AI-Hypercomputer / AI-Hypercomputer/tpu-recipes
Add checkpointing guide and configurations to PyTorch TPU recipes
Open
documentation
enhancement
- Dominant language
- Shell
- Stars
- 126
- Forks
- 81
- Avg merge
- 1d 15h
- Merged PRs (30d)
- 3
Description
Add checkpointing guide and configurations to existing PyTorch TPU recipes. The guide should offer solutions for distributed checkpointing and consolidating checkpoints.
Contributor guide
Assessment
This issue has not been assessed yet.