how to analyze the performance of each operator（myelin）?
Open
Nobody has claimed this yet.
triaged
- Dominant language
- C++
- Stars
- 13.4k
- Forks
- 2.4k
- Avg merge
- 5d 3h
- Merged PRs (30d)
- 2
Description
tensorrt 8.6.10
If operators are fused into myelin, how to analyze the performance of each operator?
Are there any tools or samples?
The core purpose is to optimize the overall inference performance, such as replacing some operators fused by meylin (first we need to find it)
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 TensorRT 8.6.10 and Myelin but no files, tests, or entry points. Start by checking the available TensorRT profiling tools or samples for fused Myelin operators. Done would mean identifying how to attribute performance to individual fused operators and documenting a way to evaluate replacements for overall inference performance.
Written by the indexing model from the issue text.
Assessment
- Domain
- machine-learning, performance
- Issue type
- Documentation
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100