feat(train): start remote GPU training from instance without GPU #600
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
supersergiy
approved these changes
Jan 3, 2024
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## main #600 +/- ##
=========================================
Coverage 100.00% 100.00%
=========================================
Files 128 128
Lines 4433 4433
=========================================
Hits 4433 4433 ☔ View full report in Codecov by Sentry. |
|
Is it about layers without CPU implementations? |
|
The CPU-only scheduler complained when loading a model that was trained with fp16 that that is not supported on CPU. That was just a warning, though. I don't know if there are other situations where the CPU-only scheduler outright refuses to start remote(!) training because of such a falled check. That would be dumb, because I absolutely don't care what the scheduler thinks about my model if remote training is done on GPU. |
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Outdated suggestions cannot be applied.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.
The CPU scheduler node might still fail to load the weights file during initialization, e.g. if it contains unsupported data types. This may be a scenario where we just want to skip the sanity checks on scheduler side entirely...