There is a problem when I train models

[soutlu]

RuntimeError: An attempt has been made to start a new process before the current process has finished its bootstrapping phase.

    This probably means that you are not using fork to start your
    child processes and you have forgotten to use the proper idiom
    in the main module:

        if __name__ == '__main__':
            freeze_support()
            ...

    The "freeze_support()" line can be omitted if the program
    is not going to be frozen to produce an executable.

[chenkaixin66]

I also happend the same problem,have you fixed this?

[89x98]

I also happend the same problem,have you fixed this?

The generation of this problem is due to the problem of multi-threads under Windows, and the DataLoader class, the specific details https://github.com/pytorch/pytorch/pull/5585

You can solve this issue by modifying the Num_Workers parameter when calling the Torch.Utils.Data.DataLoader() function.

Modify num_works=4 --> num_works=0

or Place the block of code you want to run in the main function

if __name__ == '__main__':
	#your code