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

Provide GPU support #21438

Closed
damccorm opened this issue Jun 4, 2022 · 10 comments
Closed

Provide GPU support #21438

damccorm opened this issue Jun 4, 2022 · 10 comments
Labels
core done & done Issue has been reviewed after it was closed for verification, followups, etc. P2 python run-inference sub-task

Comments

@damccorm
Copy link
Contributor

damccorm commented Jun 4, 2022

Pytorch and Tensorflow have GPU trained models. Need to support this in the RunInference classes, and also make sure they get configured in Dataflow.

Imported from Jira BEAM-13986. Original Jira may contain additional context.
Reported by: yeandy.
Subtask of issue #21435
@damccorm damccorm mentioned this issue Jun 4, 2022
Open
@yeandy
Copy link
Contributor

yeandy commented Jul 15, 2022

Subtask of #22117

@damccorm
Copy link
Contributor Author

@yeandy what pieces are missing to support this?

@yeandy
Copy link
Contributor

yeandy commented Aug 31, 2022 •
edited
Loading

The main PR to address this should be #22795

@damccorm
Copy link
Contributor Author

damccorm commented Sep 6, 2022

@yeandy now that #22795 is done is there any additional work to be done here?

@yeandy
Copy link
Contributor

yeandy commented Sep 6, 2022

No, we can close this now.

@yeandy
Copy link
Contributor

yeandy commented Sep 6, 2022

.close-issue

@github-actions github-actions bot closed this as completed Sep 6, 2022
@tvalentyn
Copy link
Contributor

@yeandy please clarify the scope of this issue. Thanks!

@yeandy
Copy link
Contributor

yeandy commented Sep 6, 2022

Clarification on this issue:

PyTorch RunInference converts data and model to GPU by using the tensor.to(device) command to move a tensor to a device. We also do implicit GPU -> CPU conversion if GPU doesn't exist in the worker.

@tvalentyn
Copy link
Contributor

FYI, you can edit the description or title of the issues.

@damccorm
Copy link
Contributor Author

damccorm commented Sep 6, 2022

FYI, you can edit the description or title of the issues.

Probably not as a non-committer unfortunately

@damccorm damccorm added the done & done Issue has been reviewed after it was closed for verification, followups, etc. label Sep 8, 2022
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
core done & done Issue has been reviewed after it was closed for verification, followups, etc. P2 python run-inference sub-task
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
3 participants
@tvalentyn @yeandy @damccorm