search

HuangJunJie2017 / BEVDet

Official code base of the BEVDet series .
Apache License 2.0
1.44k stars 266 forks source link

TRT cannot be converted correctly when setting bev height larger than one #335

Open YunzheWu-404 opened 7 months ago

YunzheWu-404 commented 7 months ago

Hi

We tried to generate the model into trt with the bev height of 5 when dealing with the occupancy network . We realized that the trt was generated with the output of bev pooling to be 1x128x128x320.

However, with our setting, we successfully obtained the bev pool out as 1x5x128x128x320 during torch training, inference and trans of onnx.

We would really appreciate if we could hear back from you.

Many thanks

Thanks for your error report and we appreciate it a lot.

Checklist

  1. I have searched related issues but cannot get the expected help.
  2. The bug has not been fixed in the latest version.

Describe the bug A clear and concise description of what the bug is.

Reproduction

  1. What command or script did you run?
A placeholder for the command.
  1. Did you make any modifications on the code or config? Did you understand what you have modified?
  2. What dataset did you use?

Environment

  1. Please run python mmdet3d/utils/collect_env.py to collect necessary environment information and paste it here.
  2. You may add addition that may be helpful for locating the problem, such as
    • How you installed PyTorch [e.g., pip, conda, source]
    • Other environment variables that may be related (such as $PATH, $LD_LIBRARY_PATH, $PYTHONPATH, etc.)

Error traceback If applicable, paste the error trackback here.

A placeholder for trackback.

Bug fix If you have already identified the reason, you can provide the information here. If you are willing to create a PR to fix it, please also leave a comment here and that would be much appreciated!