FIX mislabelling multiclass target when labels is provided in top_k_accuracy_score (#19721)

joclement · glemaitre · commit ab21254c9dec · 2021-04-28T09:40:11.000+02:00
diff --git a/doc/whats_new/v0.24.rst b/doc/whats_new/v0.24.rst
@@ -59,6 +59,14 @@ Changelog
 - |Fix|: Fixed a bug in :class:`linear_model.LogisticRegression`: the
   sample_weight object is not modified anymore. :pr:`19182` by
   :user:`Yosuke KOBAYASHI <m7142yosuke>`.
+  
+:mod:`sklearn.metrics`
+......................
+
+- |Fix| :func:`metrics.top_k_accuracy_score` now supports multiclass
+  problems where only two classes appear in `y_true` and all the classes
+  are specified in `labels`.
+  :pr:`19721` by :user:`Joris Clement <flyingdutchman23>`.
 
 :mod:`sklearn.model_selection`
 ..............................
diff --git a/sklearn/metrics/_ranking.py b/sklearn/metrics/_ranking.py
@@ -1589,7 +1589,7 @@ def top_k_accuracy_score(y_true, y_score, *, k=2, normalize=True,
         non-thresholded decision values (as returned by
         :term:`decision_function` on some classifiers). The binary case expects
         scores with shape (n_samples,) while the multiclass case expects scores
-        with shape (n_samples, n_classes). In the nulticlass case, the order of
+        with shape (n_samples, n_classes). In the multiclass case, the order of
         the class scores must correspond to the order of ``labels``, if
         provided, or else to the numerical or lexicographical order of the
         labels in ``y_true``.
@@ -1646,6 +1646,8 @@ def top_k_accuracy_score(y_true, y_score, *, k=2, normalize=True,
     y_true = check_array(y_true, ensure_2d=False, dtype=None)
     y_true = column_or_1d(y_true)
     y_type = type_of_target(y_true)
+    if y_type == "binary" and labels is not None and len(labels) > 2:
+        y_type = "multiclass"
     y_score = check_array(y_score, ensure_2d=False)
     y_score = column_or_1d(y_score) if y_type == 'binary' else y_score
     check_consistent_length(y_true, y_score, sample_weight)
diff --git a/sklearn/metrics/tests/test_ranking.py b/sklearn/metrics/tests/test_ranking.py
@@ -1650,6 +1650,30 @@ def test_top_k_accuracy_score_binary(y_score, k, true_score):
     assert score == score_acc == pytest.approx(true_score)
 
 
+@pytest.mark.parametrize('y_true, true_score, labels', [
+    (np.array([0, 1, 1, 2]), 0.75, [0, 1, 2, 3]),
+    (np.array([0, 1, 1, 1]), 0.5, [0, 1, 2, 3]),
+    (np.array([1, 1, 1, 1]), 0.5, [0, 1, 2, 3]),
+    (np.array(['a', 'e', 'e', 'a']), 0.75, ['a', 'b', 'd', 'e']),
+])
+@pytest.mark.parametrize("labels_as_ndarray", [True, False])
+def test_top_k_accuracy_score_multiclass_with_labels(
+        y_true, true_score, labels, labels_as_ndarray
+):
+    """Test when labels and y_score are multiclass."""
+    if labels_as_ndarray:
+        labels = np.asarray(labels)
+    y_score = np.array([
+        [0.4, 0.3, 0.2, 0.1],
+        [0.1, 0.3, 0.4, 0.2],
+        [0.4, 0.1, 0.2, 0.3],
+        [0.3, 0.2, 0.4, 0.1],
+    ])
+
+    score = top_k_accuracy_score(y_true, y_score, k=2, labels=labels)
+    assert score == pytest.approx(true_score)
+
+
 def test_top_k_accuracy_score_increasing():
     # Make sure increasing k leads to a higher score
     X, y = datasets.make_classification(n_classes=10, n_samples=1000,
