Starter Task 2: Adding return type annotation to the _attribute_progress_setup method (#1576)

Summary:
Pull Request resolved: #1576

The problem was that the _attribute_progress_setup method in feature_ablation.py didn't have a return type annotation. After making the relevant changes in the task 1 diff to only have the progress function return a tqdm object, I added a tqdm return type annotation to the _attribute_progress_setup function.

Reviewed By: sarahtranfb

Differential Revision: D75972827

fbshipit-source-id: 469f916cf9a4cfd08e5796600da94d3abaa4bfd9

Author: ArjunR2404

captum/attr/_core/feature_ablation.py

@@ -1102,15 +1102,14 @@ def _fut_tuple_to_accumulate_fut_list_cross_tensor(
                 "_fut_tuple_to_accumulate_fut_list_cross_tensor failed"
             ) from e
 
-    # pyre-fixme[3] return type must be annotated
     def _attribute_progress_setup(
         self,
         formatted_inputs: Tuple[Tensor, ...],
         feature_mask: Tuple[Tensor, ...],
         enable_cross_tensor_attribution: bool,
         perturbations_per_eval: int,
         **kwargs: Any,
-    ):
+    ) -> tqdm:
         feature_counts = self._get_feature_counts(
             formatted_inputs, feature_mask, **kwargs
         )