Update warnings (#1346)

* Added ignored_warnings file

* Use ignored_warnings file

* Test regressors with 1d, 1d as 2d and 2d targets

* Flake'd

* Fix broken relative imports to ignore_warnings

* Removed print and updated parameter type for tests

* Added warning catches to fit methods in tests

* Added more warning catches

* Flake'd

* Created top-level module to allow relativei imports

* Deleted blank line in __init__

* Remove uneeded ignore warnings from tests

* Fix bad indent

* Fix github merge conflict editor whitespaces and indents

Author: eddiebergman

test/test_pipeline/components/classification/test_base.py

@@ -9,7 +9,7 @@
 import sklearn.metrics
 import numpy as np
 
-from test.test_pipeline.ignored_warnings import ignore_warnings, classifier_warnings
+from ...ignored_warnings import ignore_warnings, classifier_warnings
 
 
 class BaseClassificationComponentTest(unittest.TestCase):

test/test_pipeline/components/feature_preprocessing/test_liblinear.py

@@ -5,7 +5,7 @@
     get_dataset
 import sklearn.metrics
 
-from test.test_pipeline.ignored_warnings import ignore_warnings, feature_preprocessing_warnings
+from ...ignored_warnings import ignore_warnings, feature_preprocessing_warnings
 
 
 class LiblinearComponentTest(PreprocessingTestCase):

test/test_pipeline/components/regression/test_base.py

@@ -35,6 +35,12 @@ def test_default_boston(self):
 
         for _ in range(2):
 
+            with ignore_warnings(regressor_warnings):
+                predictions, targets, n_calls = _test_regressor(
+                    dataset="boston",
+                    Regressor=self.module
+                )
+
             with ignore_warnings(regressor_warnings):
                 predictions, targets, n_calls = _test_regressor(
                     dataset="boston",
@@ -331,7 +337,7 @@ def test_fit_and_predict_with_1d_targets_as_1d(
     regressor: Type[RegressorChoice],
     X: np.ndarray,
     y: np.ndarray
-):
+) -> None:
     """Test that all pipelines work with 1d target types
 
     Parameters
@@ -374,7 +380,7 @@ def test_fit_and_predict_with_1d_targets_as_2d(
     regressor: Type[RegressorChoice],
     X: np.ndarray,
     y: np.ndarray
-):
+) -> None:
     """Test that all pipelines work with 1d target types when they are wrapped as 2d
 
     Parameters
@@ -423,7 +429,7 @@ def test_fit_and_predict_with_2d_targets(
     regressor: Type[RegressorChoice],
     X: np.ndarray,
     y: np.ndarray
-):
+) -> None:
     """Test that all pipelines work with 2d target types
 
     Parameters

test/test_pipeline/ignored_warnings.py

@@ -2,6 +2,7 @@
 from typing import List, Iterator, Tuple
 
 import warnings
+
 from sklearn.exceptions import ConvergenceWarning
 
 
@@ -68,14 +69,34 @@
             r" optimization hasn't converged yet\."
         )
     ),
+    (
+        ConvergenceWarning, (  # From FastICA
+            r"FastICA did not converge\."
+            r" Consider increasing tolerance or the maximum number of iterations\."
+        )
+    ),
     (
         UserWarning, (  # From LDA (Linear Discriminant Analysis)
             r"Variables are collinear"
         )
     ),
+    (
+        UserWarning, (
+            r"Clustering metrics expects discrete values but received continuous values"
+            r" for label, and multiclass values for target"
+        )
+    )
+]
+
+feature_preprocessing_warnings = [
+    (
+        ConvergenceWarning, (  # From liblinear
+            r"Liblinear failed to converge, increase the number of iterations."
+        )
+    )
 ]
 
-ignored_warnings = regressor_warnings + classifier_warnings
+ignored_warnings = regressor_warnings + classifier_warnings + feature_preprocessing_warnings
 
 
 @contextmanager

test/test_pipeline/test_classification.py

@@ -173,7 +173,9 @@ def test_default_configuration(self):
 
         auto = SimpleClassificationPipeline(random_state=1)
 
-        auto = auto.fit(X_train, Y_train)
+        with ignore_warnings(classifier_warnings):
+            auto = auto.fit(X_train, Y_train)
+
         predictions = auto.predict(X_test)
 
         acc = sklearn.metrics.accuracy_score(predictions, Y_test)
@@ -196,7 +198,9 @@ def test_default_configuration_multilabel(self):
         default = cs.get_default_configuration()
         classifier.set_hyperparameters(default)
 
