Fix layer LRP pyre fixme issues

captum/_utils/gradient.py

@@ -28,6 +28,7 @@
 from captum._utils.sample_gradient import SampleGradientWrapper
 from captum._utils.typing import (
     ModuleOrModuleList,
+    SliceIntType,
     TargetType,
     TensorOrTupleOfTensorsGeneric,
 )
@@ -775,8 +776,11 @@ def compute_layer_gradients_and_eval(
 
 def construct_neuron_grad_fn(
     layer: Module,
-    # pyre-fixme[24]: Generic type `Callable` expects 2 type parameters.
-    neuron_selector: Union[int, Tuple[Union[int, slice], ...], Callable],
+    neuron_selector: Union[
+        int,
+        Tuple[Union[int, SliceIntType], ...],
+        Callable[[Union[Tensor, Tuple[Tensor, ...]]], Tensor],
+    ],
     device_ids: Union[None, List[int]] = None,
     attribute_to_neuron_input: bool = False,
     # pyre-fixme[24]: Generic type `Callable` expects 2 type parameters.

captum/_utils/typing.py

@@ -41,6 +41,13 @@
     TensorLikeList5D,
 ]
 
+try:
+    # Subscripted slice syntax is not supported in previous Python versions,
+    # falling back to slice type.
+    SliceIntType = slice[int, int, int]
+except TypeError:
+    # pyre-fixme[24]: Generic type `slice` expects 3 type parameters.
+    SliceIntType = slice  # type: ignore
 
 # Necessary for Python >=3.7 and <3.9!
 if TYPE_CHECKING:

captum/attr/_core/layer/grad_cam.py

@@ -1,7 +1,7 @@
 #!/usr/bin/env python3
 
 # pyre-strict
-from typing import Any, Callable, Dict, List, Optional, Tuple, Union
+from typing import Any, Callable, cast, Dict, List, Optional, Tuple, Union
 
 import torch
 import torch.nn.functional as F
@@ -54,8 +54,7 @@ class LayerGradCam(LayerAttribution, GradientAttribution):
 
     def __init__(
         self,
-        # pyre-fixme[24]: Generic type `Callable` expects 2 type parameters.
-        forward_func: Callable,
+        forward_func: Callable[..., Tensor],
         layer: Module,
         device_ids: Union[None, List[int]] = None,
     ) -> None:
@@ -201,7 +200,7 @@ def attribute(
         # hidden layer and hidden layer evaluated at each input.
         layer_gradients, layer_evals = compute_layer_gradients_and_eval(
             self.forward_func,
-            self.layer,
+            cast(Module, self.layer),
             inputs,
             target,
             additional_forward_args,
@@ -213,10 +212,7 @@ def attribute(
         summed_grads = tuple(
             (
                 torch.mean(
-                    # pyre-fixme[6]: For 1st argument expected `Tensor` but got
-                    #  `Tuple[Tensor, ...]`.
                     layer_grad,
-                    # pyre-fixme[16]: `tuple` has no attribute `shape`.
                     dim=tuple(x for x in range(2, len(layer_grad.shape))),
                     keepdim=True,
                 )
@@ -228,27 +224,15 @@ def attribute(
 
         if attr_dim_summation:
             scaled_acts = tuple(
-                # pyre-fixme[58]: `*` is not supported for operand types
-                #  `Union[tuple[torch._tensor.Tensor], torch._tensor.Tensor]` and
-                #  `Tuple[Tensor, ...]`.
-                # pyre-fixme[6]: For 1st argument expected `Tensor` but got
-                #  `Tuple[Tensor, ...]`.
                 torch.sum(summed_grad * layer_eval, dim=1, keepdim=True)
                 for summed_grad, layer_eval in zip(summed_grads, layer_evals)
             )
         else:
             scaled_acts = tuple(
-                # pyre-fixme[58]: `*` is not supported for operand types
-                #  `Union[tuple[torch._tensor.Tensor], torch._tensor.Tensor]` and
-                #  `Tuple[Tensor, ...]`.
                 summed_grad * layer_eval
                 for summed_grad, layer_eval in zip(summed_grads, layer_evals)
             )
 
         if relu_attributions:
-            # pyre-fixme[6]: For 1st argument expected `Tensor` but got
-            #  `Union[tuple[Tensor], Tensor]`.
             scaled_acts = tuple(F.relu(scaled_act) for scaled_act in scaled_acts)
-        # pyre-fixme[6]: For 2nd argument expected `Tuple[Tensor, ...]` but got
-        #  `Tuple[Union[tuple[Tensor], Tensor], ...]`.
         return _format_output(len(scaled_acts) > 1, scaled_acts)

