[Feature request] Save optimizer states of learnable sparse embedding.
Open
enhancement
- Dominant language
- Python
- Stars
- 452
- Forks
- 76
- PR merge metrics
- No merged PRs in 30d
Description
Currently, GraphStorm does not save/load optimizer states of learnable sparse embeddings. (https://github.com/awslabs/graphstorm/blob/main/python/graphstorm/model/utils.py#L336-L337). We need to support it.
Contributor guide
Research direction
Start in python/graphstorm/model/utils.py at lines 336-337 and trace the existing save/load handling for learnable sparse embeddings and optimizer state. Compare the surrounding checkpoint flow to determine where support is missing; done means sparse embedding optimizer states are saved and restored correctly.
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
- Needs clarification
- Newbie friendliness
- 35/100