Hello,
I am getting following error, while running the Spectral Normlization Code at https://github.com/pytorch/pytorch/blob/master/torch/nn/utils/spectral_norm.py
File "/usr/local/torch4/lib/python3.5/site-packages/torch/nn/modules/module.py", line 487, in __call__
hook(self, input)
File "/home/bansa01/pytorch_wideres/tmp_spectral_norm/WideResNet-pytorch/spectral_norm.py", line 43, in __call__
setattr(module, self.name, weight)
File "/usr/local/torch4/lib/python3.5/site-packages/torch/nn/modules/module.py", line 551, in __setattr__
.format(torch.typename(value), name))
TypeError: cannot assign 'torch.cuda.FloatTensor' as parameter 'weight' (torch.nn.Parameter or None expected)
While this function is called:
def __call__(self, module, inputs):
print ("In Call Function")
weight, u = self.compute_weight(module)
setattr(module, self.name, weight)
setattr(module, self.name + '_u', u)
Not sure what I am doing incorrect here. Please suggest!
Thanks!