captum/attr/_core/layer/internal_influence.py

@@ -1,7 +1,7 @@
 #!/usr/bin/env python3
 
 # pyre-strict
-from typing import Any, Callable, Dict, List, Optional, Tuple, Union
+from typing import Any, Callable, cast, Dict, List, Optional, Tuple, Union
 
 import torch
 from captum._utils.common import (
@@ -41,8 +41,7 @@ class InternalInfluence(LayerAttribution, GradientAttribution):
 
     def __init__(
         self,
-        # pyre-fixme[24]: Generic type `Callable` expects 2 type parameters.
-        forward_func: Callable,
+        forward_func: Callable[..., Tensor],
         layer: Module,
         device_ids: Union[None, List[int]] = None,
     ) -> None:
@@ -293,7 +292,7 @@ def _attribute(
         # Returns gradient of output with respect to hidden layer.
         layer_gradients, _ = compute_layer_gradients_and_eval(
             forward_fn=self.forward_func,
-            layer=self.layer,
+            layer=cast(Module, self.layer),
             inputs=scaled_features_tpl,
             target_ind=expanded_target,
             additional_forward_args=input_additional_args,
@@ -304,9 +303,7 @@ def _attribute(
         # flattening grads so that we can multiply it with step-size
         # calling contiguous to avoid `memory whole` problems
         scaled_grads = tuple(
-            # pyre-fixme[16]: `tuple` has no attribute `contiguous`.
             layer_grad.contiguous().view(n_steps, -1)
-            # pyre-fixme[16]: `tuple` has no attribute `device`.
             * torch.tensor(step_sizes).view(n_steps, 1).to(layer_grad.device)
             for layer_grad in layer_gradients
         )
@@ -317,8 +314,7 @@ def _attribute(
                 scaled_grad,
                 n_steps,
                 inputs[0].shape[0],
-                # pyre-fixme[16]: `tuple` has no attribute `shape`.
-                layer_grad.shape[1:],
+                tuple(layer_grad.shape[1:]),
             )
             for scaled_grad, layer_grad in zip(scaled_grads, layer_gradients)
         )

captum/attr/_core/layer/layer_activation.py

@@ -20,8 +20,7 @@ class LayerActivation(LayerAttribution):
 
     def __init__(
         self,
-        # pyre-fixme[24]: Generic type `Callable` expects 2 type parameters.
-        forward_func: Callable,
+        forward_func: Callable[..., Union[int, float, Tensor]],
         layer: ModuleOrModuleList,
         device_ids: Union[None, List[int]] = None,
     ) -> None:
@@ -132,8 +131,6 @@ def attribute(
             )
         else:
             return [
-                # pyre-fixme[6]: For 2nd argument expected `Tuple[Tensor, ...]` but
-                #  got `Tensor`.
                 _format_output(len(single_layer_eval) > 1, single_layer_eval)
                 for single_layer_eval in layer_eval
             ]

captum/attr/_core/layer/layer_conductance.py

@@ -2,7 +2,7 @@
 
 # pyre-strict
 import typing
-from typing import Any, Callable, Dict, List, Literal, Optional, Tuple, Union
+from typing import Any, Callable, cast, Dict, List, Literal, Optional, Tuple, Union
 
 import torch
 from captum._utils.common import (
@@ -44,8 +44,7 @@ class LayerConductance(LayerAttribution, GradientAttribution):
 
     def __init__(
         self,
-        # pyre-fixme[24]: Generic type `Callable` expects 2 type parameters.
-        forward_func: Callable,
+        forward_func: Callable[..., Tensor],
         layer: Module,
         device_ids: Union[None, List[int]] = None,
     ) -> None:
@@ -73,8 +72,6 @@ def has_convergence_delta(self) -> bool:
         return True
 
     @typing.overload
-    # pyre-fixme[43]: The implementation of `attribute` does not accept all possible
-    #  arguments of overload defined on line `75`.
     def attribute(
         self,
         inputs: Union[Tensor, Tuple[Tensor, ...]],
@@ -91,8 +88,6 @@ def attribute(
     ) -> Tuple[Union[Tensor, Tuple[Tensor, ...]], Tensor]: ...
 
