deepspeedai / deepspeedai/DeepSpeed
Run preprocessing steps only once
Open
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 43.1k
- Forks
- 5k
- Avg merge
- 4d 15h
- Merged PRs (30d)
- 112
Description
In a large language model code which is launched using DeepSpeed, how do we run the initial loading and data preprocessing steps only once and then share them with all the processes, instead of replicating the same data preprocessing steps in each of the GPUs?
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
The issue names no files, tests, or entry points. Begin by locating the DeepSpeed process and data-loading path; done should be an agreed implementation that preprocesses data once, shares the result across processes, and verifies that behavior.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- distributed-systems, machine-learning
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100