grimoire / mmdetection-to-tensorrt

convert mmdetection model to tensorrt, support fp16, int8, batch input, dynamic shape etc.
Apache License 2.0
590 stars 85 forks source link

Fixed build on OpenCV 4.x, TensorRT 8.x. Fixed TensorRT version cmake info #101

Closed tehkillerbee closed 2 years ago

tehkillerbee commented 2 years ago

Builds successfully on TensorRT 8.2.1, Opencv 4.5.2

tehkillerbee commented 2 years ago

I noticed some issues when building the cpp demo code as per your instructions. This PR fixes these small issues.

I ran link check on my PC and it did not fail, however, it fails when creating the PR. I am not sure why.

grimoire commented 2 years ago

Thanks for the contribution! I will review this PR ASAP. By the way, I am working on mmdeploy. Which supports a lot of networks in OpenMMLab. Please have a try. PRs and Issues are welcome!

grimoire commented 2 years ago

And, errr, you can fix the lint.

tehkillerbee commented 2 years ago

@grimoire Good idea, of course we should make sure trt7 is still supported. I have now fixed it as you suggested. Also thanks for letting me know about mmdeploy! I will check it out asap.

Regarding the failing lint check. On my PC, all tests pass so I am not sure what causes the issue?

pre-commit run --all-files
flake8...................................................................Passed
seed isort known_third_party.............................................Passed
isort....................................................................Passed
yapf.....................................................................Passed
Trim Trailing Whitespace.................................................Passed
Check Yaml...............................................................Passed
Fix End of Files.........................................................Passed
Fix requirements.txt.................................(no files to check)Skipped
Fix double quoted strings................................................Passed
Check for merge conflicts................................................Passed
Fix python encoding pragma...............................................Passed
Mixed line ending........................................................Passed