DataParallel support for torchvision
#1332
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
bfineran
previously approved these changes
Jan 23, 2023
|
@rahul-tuli looks like there are quality issues |
KSGulin
previously approved these changes
Jan 23, 2023
rahul-tuli
force-pushed
the
torchvision-dp-support
branch
from
f3550bb
to
2c07ca7
Compare
Update: all _create_model calling code to accept a third argument Update: device to maybe_dp_device after teacher creation
corey-nm
approved these changes
Feb 9, 2023
dbogunowicz
approved these changes
Feb 9, 2023
bfineran
approved these changes
Feb 10, 2023
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This PR adds support for DataParallel in sparseml's torchvision integration, if no devices are specified, all devices are used by default, else devices can be specified as "cuda:0,1,2,3"
The bug w.r.t multi-gpu environments occurred because the device info wasn't propagated correctly after wrapping the model in DataParallel
The bug has been fixed and the following command was run in a multi-gpu setup to verify all GPU's were picked up; manually verified GPU usage