How to implement GRADCAM for object detection model

[Monk5088]

Hey author,
Can you explain how can we implement GRADCAM for object detection result model?

[Monk5088]

It gives the following error:

att_shot = gc(data_test.valid_ds[0][0], index=10)

ERROR:

TypeError                                 Traceback (most recent call last)
[<ipython-input-77-0a01d7c8cfc1>](https://localhost:8080/#) in <module>
      1 #All inputs for model and index of chosen class
      2 #If index is None -> will be provided attention to the most chosen class
----> 3 att_shot = gc(data_test.valid_ds[0][0], index=10)

7 frames
[/content/pytorchGradCAM/gradCAM.py](https://localhost:8080/#) in __call__(self, index, *input)
     29         self.output_gradients : List[torch.Tensor] = []
     30 
---> 31         output = self.model(*input)
     32 
     33         if index == None:

[/usr/local/lib/python3.7/dist-packages/torch/nn/modules/module.py](https://localhost:8080/#) in _call_impl(self, *input, **kwargs)
   1108         if not (self._backward_hooks or self._forward_hooks or self._forward_pre_hooks or _global_backward_hooks
   1109                 or _global_forward_hooks or _global_forward_pre_hooks):
-> 1110             return forward_call(*input, **kwargs)
   1111         # Do not call functions when jit is used
   1112         full_backward_hooks, non_full_backward_hooks = [], []

[/usr/local/lib/python3.7/dist-packages/object_detection_fastai/models/RetinaNet.py](https://localhost:8080/#) in forward(self, x)
     70 
     71     def forward(self, x):
---> 72         c5 = self.encoder(x)
     73         p_states = [self.c5top5(c5.clone()), self.c5top6(c5)]
     74         p_states.append(self.p6top7(p_states[-1]))

[/usr/local/lib/python3.7/dist-packages/torch/nn/modules/module.py](https://localhost:8080/#) in _call_impl(self, *input, **kwargs)
   1126             input = bw_hook.setup_input_hook(input)
   1127 
-> 1128         result = forward_call(*input, **kwargs)
   1129         if _global_forward_hooks or self._forward_hooks:
   1130             for hook in (*_global_forward_hooks.values(), *self._forward_hooks.values()):

[/usr/local/lib/python3.7/dist-packages/torch/nn/modules/container.py](https://localhost:8080/#) in forward(self, input)
    139     def forward(self, input):
    140         for module in self:
--> 141             input = module(input)
    142         return input
    143 

[/usr/local/lib/python3.7/dist-packages/torch/nn/modules/module.py](https://localhost:8080/#) in _call_impl(self, *input, **kwargs)
   1126             input = bw_hook.setup_input_hook(input)
   1127 
-> 1128         result = forward_call(*input, **kwargs)
   1129         if _global_forward_hooks or self._forward_hooks:
   1130             for hook in (*_global_forward_hooks.values(), *self._forward_hooks.values()):

[/usr/local/lib/python3.7/dist-packages/torch/nn/modules/conv.py](https://localhost:8080/#) in forward(self, input)
    445 
    446     def forward(self, input: Tensor) -> Tensor:
--> 447         return self._conv_forward(input, self.weight, self.bias)
    448 
    449 class Conv3d(_ConvNd):

[/usr/local/lib/python3.7/dist-packages/torch/nn/modules/conv.py](https://localhost:8080/#) in _conv_forward(self, input, weight, bias)
    442                             _pair(0), self.dilation, self.groups)
    443         return F.conv2d(input, weight, bias, self.stride,
--> 444                         self.padding, self.dilation, self.groups)
    445 
    446     def forward(self, input: Tensor) -> Tensor:

TypeError: conv2d() received an invalid combination of arguments - got (Image, Parameter, NoneType, tuple, tuple, tuple, int), but expected one of:
 * (Tensor input, Tensor weight, Tensor bias, tuple of ints stride, tuple of ints padding, tuple of ints dilation, int groups)
      didn't match because some of the arguments have invalid types: (!Image!, !Parameter!, !NoneType!, !tuple!, !tuple!, !tuple!, int)
 * (Tensor input, Tensor weight, Tensor bias, tuple of ints stride, str padding, tuple of ints dilation, int groups)
      didn't match because some of the arguments have invalid types: (!Image!, !Parameter!, !NoneType!, !tuple!, !tuple!, !tuple!, int)