Fix mypy issue introduced in 1.11.0

monai/apps/auto3dseg/hpo_gen.py

@@ -53,7 +53,7 @@ def update_params(self, *args, **kwargs):
         raise NotImplementedError
 
     @abstractmethod
-    def set_score(self):
+    def set_score(self, *args, **kwargs):
         """Report the evaluated results to HPO."""
         raise NotImplementedError
 

monai/apps/pathology/transforms/post/array.py

@@ -28,7 +28,7 @@
     SobelGradients,
 )
 from monai.transforms.transform import Transform
-from monai.transforms.utils_pytorch_numpy_unification import max, maximum, min, sum, unique
+from monai.transforms.utils_pytorch_numpy_unification import max, maximum, min, sum, unique, where
 from monai.utils import TransformBackends, convert_to_numpy, optional_import
 from monai.utils.misc import ensure_tuple_rep
 from monai.utils.type_conversion import convert_to_dst_type, convert_to_tensor
@@ -162,7 +162,8 @@ def __call__(self, prob_map: NdarrayOrTensor) -> NdarrayOrTensor:
         pred = label(pred)[0]
         if self.remove_small_objects is not None:
             pred = self.remove_small_objects(pred)
-        pred[pred > 0] = 1
+        pred_indices = np.where(pred > 0)
+        pred[pred_indices] = 1
 
         return convert_to_dst_type(pred, prob_map, dtype=self.dtype)[0]
 
@@ -338,7 +339,8 @@ def __call__(self, mask: NdarrayOrTensor, instance_border: NdarrayOrTensor) -> N
         instance_border = instance_border >= self.threshold  # uncertain area
 
         marker = mask - convert_to_dst_type(instance_border, mask)[0]  # certain foreground
-        marker[marker < 0] = 0
+        marker_indices = where(marker < 0)
+        marker[marker_indices] = 0  # type: ignore[index]
         marker = self.postprocess_fn(marker)
         marker = convert_to_numpy(marker)
 
@@ -635,7 +637,7 @@ def __call__(  # type: ignore
 
         seg_map_crop = convert_to_dst_type(seg_map_crop == instance_id, type_map_crop, dtype=bool)[0]
 
-        inst_type = type_map_crop[seg_map_crop]
+        inst_type = type_map_crop[seg_map_crop]  # type: ignore[index]
         type_list, type_pixels = unique(inst_type, return_counts=True)
         type_list = list(zip(type_list, type_pixels))
         type_list = sorted(type_list, key=lambda x: x[1], reverse=True)

monai/data/meta_tensor.py

@@ -505,7 +505,7 @@ def peek_pending_rank(self):
         a = self.pending_operations[-1].get(LazyAttr.AFFINE, None) if self.pending_operations else self.affine
         return 1 if a is None else int(max(1, len(a) - 1))
 
-    def new_empty(self, size, dtype=None, device=None, requires_grad=False):
+    def new_empty(self, size, dtype=None, device=None, requires_grad=False):  # type: ignore[override]
         """
         must be defined for deepcopy to work
 
@@ -580,7 +580,7 @@ def ensure_torch_and_prune_meta(
             img.affine = MetaTensor.get_default_affine()
         return img
 
-    def __repr__(self):
+    def __repr__(self):  # type: ignore[override]
         """
         Prints a representation of the tensor.
         Prepends "meta" to ``torch.Tensor.__repr__``.

monai/networks/layers/simplelayers.py

@@ -452,7 +452,7 @@ def get_binary_kernel(window_size: Sequence[int], dtype=torch.float, device=None
 
 def median_filter(
     in_tensor: torch.Tensor,
-    kernel_size: Sequence[int] = (3, 3, 3),
+    kernel_size: Sequence[int] | int = (3, 3, 3),
     spatial_dims: int = 3,
     kernel: torch.Tensor | None = None,
     **kwargs,

monai/networks/nets/quicknat.py

@@ -42,7 +42,7 @@ class SkipConnectionWithIdx(SkipConnection):
     Inherits from SkipConnection but provides the indizes with each forward pass.
     """
 
-    def forward(self, input, indices):
+    def forward(self, input, indices):  # type: ignore[override]
         return super().forward(input), indices
 
 
@@ -57,7 +57,7 @@ class SequentialWithIdx(nn.Sequential):
     def __init__(self, *args):
         super().__init__(*args)
 
