RuntimeError: set_ sizes_ contiguous is not allowed on a Tensor created from .data or .detach().

The error information is as follows ：

RuntimeError                              Traceback (most recent call last)
~\AppData\Local\Temp/ipykernel_19156/279535578.py in <module>
     59 
     60         #2、 Generate a fake picture with noise 
---> 61         noise.data.resize_(data.size()[0], input_dim, 1, 1).normal_(0, 1) # Noise is a input_dim The vector of the dimension 
     62         # Feed the generator to generate images 
     63         fake_pic=netG(noise).detach() # there detach It is to make the generator not participate in gradient update 

RuntimeError: set_sizes_contiguous is not allowed on a Tensor created from .data or .detach().
If your intent is to change the metadata of a Tensor (such as sizes / strides / storage / storage_offset)
without autograd tracking the change, remove the .data / .detach() call and wrap the change in a `with torch.no_grad():` block.
For example, change:
    x.data.set_(y)
to:
    with torch.no_grad():
        x.set_(y)

According to the error prompt, change and remove .data After that, the operation still reports an error , The error information is as follows ：

RuntimeError                              Traceback (most recent call last)
~\AppData\Local\Temp/ipykernel_19156/1757806420.py in <module>
     59 
     60         #2、 Generate a fake picture with noise 
---> 61         noise.resize_(data.size()[0], input_dim, 1, 1).normal_(0, 1) # Noise is a input_dim The vector of the dimension 
     62         # Feed the generator to generate images 
     63         fake_pic=netG(noise).detach() # there detach It is to make the generator not participate in gradient update 

RuntimeError: cannot resize variables that require grad

Then I checked a lot of information on the Internet and failed to solve the problem , Finally, by removing the front noise Parameters of require_grad Option or make it false

#  Input and output of the model 
#  Generate a random noise input to the generator 
noise=torch.tensor((batch_size, input_dim, 1, 1), dtype = torch.float, requires_grad = True)

Modify it as follows ：

#  Input and output of the model 
#  Generate a random noise input to the generator 
noise=torch.tensor((batch_size, input_dim, 1, 1), dtype = torch.float)
# Fixed noise is used to evaluate the results of the generator , It remains the same throughout the training process 
fixed_noise=torch.FloatTensor(batch_size, input_dim, 1, 1).normal_(0,1).requires_grad_(True)

Problem solvable .