Search results

Failure to Detection objects in the engine model(yolov9)

Status: Open.

#1643 In wang-xinyu/tensorrtx;

· jackkal2 opened on Jul 19, 2025

您好，我按照您的方法将自己训练的仅检测行人一个类别的yolov7-tiny.pt转化得到.engine,用它推理图片时，发现每个行人的都有好几个大小不一的检测框，唯一的修改是在config.h将类别改为1了，请问这是为什么？

Status: Open.

#1640 In wang-xinyu/tensorrtx;

· zzzzzz1112 opened on Jun 27, 2025

运行yolov11_cls.cpp和yolov11(ultralytics)的predict.py的运行结果存在不一致

Status: Open.

#1639 In wang-xinyu/tensorrtx;

· yuan-kai-design opened on Jun 13, 2025

rcnn 里面是c4的,要FPN的,改config就可以了,里面的cpp和cu代码不用改吧?

Status: Open.

#1638 In wang-xinyu/tensorrtx;

· zouwen198317 opened on Jun 11, 2025

yolov12用cmake配置好环境后，能正常生产exe文件，但在wts转engine时报实际加载的权重数量只有是预期的一半，请问怎么处理？

Status: Open.

#1636 In wang-xinyu/tensorrtx;

· LDQ520LDQ opened on May 27, 2025

yolov8及11等，多batch推理时，context.enqueue纯推理耗时是单batch的成倍增加，并没有加速？

wontfixThis will not be worked onThis will not be worked on

Status: Open.

#1634 In wang-xinyu/tensorrtx;

· Mapalas opened on May 15, 2025