FIX SkLearn .score() method generating error with Dask DataFrames (scikit-learn#12462)

ZWMiller · jnothman · commit 5d8dfc9677e4 · 2018-11-06T21:46:13.000+11:00
diff --git a/doc/whats_new/v0.20.rst b/doc/whats_new/v0.20.rst
@@ -1314,6 +1314,9 @@ Miscellaneous
   happens immediately (i.e., without a deprecation cycle).
   :issue:`11741` by `Olivier Grisel`_.
 
+- |Fix| Fixed a bug in validation helpers where passing a Dask DataFrame results
+  in an error. :issue:`12462` by :user:`Zachariah Miller <zwmiller>`
+
 Changes to estimator checks
 ---------------------------
 
diff --git a/sklearn/utils/tests/test_validation.py b/sklearn/utils/tests/test_validation.py
@@ -41,6 +41,7 @@
     check_memory,
     check_non_negative,
     LARGE_SPARSE_SUPPORTED,
+    _num_samples
 )
 import sklearn
 
@@ -786,3 +787,15 @@ def test_check_X_y_informative_error():
     X = np.ones((2, 2))
     y = None
     assert_raise_message(ValueError, "y cannot be None", check_X_y, X, y)
+
+
+def test_retrieve_samples_from_non_standard_shape():
+    class TestNonNumericShape:
+        def __init__(self):
+            self.shape = ("not numeric",)
+
+        def __len__(self):
+            return len([1, 2, 3])
+
+    X = TestNonNumericShape()
+    assert _num_samples(X) == len(X)
diff --git a/sklearn/utils/validation.py b/sklearn/utils/validation.py
@@ -140,7 +140,12 @@ def _num_samples(x):
         if len(x.shape) == 0:
             raise TypeError("Singleton array %r cannot be considered"
                             " a valid collection." % x)
-        return x.shape[0]
+        # Check that shape is returning an integer or default to len
+        # Dask dataframes may not return numeric shape[0] value
+        if isinstance(x.shape[0], numbers.Integral):
+            return x.shape[0]
+        else:
+            return len(x)
     else:
         return len(x)
 
