the result on TensorRT10.3 is difference from TensorRT10.8 and is wrong
Open
Nobody has claimed this yet.
triaged
wontfix
- Dominant language
- C++
- Stars
- 13.4k
- Forks
- 2.4k
- Avg merge
- 5d 3h
- Merged PRs (30d)
- 2
Description
The results of tensorRT10.3 are different from those of TensorRT 10.8. The results of 10.8 are as expected, so the results of 10.3 are wrong. I want to use 10.3 because jetpack supports the version of 10.3. My network contains transformers with dynamic dimensions.
I don't know if there is any bug in version 10.3.
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
Reproduce the discrepancy using the reported TensorRT 10.3 and 10.8 environments with the transformer network and its dynamic dimensions. Compare outputs and reduce the case to a minimal reproducible example; the issue is ready for diagnosis when the differing behavior and expected result are documented.
Written by the indexing model from the issue text.
Assessment
- Domain
- machine-learning
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100