-    def forward(self, input, indices):
+    def forward(self, input, indices):  # type: ignore[override]
         for module in self:
             input, indices = module(input, indices)
         return input, indices
@@ -165,7 +165,7 @@ def _get_layer(self, in_channels, out_channels, dilation):
         )
         return nn.Sequential(conv.get_submodule("adn"), conv.get_submodule("conv"))
 
-    def forward(self, input, _):
+    def forward(self, input, _):  # type: ignore[override]
         i = 0
         result = input
         result1 = input  # this will not stay this value, needed here for pylint/mypy
@@ -215,7 +215,7 @@ def __init__(self, in_channels: int, max_pool, se_layer, dropout, kernel_size, n
         super().__init__(in_channels, se_layer, dropout, kernel_size, num_filters)
         self.max_pool = max_pool
 
-    def forward(self, input, indices=None):
+    def forward(self, input, indices=None):  # type: ignore[override]
         input, indices = self.max_pool(input)
 
         out_block, _ = super().forward(input, None)
@@ -243,7 +243,7 @@ def __init__(self, in_channels: int, un_pool, se_layer, dropout, kernel_size, nu
         super().__init__(in_channels, se_layer, dropout, kernel_size, num_filters)
         self.un_pool = un_pool
 
-    def forward(self, input, indices):
+    def forward(self, input, indices):  # type: ignore[override]
         out_block, _ = super().forward(input, None)
         out_block = self.un_pool(out_block, indices)
         return out_block, None
@@ -270,7 +270,7 @@ def __init__(self, in_channels: int, se_layer, dropout, max_pool, un_pool, kerne
         self.max_pool = max_pool
         self.un_pool = un_pool
 
-    def forward(self, input, indices):
+    def forward(self, input, indices):  # type: ignore[override]
         out_block, indices = self.max_pool(input)
         out_block, _ = super().forward(out_block, None)
         out_block = self.un_pool(out_block, indices)

monai/transforms/croppad/array.py

@@ -362,10 +362,10 @@ def __init__(self, lazy: bool = False):
 
     @staticmethod
     def compute_slices(
-        roi_center: Sequence[int] | NdarrayOrTensor | None = None,
-        roi_size: Sequence[int] | NdarrayOrTensor | None = None,
-        roi_start: Sequence[int] | NdarrayOrTensor | None = None,
-        roi_end: Sequence[int] | NdarrayOrTensor | None = None,
+        roi_center: Sequence[int] | int | NdarrayOrTensor | None = None,
+        roi_size: Sequence[int] | int | NdarrayOrTensor | None = None,
+        roi_start: Sequence[int] | int | NdarrayOrTensor | None = None,
+        roi_end: Sequence[int] | int | NdarrayOrTensor | None = None,
         roi_slices: Sequence[slice] | None = None,
     ) -> tuple[slice]:
         """
@@ -459,10 +459,10 @@ class SpatialCrop(Crop):
 
     def __init__(
         self,
-        roi_center: Sequence[int] | NdarrayOrTensor | None = None,
-        roi_size: Sequence[int] | NdarrayOrTensor | None = None,
-        roi_start: Sequence[int] | NdarrayOrTensor | None = None,
-        roi_end: Sequence[int] | NdarrayOrTensor | None = None,
+        roi_center: Sequence[int] | int | NdarrayOrTensor | None = None,
+        roi_size: Sequence[int] | int | NdarrayOrTensor | None = None,
+        roi_start: Sequence[int] | int | NdarrayOrTensor | None = None,
+        roi_end: Sequence[int] | int | NdarrayOrTensor | None = None,
         roi_slices: Sequence[slice] | None = None,
         lazy: bool = False,
     ) -> None:

monai/transforms/croppad/dictionary.py

@@ -438,10 +438,10 @@ class SpatialCropd(Cropd):
     def __init__(
         self,
         keys: KeysCollection,
-        roi_center: Sequence[int] | None = None,
-        roi_size: Sequence[int] | None = None,
-        roi_start: Sequence[int] | None = None,
-        roi_end: Sequence[int] | None = None,
+        roi_center: Sequence[int] | int | None = None,
+        roi_size: Sequence[int] | int | None = None,
+        roi_start: Sequence[int] | int | None = None,
+        roi_end: Sequence[int] | int | None = None,
         roi_slices: Sequence[slice] | None = None,
         allow_missing_keys: bool = False,
         lazy: bool = False,

