I have a problem training the model with my own dataset when using Distributed Mode. I wish to train the model on 2 GPUs and the message I get is:
RuntimeError: Expected to have finished reduction in the prior iteration before starting a new one. This error indicates that your module has parameters that were not used in producing loss. You can enable unused parameter detection by passing the keyword argument find_unused_parameters=True to torch.nn.parallel.DistributedDataParallel, and by
making sure all forward function outputs participate in calculating loss.
If you already have done the above, then the distributed data parallel module wasn't able to locate the output tensors in the return value of your module's forward function. Please include the loss function and the structure of the return value of forward of your module when reporting this issue (e.g. list, dict, iterable).
Parameter indices which did not receive grad for rank 0: 28 29
However when I set nproc_per_node to 1, it only uses 1 GPU and the model trains. How can I fix this problem?
I have a problem training the model with my own dataset when using Distributed Mode. I wish to train the model on 2 GPUs and the message I get is:
RuntimeError: Expected to have finished reduction in the prior iteration before starting a new one. This error indicates that your module has parameters that were not used in producing loss. You can enable unused parameter detection by passing the keyword argument
find_unused_parameters=True
totorch.nn.parallel.DistributedDataParallel
, and by making sure allforward
function outputs participate in calculating loss. If you already have done the above, then the distributed data parallel module wasn't able to locate the output tensors in the return value of your module'sforward
function. Please include the loss function and the structure of the return value offorward
of your module when reporting this issue (e.g. list, dict, iterable). Parameter indices which did not receive grad for rank 0: 28 29However when I set nproc_per_node to 1, it only uses 1 GPU and the model trains. How can I fix this problem?