Lightning-AI / Lightning-AI/pytorch-lightning

Directly save checkpoint to specified disk location

Open
#21,253 4 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

checkpointing question
Dominant language
Python
Stars
31.4k
Forks
3.8k
Avg merge
6d 7h
Merged PRs (30d)
6

Description

### Description & Motivation

Can lightning save the model checkpoints directly to the target location on disk, instead of first writing it to some tmp dir like `/tmp/blah` then trying to move it to the target location?

### Pitch

_No response_

### Alternatives

_No response_

### Additional context

_No response_

cc @lantiga @borda

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start by locating the checkpoint-saving path responsible for writing to a temporary directory and moving the result to its target location. Determine how direct writes should behave for the specified disk location, then add or update coverage showing that checkpoints are saved there without the temporary move; the issue names no file or test to begin with.

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
Quiet
Clarity
Mostly clear
Newbie friendliness
48/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.