pytorch-smoothgrad
pytorch-smoothgrad copied to clipboard
pkmr06
SmoothGrad implementation in PyTorch
--------------------------------------------------------------------------- AttributeError Traceback (most recent call last) [](https://localhost:8080/#) in () 94 95 if __name__ == '__main__': ---> 96 main() [](https://localhost:8080/#) in __call__(self, x, index) 45 # print('one_hot after backward:', one_hot)...
[](https://localhost:8080/#) in __call__(self, x, index) 45 # print('one_hot after backward:', one_hot) 46 ---> 47 x.grad.data.zero_() # Reset the gradient to zero 48 49 # print('x=', x) AttributeError: 'NoneType' object has...
While running one_hot.backward(retain_graph=True) , an error appears. RuntimeError: Trying to backward through the graph a second time, but the buffers have already been freed. Specify retain_graph=True when calling backward the...
