-
Notifications
You must be signed in to change notification settings - Fork 65
fix: save model dir hf moe checkpoint #503
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
fix: save model dir hf moe checkpoint #503
Conversation
Signed-off-by: Will Johnson <[email protected]>
Signed-off-by: Will Johnson <[email protected]>
Signed-off-by: Will Johnson <[email protected]>
Signed-off-by: Will Johnson <[email protected]>
Signed-off-by: Will Johnson <[email protected]>
Signed-off-by: Will Johnson <[email protected]>
Signed-off-by: Will Johnson <[email protected]>
|
Thanks for making a pull request! 😃 |
| checkpoint(checkpoint_dir, self.save_model_dir) | ||
|
|
||
| callbacks.append( | ||
| ConvertAndSaveHFCheckpointAtEverySave( |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Can we pass save_model_dir here and pick it up in __init__ ?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Fixed, good catch!
kmehant
left a comment
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@willmj apart from one comment I have, rest all looks good to me. Can you also confirm if you are able to pass both the unit tests for this feature on your end? Since I dont think these would run in our GitHub Action runners.
Signed-off-by: Will Johnson <[email protected]>
Signed-off-by: Will Johnson <[email protected]>
Signed-off-by: Will Johnson <[email protected]>
Signed-off-by: Will Johnson <[email protected]>
Description of the change
Building on PR #486, saving
hf_converted_checkpointtosave_model_dirif it exists.Was the PR tested