export error: export detection model from pytorch to tensorrt

Open angiewlz opened this issue 2 years ago • 3 comments

Hello! I'm attempting to export detection model from pytorch to tensorrt, but there is an error below: ImportError: cannot import name 'register_all_modules' from 'mmdet.utils.setup_env' (/xxx/.conda/envs/internimage/lib/python3.7/site-packages/mmdet/utils/setup_env.py)

Is this mmdet version problem?

Apr 19 '23 00:04 angiewlz

@angiewlz I meet the same problem, can you save the problem?

May 29 '23 02:05 xuweidongkobe

It is a problem of the mmdeploy version. I solved it by using the version 0.14.0.

Dec 08 '23 02:12 goldentimecoolk

It is a problem of the mmdeploy version. I solved it by using the version 0.14.0.

@goldentimecoolk When I convert the model to tensorrt, I want to infer it with C++ API something was wrong,is the version problem?

Jan 30 '24 01:01 GitHubYuxiao