gaojie lu
gaojie lu
Eventually the code runs out, but all the labels are in the upper left corner.
Thank you for your reply. I'm using yolov5 v5.0. Running programs are stuck on Ubuntu. The source code also worked on Windows, so I used VS debugging and found data[4]...
I used yolov5-6.1 to export the ONNX model and it worked successfully. (And yolov5n.pt is so fast ) I don't know why. 
I found this reason. The ONNX model output exported by YoloV5-5.0 consists of three output layers. it's my fault. 
啥annox文件?是yolox吗?太久了,不是很清楚,你可以看看我写的这篇博客,是关于这个项目的https://blog.csdn.net/weixin_46596757/article/details/123409536 ------------------ 原始邮件 ------------------ 发件人: ***@***.***>; 发送时间: 2022年9月4日(星期天) 凌晨1:05 收件人: ***@***.***>; 抄送: ***@***.***>; ***@***.***>; 主题: Re: [doleron/yolov5-opencv-cpp-python] the vsion of yolov5 ? (Issue #6) 老哥可以分享下6.1版本的annox文件吗?刚入坑.... — Reply to this email directly, view...