monai/transforms/spatial/array.py

@@ -3441,7 +3441,7 @@ def filter_count(self, image_np: NdarrayOrTensor, locations: np.ndarray) -> tupl
             idx = self.R.permutation(image_np.shape[0])
             idx = idx[: self.num_patches]
             idx_np = convert_data_type(idx, np.ndarray)[0]
-            image_np = image_np[idx]
+            image_np = image_np[idx]  # type: ignore[index]
             locations = locations[idx_np]
             return image_np, locations
         elif self.sort_fn not in (None, GridPatchSort.MIN, GridPatchSort.MAX):

monai/visualize/class_activation_maps.py

@@ -290,7 +290,7 @@ def __init__(
         )
         self.fc_layers = fc_layers
 
-    def compute_map(self, x, class_idx=None, layer_idx=-1, **kwargs):
+    def compute_map(self, x, class_idx=None, layer_idx=-1, **kwargs):  # type: ignore[override]
         logits, acti, _ = self.nn_module(x, **kwargs)
         acti = acti[layer_idx]
         if class_idx is None:
@@ -302,7 +302,7 @@ def compute_map(self, x, class_idx=None, layer_idx=-1, **kwargs):
         output = torch.stack([output[i, b : b + 1] for i, b in enumerate(class_idx)], dim=0)
         return output.reshape(b, 1, *spatial)  # resume the spatial dims on the selected class
 
-    def __call__(self, x, class_idx=None, layer_idx=-1, **kwargs):
+    def __call__(self, x, class_idx=None, layer_idx=-1, **kwargs):  # type: ignore[override]
         """
         Compute the activation map with upsampling and postprocessing.
 
@@ -361,15 +361,15 @@ class GradCAM(CAMBase):
 
     """
 
-    def compute_map(self, x, class_idx=None, retain_graph=False, layer_idx=-1, **kwargs):
+    def compute_map(self, x, class_idx=None, retain_graph=False, layer_idx=-1, **kwargs):  # type: ignore[override]
         _, acti, grad = self.nn_module(x, class_idx=class_idx, retain_graph=retain_graph, **kwargs)
         acti, grad = acti[layer_idx], grad[layer_idx]
         b, c, *spatial = grad.shape
         weights = grad.view(b, c, -1).mean(2).view(b, c, *[1] * len(spatial))
         acti_map = (weights * acti).sum(1, keepdim=True)
         return F.relu(acti_map)
 
-    def __call__(self, x, class_idx=None, layer_idx=-1, retain_graph=False, **kwargs):
+    def __call__(self, x, class_idx=None, layer_idx=-1, retain_graph=False, **kwargs):  # type: ignore[override]
         """
         Compute the activation map with upsampling and postprocessing.
 
@@ -401,7 +401,7 @@ class GradCAMpp(GradCAM):
 
     """
 
-    def compute_map(self, x, class_idx=None, retain_graph=False, layer_idx=-1, **kwargs):
+    def compute_map(self, x, class_idx=None, retain_graph=False, layer_idx=-1, **kwargs):  # type: ignore[override]
         _, acti, grad = self.nn_module(x, class_idx=class_idx, retain_graph=retain_graph, **kwargs)
         acti, grad = acti[layer_idx], grad[layer_idx]
         b, c, *spatial = grad.shape

tests/test_subpixel_upsample.py

@@ -55,9 +55,9 @@
     (2, 1, 32, 16, 8),
 ]
 
-TEST_CASE_SUBPIXEL.append(TEST_CASE_SUBPIXEL_2D_EXTRA)
-TEST_CASE_SUBPIXEL.append(TEST_CASE_SUBPIXEL_3D_EXTRA)
-TEST_CASE_SUBPIXEL.append(TEST_CASE_SUBPIXEL_CONV_BLOCK_EXTRA)
+TEST_CASE_SUBPIXEL.append(TEST_CASE_SUBPIXEL_2D_EXTRA)  # type: ignore
+TEST_CASE_SUBPIXEL.append(TEST_CASE_SUBPIXEL_3D_EXTRA)  # type: ignore
+TEST_CASE_SUBPIXEL.append(TEST_CASE_SUBPIXEL_CONV_BLOCK_EXTRA)  # type: ignore
 
 # add every test back with the pad/pool sequential component omitted
 for tests in list(TEST_CASE_SUBPIXEL):