ENH Improve column_or_1d error message (#15926)

scikit-learn · Dec 20, 2019 · 9626583 · 9626583
1 parent a5542e9
commit 9626583
Show file tree

Hide file tree

Showing 4 changed files with 12 additions and 4 deletions.
diff --git a/doc/whats_new/v0.23.rst b/doc/whats_new/v0.23.rst
@@ -106,3 +106,9 @@ Changelog
 - |Fix| :func:`tree.plot_tree` `rotate` parameter was unused and has been
   deprecated.
   :pr:`15806` by :user:`Chiara Marmo <cmarmo>`.
+
+:mod:`sklearn.utils`
+....................
+
+- |Enhancement| improve error message in :func:`utils.validation.column_or_1d`.
+  :pr:`15926` by :user:`Loïc Estève <lesteve>`.
diff --git a/sklearn/metrics/tests/test_classification.py b/sklearn/metrics/tests/test_classification.py
@@ -486,7 +486,7 @@ def test_multilabel_confusion_matrix_errors():
     # Bad sample_weight
     with pytest.raises(ValueError, match="inconsistent numbers of samples"):
         multilabel_confusion_matrix(y_true, y_pred, sample_weight=[1, 2])
-    with pytest.raises(ValueError, match="bad input shape"):
+    with pytest.raises(ValueError, match="should be a 1d array"):
         multilabel_confusion_matrix(y_true, y_pred,
                                     sample_weight=[[1, 2, 3],
                                                    [2, 3, 4],

diff --git a/sklearn/preprocessing/tests/test_label.py b/sklearn/preprocessing/tests/test_label.py
@@ -222,7 +222,7 @@ def test_label_encoder_negative_ints():
 def test_label_encoder_str_bad_shape(dtype):
     le = LabelEncoder()
     le.fit(np.array(["apple", "orange"], dtype=dtype))
-    msg = "bad input shape"
+    msg = "should be a 1d array"
     with pytest.raises(ValueError, match=msg):
         le.transform("apple")
 
@@ -245,7 +245,7 @@ def test_label_encoder_errors():
         le.inverse_transform([-2, -3, -4])
 
     # Fail on inverse_transform("")
-    msg = "bad input shape ()"
+    msg = r"should be a 1d array.+shape \(\)"
     with pytest.raises(ValueError, match=msg):
         le.inverse_transform("")
 

diff --git a/sklearn/utils/validation.py b/sklearn/utils/validation.py
@@ -743,7 +743,9 @@ def column_or_1d(y, warn=False):
                           DataConversionWarning, stacklevel=2)
         return np.ravel(y)
 
-    raise ValueError("bad input shape {0}".format(shape))
+    raise ValueError(
+        "y should be a 1d array, "
+        "got an array of shape {} instead.".format(shape))
 
 
 def check_random_state(seed):