tensorrtx icon indicating copy to clipboard operation
tensorrtx copied to clipboard

Published 20 hours ago • verified publisher icon wang-xinyu

修改efficientnet输出层

Open stealth0414 opened this issue 3 years ago • 2 comments

Env

  • GPU, nano
  • OS, Ubuntu18.04
  • Cuda version 10.2
  • TensorRT version 8.0

About this repo

  • which branch/tag/commit are you using? efficientnet
  • which model? yolov5, retinaface?

Your problem

我训练了一个输出层为71280的efficientnet模型，原efficientnet网络的输出层为10001280 请问如何修改代码生成我的wts文件

stealth0414 avatar Jul 22 '22 14:07 stealth0414

Normally, the gen_wts.py doesn't needs to be modified. But the .cpp code needs, especially the createEngine() function.

wang-xinyu avatar Jul 25 '22 03:07 wang-xinyu

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

stale[bot] avatar Sep 23 '22 03:09 stale[bot]