-
Notifications
You must be signed in to change notification settings - Fork 3.4k
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
Split train- and val progress into separate bars #16695
Conversation
for more information, see https://pre-commit.ci
… feature/split-progress-bar
⚡ Required checks status: All passing 🟢Groups summary🟢 pytorch_lightning: Tests workflow
These checks are required after the changes to 🟢 pytorch_lightning: Azure GPU
These checks are required after the changes to 🟢 pytorch_lightning: Azure HPU
These checks are required after the changes to 🟢 pytorch_lightning: Azure IPU
These checks are required after the changes to 🟢 pytorch_lightning: Docs
These checks are required after the changes to 🟢 mypy
These checks are required after the changes to 🟢 installThese checks are required after the changes to 🟢 link-check
These checks are required after the changes to Thank you for your contribution! 💜
|
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.
Just curious do we have also a bar for testing and predictions?
@Borda These are already there and remain unchanged. They were never coupled to the "main" bar, so no changes there. |
What does this PR do?
Fixes #12623
A long pending request from the community is an improved progress bar. So far, the training and validation counts were always added together in one progress bar. This has the drawback that more often than not, the time estimate for completion is wrong, because the validation loop typically runs much faster than the regular training loop. Furthermore, users were often confused about the total iteration count, because they didn't realize that the total number was train + val batches. See #6735, #1121, #6406
This PR splits the bars into two separate ones. The validation bar appears and disappears as needed as before. The PR introduces the change for both TQDM and Rich.
Video recording with 10 train batches 5 val batches, validation every 2 train batches:
TQDM:
Screen.Recording.2023-02-09.at.00.11.26.mov
Rich:
Screen.Recording.2023-02-09.at.00.12.49.mov
Before submitting
PR review
Anyone in the community is free to review the PR once the tests have passed.
Before you start reviewing make sure you have read Review guidelines. In short, see the following bullet-list:
Did you have fun?
I made sure I had fun coding 🙃
cc @Borda @justusschock @awaelchli