TST common test for empty input data

ogrisel · ogrisel · commit 25898ded2537 · 2015-02-16T11:01:38.000+01:00
diff --git a/sklearn/tests/test_common.py b/sklearn/tests/test_common.py
@@ -40,6 +40,7 @@
     check_regressors_pickle,
     check_transformer_pickle,
     check_transformers_unfitted,
+    check_estimators_empty_data_messages,
     check_estimators_nan_inf,
     check_estimators_unfitted,
     check_classifiers_one_label,
@@ -103,6 +104,10 @@ def test_non_meta_estimators():
             yield check_pipeline_consistency, name, Estimator
 
         if name not in CROSS_DECOMPOSITION + ['Imputer']:
+            # Check that all estimator yield informative messages when
+            # trained on empty datasets
+            yield check_estimators_empty_data_messages, name, Estimator
+
             # Test that all estimators check their input for NaN's and infs
             yield check_estimators_nan_inf, name, Estimator
 
diff --git a/sklearn/utils/estimator_checks.py b/sklearn/utils/estimator_checks.py
@@ -12,6 +12,7 @@
 
 from sklearn.externals.six.moves import zip
 from sklearn.utils.testing import assert_raises
+from sklearn.utils.testing import assert_raise_message
 from sklearn.utils.testing import assert_equal
 from sklearn.utils.testing import assert_true
 from sklearn.utils.testing import assert_false
@@ -318,6 +319,23 @@ def check_estimators_dtypes(name, Estimator):
                 pass
 
 
+def check_estimators_empty_data_messages(name, Estimator):
+    with warnings.catch_warnings(record=True):
+        e = Estimator()
+        set_fast_parameters(e)
+        set_random_state(e, 1)
+
+        X_zero_samples = np.empty(0).reshape(0, 3)
+        y = []
+        msg = "0 sample(s) (shape=(0, 3)) while a minimum of 1 is required."
+        assert_raise_message(ValueError, msg, e.fit, X_zero_samples, y)
+
+        X_zero_features = np.empty(0).reshape(3, 0)
+        y = [1, 2, 3]
+        msg = "0 feature(s) (shape=(3, 0)) while a minimum of 1 is required."
+        assert_raise_message(ValueError, msg, e.fit, X_zero_features, y)
+
+
 def check_estimators_nan_inf(name, Estimator):
     rnd = np.random.RandomState(0)
     X_train_finite = rnd.uniform(size=(10, 3))
