Lightning-AI / Lightning-AI/pytorch-lightning
best-k-metrics in ModelCheckpoint
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 31.4k
- Forks
- 3.8k
- Avg merge
- 6d 7h
- Merged PRs (30d)
- 6
Description
### Description & Motivation
As part of the ModelCheckpoint, it would be nice to have access to all the best metrics, not just the monitor one. You can check an example in #feature/best-k-metrics. Implementation here: [https://github.com/gonzachiar/pytorch-lightning/commit/485c367fa37093749e5f9f59305c56d2ef284198](https://github.com/gonzachiar/pytorch-lightning/commit/485c367fa37093749e5f9f59305c56d2ef284198).
This would be useful for registering experiments, similar to [dvc experiment tracking](https://github.com/gonzachiar/pytorch-lightning/commit/485c367fa37093749e5f9f59305c56d2ef284198)
### Pitch
_No response_
### Alternatives
_No response_
### Additional context
_No response_
cc @lantiga @borda
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start at the ModelCheckpoint entry point and inspect the implementation linked in commit 485c367fa37093749e5f9f59305c56d2ef284198. Determine how best metrics are stored and exposed beyond the monitored metric; done means experiments can access all metrics associated with the best checkpoints.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python, pytorch
- Domain
- machine-learning
- Issue type
- Feature
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 35/100