Put values and offsets on the same device on gpu #45
Conversation
This file contains hidden or 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
Documentation preview |
edknv
commented
Nov 10, 2022
Comment on lines
+26
to
+27
| passenv = | ||
| NR_USER |
Contributor
Author
There was a problem hiding this comment.
Tox blocks all environment variables by default. We need to pass this in order to skip this test that uses horovod. The test was skipped previously because horovod was not installed in the ci runner image, but horovod was recently installed in the ci runner image because of multi-gpu support in Models.
jperez999
approved these changes
Nov 10, 2022
nzarif
reviewed
Nov 10, 2022
nzarif
left a comment
nzarif
left a comment
There was a problem hiding this comment.
I checked with T4Rec multi-gpu training and the fix seems to work.
edknv
added a commit
that referenced
this pull request
Nov 14, 2022
* Put values and offsets on the same device on gpu * pass environ var in tox
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
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Outdated suggestions cannot be applied.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.
Hot fix for NVIDIA-Merlin/Transformers4Rec#526.
Implementation
Testing
python -m pytest tests/unit/loader/test_torch_dataloader.pyManually tested the notebook from T4R as described in NVIDIA-Merlin/Transformers4Rec#526.
Ideally we probably want to set up automated tests for multi-gpu settings but it will probably have to be a follow-up work after the current release.