deepspeedai / deepspeedai/DeepSpeed
[REQUEST] LOMO Low Memory Optimization
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 43.1k
- Forks
- 5k
- Avg merge
- 4d 15h
- Merged PRs (30d)
- 112
Description
Is your feature request related to a problem? Please describe.
LOMO is a method that offers substantial memory saving benefits by fusing the gradient backpropagation and update step together.
Describe the solution you'd like
Implementation of the solution provided here https://github.com/OpenLMLab/LOMO
Context:
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 reading the linked OpenLMLab/LOMO implementation and inspecting the DeepSpeed repository for the training and optimizer entry points where such an integration could belong. No DeepSpeed files, tests, or acceptance criteria are named; done would require a defined implementation and validation of the requested low-memory optimization.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python, pytorch
- Domain
- machine-learning, performance
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 20/100