deepspeedai / deepspeedai/DeepSpeed
[Question] Deepspeed Transformer Kernel benchmarks
Open
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 43.1k
- Forks
- 5k
- Avg merge
- 4d 15h
- Merged PRs (30d)
- 112
Description
Hi I was looking around the docs and blogs on the fast BERT training but I couldn't find anything showing speedups from the transformer kernel in isolation from the other optimizations. I was wondering if this is available somewhere?
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 reviewing the DeepSpeed documentation and blogs covering fast BERT training and the transformer kernel. Determine whether isolated transformer-kernel speedup benchmarks already exist; done means locating them or documenting reproducible benchmark results that separate this optimization from the others.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- documentation, machine-learning, performance
- Issue type
- Documentation
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100