[DLMED] fix mypy

Signed-off-by: Nic Ma <[email protected]>

Author: Nic-Ma

monai/transforms/intensity/array.py

@@ -1688,7 +1688,7 @@ def __call__(self, img: np.ndarray):
                         raise ValueError("fill_value should contain 2 numbers if providing the `min` and `max`.")
                     img[h] = self.R.uniform(self.fill_value[0], self.fill_value[1], size=img[h].shape)
                 else:
-                    img[h] = self.fill_value
+                    img[h] = self.fill_value  # type: ignore
 
         return img
 

monai/transforms/intensity/dictionary.py

@@ -1489,7 +1489,7 @@ def __call__(self, data):
                             raise ValueError("fill_value should contain 2 numbers if providing the `min` and `max`.")
                         d[key][h] = self.R.uniform(self.fill_value[0], self.fill_value[1], size=d[key][h].shape)
                     else:
-                        d[key][h] = self.fill_value
+                        d[key][h] = self.fill_value  # type: ignore
         return d
 
 

tests/test_rand_coarse_dropoutd.py

@@ -51,9 +51,14 @@
     {"img": np.random.randint(0, 2, size=[3, 3, 3, 4])},
 ]
 
+TEST_CASE_4 = [
+    {"keys": "img", "holes": 2, "spatial_size": [2, 2, 2], "fill_value": (0.2, 0.6), "prob": 1.0},
+    {"img": np.random.rand((3, 3, 3, 4))},
+]
+
 
 class TestRandCoarseDropoutd(unittest.TestCase):
-    @parameterized.expand([TEST_CASE_0, TEST_CASE_1, TEST_CASE_2, TEST_CASE_3])
+    @parameterized.expand([TEST_CASE_0, TEST_CASE_1, TEST_CASE_2, TEST_CASE_3, TEST_CASE_4])
     def test_value(self, input_param, input_data):
         dropout = RandCoarseDropoutd(**input_param)
         result = dropout(input_data)["img"]
@@ -74,6 +79,7 @@ def test_value(self, input_param, input_data):
             if isinstance(fill_value, (int, float)):
                 np.testing.assert_allclose(data, fill_value)
             else:
+                print("hole data:", data)
                 min_value = data.min()
                 max_value = data.max()
                 self.assertGreaterEqual(max_value, min_value)