How to use PyTorch multiprocessing?

Question

As stated in pytorch documentation the best practice to handle multiprocessing is to use torch.multiprocessing instead of multiprocessing.

Be aware that sharing CUDA tensors between processes is supported only in Python 3, either with spawn or forkserver as start method.

Without touching your code, a workaround for the error you got is replacing

from multiprocessing import Process, Pool

with:

from torch.multiprocessing import Pool, Process, set_start_method
try:
     set_start_method('spawn')
except RuntimeError:
    pass