     @typing.overload
-    # pyre-fixme[43]: The implementation of `attribute` does not accept all possible
-    #  arguments of overload defined on line `91`.
     def attribute(
         self,
         inputs: Union[Tensor, Tuple[Tensor, ...]],
@@ -376,7 +371,7 @@ def _attribute(
             layer_evals,
         ) = compute_layer_gradients_and_eval(
             forward_fn=self.forward_func,
-            layer=self.layer,
+            layer=cast(Module, self.layer),
             inputs=scaled_features_tpl,
             additional_forward_args=input_additional_args,
             target_ind=expanded_target,
@@ -389,8 +384,6 @@ def _attribute(
         # This approximates the total input gradient of each step multiplied
         # by the step size.
         grad_diffs = tuple(
-            # pyre-fixme[58]: `-` is not supported for operand types `Tuple[Tensor,
-            #  ...]` and `Tuple[Tensor, ...]`.
             layer_eval[num_examples:] - layer_eval[:-num_examples]
             for layer_eval in layer_evals
         )
@@ -403,8 +396,7 @@ def _attribute(
                 grad_diff * layer_gradient[:-num_examples],
                 n_steps,
                 num_examples,
-                # pyre-fixme[16]: `tuple` has no attribute `shape`.
-                layer_eval.shape[1:],
+                tuple(layer_eval.shape[1:]),
             )
             for layer_gradient, layer_eval, grad_diff in zip(
                 layer_gradients, layer_evals, grad_diffs

captum/attr/_core/layer/layer_deep_lift.py

@@ -321,8 +321,9 @@ def attribute(
                 additional_forward_args,
             )
 
-            # pyre-fixme[24]: Generic type `Sequence` expects 1 type parameter.
-            def chunk_output_fn(out: TensorOrTupleOfTensorsGeneric) -> Sequence:
+            def chunk_output_fn(
+                out: TensorOrTupleOfTensorsGeneric,
+            ) -> Sequence[Union[Tensor, Sequence[Tensor]]]:
                 if isinstance(out, Tensor):
                     return out.chunk(2)
                 return tuple(out_sub.chunk(2) for out_sub in out)
@@ -434,8 +435,6 @@ def __init__(
 
     # Ignoring mypy error for inconsistent signature with DeepLiftShap
     @typing.overload  # type: ignore
-    # pyre-fixme[43]: The implementation of `attribute` does not accept all possible
-    #  arguments of overload defined on line `453`.
     def attribute(
         self,
         inputs: Union[Tensor, Tuple[Tensor, ...]],
@@ -450,9 +449,7 @@ def attribute(
         custom_attribution_func: Union[None, Callable[..., Tuple[Tensor, ...]]] = None,
     ) -> Tuple[Union[Tensor, Tuple[Tensor, ...]], Tensor]: ...
 
-    @typing.overload
-    # pyre-fixme[43]: The implementation of `attribute` does not accept all possible
-    #  arguments of overload defined on line `439`.
+    @typing.overload  # type: ignore
     def attribute(
         self,
         inputs: Union[Tensor, Tuple[Tensor, ...]],
@@ -654,7 +651,7 @@ def attribute(
         ) = DeepLiftShap._expand_inputs_baselines_targets(
             self, baselines, inputs, target, additional_forward_args
         )
-        attributions = LayerDeepLift.attribute.__wrapped__(  # type: ignore
+        attribs_layer_deeplift = LayerDeepLift.attribute.__wrapped__(  # type: ignore
             self,
             exp_inp,
             exp_base,
@@ -667,8 +664,12 @@ def attribute(
             attribute_to_layer_input=attribute_to_layer_input,
             custom_attribution_func=custom_attribution_func,
         )
+        delta: Tensor
+        attributions: Union[Tensor, Tuple[Tensor, ...]]
         if return_convergence_delta:
-            attributions, delta = attributions
+            attributions, delta = attribs_layer_deeplift
+        else:
+            attributions = attribs_layer_deeplift
         if isinstance(attributions, tuple):
             attributions = tuple(
                 DeepLiftShap._compute_mean_across_baselines(
@@ -681,15 +682,17 @@ def attribute(
                 self, inp_bsz, base_bsz, attributions
             )
         if return_convergence_delta:
-            # pyre-fixme[61]: `delta` is undefined, or not always defined.
             return attributions, delta
         else:
-            # pyre-fixme[7]: Expected `Union[Tuple[Union[Tensor,
-            #  typing.Tuple[Tensor, ...]], Tensor], Tensor, typing.Tuple[Tensor, ...]]`
-            #  but got `Union[tuple[Tensor], Tensor]`.
-            return attributions
+            return cast(
+                Union[
+                    Tensor,
+                    Tuple[Tensor, ...],
+                    Tuple[Union[Tensor, Tuple[Tensor, ...]], Tensor],
+                ],
+                attributions,
+            )
 
     @property
-    # pyre-fixme[3]: Return type must be annotated.
     def multiplies_by_inputs(self) -> bool:
         return self._multiply_by_inputs