FIX allow object dtype arrays in clustering metrics (scikit-learn#15535)

amueller · adrinjalali · commit d4e082679d97 · 2019-11-06T11:35:36.000+01:00
* allow object dtype arrays in clustering metrics

* pep8
diff --git a/sklearn/metrics/cluster/_supervised.py b/sklearn/metrics/cluster/_supervised.py
@@ -43,10 +43,10 @@ def check_clusterings(labels_true, labels_pred):
         The predicted labels.
     """
     labels_true = check_array(
-        labels_true, ensure_2d=False, ensure_min_samples=0
+        labels_true, ensure_2d=False, ensure_min_samples=0, dtype=None,
     )
     labels_pred = check_array(
-        labels_pred, ensure_2d=False, ensure_min_samples=0
+        labels_pred, ensure_2d=False, ensure_min_samples=0, dtype=None,
     )
 
     # input checks
diff --git a/sklearn/metrics/cluster/tests/test_common.py b/sklearn/metrics/cluster/tests/test_common.py
@@ -161,7 +161,9 @@ def generate_formats(y):
         y = np.array(y)
         yield y, 'array of ints'
         yield y.tolist(), 'list of ints'
-        yield [str(x) for x in y.tolist()], 'list of strs'
+        yield [str(x) + "-a" for x in y.tolist()], 'list of strs'
+        yield (np.array([str(x) + "-a" for x in y.tolist()], dtype=object),
+               'array of strs')
         yield y - 1, 'including negative ints'
         yield y + 1, 'strictly positive ints'
 

Original file line number	Diff line number	Diff line change
`@@ -43,10 +43,10 @@ def check_clusterings(labels_true, labels_pred):`
`43`	`43`	`The predicted labels.`
`44`	`44`	`"""`
`45`	`45`	`labels_true = check_array(`
`46`		`- labels_true, ensure_2d=False, ensure_min_samples=0`
	`46`	`+ labels_true, ensure_2d=False, ensure_min_samples=0, dtype=None,`
`47`	`47`	`)`
`48`	`48`	`labels_pred = check_array(`
`49`		`- labels_pred, ensure_2d=False, ensure_min_samples=0`
	`49`	`+ labels_pred, ensure_2d=False, ensure_min_samples=0, dtype=None,`
`50`	`50`	`)`
`51`	`51`
`52`	`52`	`# input checks`