microsoft / microsoft/onnxruntime
[Feature Request] Provide operator blacklist setting option when the operator does not support tensorrt
Nobody has claimed this yet.
- Dominant language
- C++
- Stars
- 21.9k
- Forks
- 4.2k
- Avg merge
- 4d 11h
- Merged PRs (30d)
- 184
Description
### Describe the feature request
For a model, if there is an operator not supported by tensorrt, it is expected that the operator will degenerate to GPU implementation, and other operators will use tensorrt to accelerate calculation.
### Describe scenario use case
For a model, if there is an operator not supported by tensorrt, it is expected that the operator will degenerate to GPU implementation, and other operators will use tensorrt to accelerate calculation.
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
No files, tests, or entry points are identified in the issue. First clarify the proposed operator blacklist setting and its TensorRT fallback behavior, then define the affected configuration surface and tests needed to verify that unsupported operators use the GPU while other operators use TensorRT.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- cpp
- Domain
- machine-learning, performance
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100