-        classifier = classifier.fit(X_train, Y_train)
+        with ignore_warnings(classifier_warnings):
+            classifier = classifier.fit(X_train, Y_train)
+
         predictions = classifier.predict(X_test)
 
         acc = sklearn.metrics.accuracy_score(predictions, Y_test)
@@ -221,10 +225,12 @@ def test_default_configuration_iterative_fit(self):
             random_state=0
         )
         classifier.fit_transformer(X_train, Y_train)
-        for i in range(1, 11):
-            classifier.iterative_fit(X_train, Y_train)
-            n_estimators = classifier.steps[-1][-1].choice.estimator.n_estimators
-            self.assertEqual(n_estimators, i)
+
+        with ignore_warnings(classifier_warnings):
+            for i in range(1, 11):
+                classifier.iterative_fit(X_train, Y_train)
+                n_estimators = classifier.steps[-1][-1].choice.estimator.n_estimators
+                self.assertEqual(n_estimators, i)
 
     def test_repr(self):
         """Test that the default pipeline can be converted to its representation and
@@ -727,7 +733,9 @@ def test_predict_batched(self):
 
         # Multiclass
         X_train, Y_train, X_test, Y_test = get_dataset(dataset='digits')
-        cls.fit(X_train, Y_train)
+
+        with ignore_warnings(classifier_warnings):
+            cls.fit(X_train, Y_train)
 
         X_test_ = X_test.copy()
         prediction_ = cls.predict_proba(X_test_)
@@ -759,7 +767,8 @@ def test_predict_batched_sparse(self):
 
         # Multiclass
         X_train, Y_train, X_test, Y_test = get_dataset(dataset='digits', make_sparse=True)
-        cls.fit(X_train, Y_train)
+        with ignore_warnings(classifier_warnings):
+            cls.fit(X_train, Y_train)
 
         X_test_ = X_test.copy()
         prediction_ = cls.predict_proba(X_test_)
@@ -788,7 +797,8 @@ def test_predict_proba_batched(self):
         cls = SimpleClassificationPipeline(include={'classifier': ['sgd']})
         X_train, Y_train, X_test, Y_test = get_dataset(dataset='digits')
 
-        cls.fit(X_train, Y_train)
+        with ignore_warnings(classifier_warnings):
+            cls.fit(X_train, Y_train)
 
         X_test_ = X_test.copy()
         prediction_ = cls.predict_proba(X_test_)
@@ -808,7 +818,9 @@ def test_predict_proba_batched(self):
         X_train, Y_train, X_test, Y_test = get_dataset(dataset='digits')
         Y_train = np.array(list([(list([1 if i != y else 0 for i in range(10)]))
                                  for y in Y_train]))
-        cls.fit(X_train, Y_train)
+
+        with ignore_warnings(classifier_warnings):
+            cls.fit(X_train, Y_train)
 
         X_test_ = X_test.copy()
         prediction_ = cls.predict_proba(X_test_)
@@ -842,7 +854,9 @@ def test_predict_proba_batched_sparse(self):
         X_train, Y_train, X_test, Y_test = get_dataset(dataset='digits', make_sparse=True)
         X_test_ = X_test.copy()
 
-        cls.fit(X_train, Y_train)
+        with ignore_warnings(classifier_warnings):
+            cls.fit(X_train, Y_train)
+
         prediction_ = cls.predict_proba(X_test_)
 
         # The object behind the last step in the pipeline
@@ -861,10 +875,13 @@ def test_predict_proba_batched_sparse(self):
             include={'classifier': ['lda']}
         )
         X_train, Y_train, X_test, Y_test = get_dataset(dataset='digits', make_sparse=True)
+
         X_test_ = X_test.copy()
         Y_train = np.array([[1 if i != y else 0 for i in range(10)] for y in Y_train])
 
-        cls.fit(X_train, Y_train)
+        with ignore_warnings(classifier_warnings):
+            cls.fit(X_train, Y_train)
+
         prediction_ = cls.predict_proba(X_test_)
 
         # The object behind the last step in the pipeline
@@ -889,7 +906,9 @@ def test_pipeline_clonability(self):
         X_train, Y_train, X_test, Y_test = get_dataset(dataset='iris')
 
         auto = SimpleClassificationPipeline()
-        auto = auto.fit(X_train, Y_train)
+
+        with ignore_warnings(classifier_warnings):
+            auto = auto.fit(X_train, Y_train)
 
         auto_clone = clone(auto)
         auto_clone_params = auto_clone.get_params()