Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Multi-GPU training with PyTorch Distributed #775

Closed
viswa-nvidia opened this issue Apr 26, 2021 · 1 comment · Fixed by #1099
Closed

Multi-GPU training with PyTorch Distributed #775

viswa-nvidia opened this issue Apr 26, 2021 · 1 comment · Fixed by #1099
Assignees
@jperez999
Labels
P0 PyTorch
Milestone
NVTabular-v21.09

Comments

@viswa-nvidia
Copy link

No description provided.

@viswa-nvidia viswa-nvidia added this to the NVTabular v0.9 milestone Apr 26, 2021
@karlhigley karlhigley mentioned this issue May 14, 2021
Open
@karlhigley
Copy link
Contributor

Someone on our Slack tried to do this by combining PyTorch DistributedDataParallel with the NVT dataloaders, and reported that the NVT dataloaders for PyTorch are not picklable, which may present an obstacle to native support for PyTorch Distributed:
image

@benfred benfred added the P0 label Jul 27, 2021
@jperez999 jperez999 linked a pull request Sep 8, 2021 that will close this issue
Working pytorch distributed + ddp example for multigpu #1099
Merged
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@jperez999 jperez999

Labels
P0 PyTorch
Projects
None yet
Milestone
NVTabular-v21.09
Development

Successfully merging a pull request may close this issue.

Working pytorch distributed + ddp example for multigpu jperez999/NVTabular
5 participants
@benfred @karlhigley @EvenOldridge @jperez999 @viswa-nvidia