-
Notifications
You must be signed in to change notification settings - Fork 861
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
Improve save_path in AutoMM #2635
Conversation
Job PR-2635-af78a0e is done. |
Job PR-2635-6c05703 is done. |
Job PR-2635-6a33bad is done. |
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.
LGTM, thanks for the fix!
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.
LGTM
save_path = process_save_path(path=old_save_path, raise_if_exist=False) | ||
else: | ||
save_path = os.path.abspath(os.path.expanduser(old_save_path)) | ||
save_path = process_save_path(path=old_save_path, raise_if_exist=(raise_if_exist and rank == 0)) |
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.
We may need a multi-GPU machine in the CI to test for multi-GPU training next time.
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.
Agree. Maybe select partial tests to avoid slow down CI.
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.
Awesome, thanks for the fix.
Job PR-2635-51206e7 is done. |
Issue #, if available:
.predict()
would create a new save folder and change_save_path
.init()
complicates the logic of save_path.Description of changes:
init()
. Create saving folder only when it's used._model_loaded
flag. For saving the prediction results of object detection, we can directly save to the path_save_path
, whether the model is loaded or fitted.By submitting this pull request, I confirm that you can use, modify, copy, and redistribute this contribution, under the terms of your choice.