tensorrtx icon indicating copy to clipboard operation
tensorrtx copied to clipboard

about yolov7 with wts and engine

Open QIANXUNZDL123 opened this issue 1 year ago • 2 comments

我是用gen_wts.py 获得了模型的wts文件(该模型基于yolov7-tiny训练),成功的使用api搭建了网络,同时也生成了engine文件,但是我推理的时候出现什么也检测不到,我的推理后处理均是基于该项目的yolov5,这是什么原因导致的呢。。。 image image

QIANXUNZDL123 avatar Jul 13 '22 08:07 QIANXUNZDL123

You can debug it by comparing the layer output value between trt and pytorch.

For example, if conv54 output values are the same, it means all the layers before conv54 are correct.

wang-xinyu avatar Jul 18 '22 04:07 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 16 '22 07:09 stale[bot]