Add test for CNN where X td.data.Dataset and sparse_y=True

tests/test_cnn.py

@@ -247,3 +247,30 @@ def test_XY_dataset():
 
     clf.fit(data)
     assert clf.score(data, Y) > 0.6
+
+
+def test_XY_dataset_sparse_y():
+    X = [
+        "One and two",
+        "One only",
+        "Two nothing else",
+        "Two and three"
+    ]
+    Y = np.array([
+        [1, 1, 0, 0],
+        [1, 0, 0, 0],
+        [0, 1, 0, 0],
+        [0, 1, 1, 0]
+    ])
+    Y_sparse = csr_matrix(Y)
+
+    vec = KerasVectorizer()
+    X_vec = vec.fit_transform(X)
+
+    data = tf.data.Dataset.from_tensor_slices((X_vec, Y))
+    data = data.shuffle(100)
+    clf = CNNClassifier(
+        batch_size=2, sparse_y=True, multilabel=True
+    )
+    clf.fit(data)
+    assert clf.score(data, Y_sparse) > 0.3