Conversation
Signed-off-by: Jan Lasek <janek.lasek@gmail.com>
Signed-off-by: Jan Lasek <janek.lasek@gmail.com>
Signed-off-by: Jan Lasek <janek.lasek@gmail.com>
2da19b8 to
21b21ee
Compare
suiyoubi
previously approved these changes
Mar 19, 2024
ericharper
reviewed
Mar 19, 2024
nemo/utils/model_utils.py
Outdated
| model_file = app_state.model_restore_path | ||
| model_cfg = copy.deepcopy(model.cfg) | ||
| if not hasattr(model, "artifacts"): | ||
| OmegaConf.save(model_cfg.tokenizer, os.path.join(output_dir, "tokenizer_config.yaml")) |
Collaborator
There was a problem hiding this comment.
there's no guarantee that model_cfg has an attribute tokenizer
Collaborator
Author
There was a problem hiding this comment.
Thanks - addressed here 8adc16d.
Do you have an example LLM w/o tokenizer? I'll put together a test for this.
Collaborator
|
jenkins |
Signed-off-by: Jan Lasek <janek.lasek@gmail.com>
Signed-off-by: Jan Lasek <janek.lasek@gmail.com>
Signed-off-by: Jan Lasek <janek.lasek@gmail.com>
Signed-off-by: Jan Lasek <janek.lasek@gmail.com>
f0d07e0 to
a23a123
Compare
Collaborator
Author
|
jenkins |
Signed-off-by: Jan Lasek <janek.lasek@gmail.com>
Signed-off-by: Jan Lasek <janek.lasek@gmail.com>
Collaborator
Author
|
jenkins |
rohitrango
pushed a commit
to rohitrango/NeMo
that referenced
this pull request
Jun 25, 2024
* Handle model w/o artifacts Signed-off-by: Jan Lasek <janek.lasek@gmail.com> * Quantization for megatron_amp_O2=True case and test for it Signed-off-by: Jan Lasek <janek.lasek@gmail.com> * Remove broken pileval dataset Signed-off-by: Jan Lasek <janek.lasek@gmail.com> * Consider model w/o tokenizer case Signed-off-by: Jan Lasek <janek.lasek@gmail.com> * Use default converter precision Signed-off-by: Jan Lasek <janek.lasek@gmail.com> * Code formatting Signed-off-by: Jan Lasek <janek.lasek@gmail.com> * Fix typo in Jenkinsfile Signed-off-by: Jan Lasek <janek.lasek@gmail.com> * Disable sequence parallel for evaluation Signed-off-by: Jan Lasek <janek.lasek@gmail.com> --------- Signed-off-by: Jan Lasek <janek.lasek@gmail.com>
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
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.
What does this PR do ?
Fixing quantization issues based on more extensive tests:
megatron_amp_O2: trueCollection: NLP
Changelog
See above.
Jenkins CI
To run Jenkins, a NeMo User with write access must comment
jenkinson the PR.Before your PR is "Ready for review"
Pre checks:
PR Type:
If you haven't finished some of the above items you can still open "Draft" PR.
Who can review?
Anyone in the community is free to review the PR once the checks have passed.
Contributor guidelines contains specific people who can review PRs to various areas.
Additional Information