Lightning-AI / Lightning-AI/pytorch-lightning
on_checkpoint_write_end callback
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 31.4k
- Forks
- 3.8k
- Avg merge
- 6d 7h
- Merged PRs (30d)
- 6
Description
## 🚀 Feature
A callback for when the checkpoint has finished to be writen.
on_save_checkpoint is indeed triggered before the checkpoint is written to disk.
### Motivation
My motivation is that i want to launch some codes once i'm sure that the checkpoint has been created (some validation on checkpoints to be runed as different processes
### Pitch
I want to be able to write callbacks which can be triggered once the callbacks has been written
cc @borda @awaelchli
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 by tracing the existing on_save_checkpoint callback and the checkpoint-writing flow to identify where writing has fully completed. The work is done when a separate on_checkpoint_write_end callback can reliably trigger after the checkpoint is created, allowing follow-up validation processes to run safely.
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
- 38/100