LM Workload TF32 #900
Closed
LM Workload TF32 #900
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
Dev -> main
…w pytorch" This reverts commit 6f7d638.
Fix slow ImageNet workloads on PyTorch
add mixed precision training for lm workload
Revert "add mixed precision training for lm workload"
|
MLCommons CLA bot All contributors have signed the MLCommons CLA ✍️ ✅ |
Contributor
|
This branch has the mixed precision code changes for all of the workloads. We have an existing PR open from the lm_workload_base branch that only has the LM workload changes. Note I did not do a final test with the mixed precision for just the LM workload after increasing the number of evals to get a better timing estimates. At this point we may just opt to do LM workload with TF32 for the new release since we don't have anymore bandwidth to test changes. |
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
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Outdated suggestions cannot be applied.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 is the PR for the LM workload with mixed-precision training on 4xA100.