You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on Feb 12, 2022. It is now read-only.
I try to train lm with multi-GPU(In this case, I use 3 GPU) by set "model = torch.nn.DataParallel(model).cuda()", and change "model.init_hidden" to "model.module.init_hidden", then meet error:
It seems that only one GPU's result has been collected, I can't explain what happened.
The text was updated successfully, but these errors were encountered:
I try to train lm with multi-GPU(In this case, I use 3 GPU) by set "model = torch.nn.DataParallel(model).cuda()", and change "model.init_hidden" to "model.module.init_hidden", then meet error:
It seems that only one GPU's result has been collected, I can't explain what happened.
The text was updated successfully, but these errors were encountered: