/under_sampling/tests/test_one_sided_selection.py

@@ -57,7 +57,7 @@ def test_oss_fit_single_class():
     # Resample the data
     # Create a wrong y
     y_single_class = np.zeros((X.shape[0], ))
-    assert_warns(RuntimeWarning, oss.fit, X, y_single_class)
+    assert_warns(UserWarning, oss.fit, X, y_single_class)
 
 
 def test_oss_fit():

imblearn/under_sampling/tests/test_random_under_sampler.py

@@ -74,7 +74,7 @@ def test_rus_fit_single_class():
     # Resample the data
     # Create a wrong y
     y_single_class = np.zeros((X.shape[0], ))
-    assert_warns(RuntimeWarning, rus.fit, X, y_single_class)
+    assert_warns(UserWarning, rus.fit, X, y_single_class)
 
 
 def test_rus_fit_invalid_ratio():

imblearn/under_sampling/tests/test_repeated_edited_nearest_neighbours.py

@@ -94,7 +94,7 @@ def test_renn_fit_single_class():
     # Resample the data
     # Create a wrong y
     y_single_class = np.zeros((X.shape[0], ))
-    assert_warns(RuntimeWarning, renn.fit, X, y_single_class)
+    assert_warns(UserWarning, renn.fit, X, y_single_class)
 
 
 def test_renn_fit():

imblearn/under_sampling/tests/test_tomek_links.py

@@ -59,7 +59,7 @@ def test_tl_fit_single_class():
     # Resample the data
     # Create a wrong y
     y_single_class = np.zeros((X.shape[0], ))
-    assert_warns(RuntimeWarning, tl.fit, X, y_single_class)
+    assert_warns(UserWarning, tl.fit, X, y_single_class)
 
 
 def test_tl_fit():