tensorrtx

tensorrtx copied to clipboard

This error appears when compiling YOLov6 6.0. Can anyone help someone help?

2

严重性 代码 说明 项目 文件 行 禁止显示状态 错误 LNK2019 无法解析的外部符号 "void __cdecl preprocess_kernel_img(unsigned char *,int,int,float *,int,int,struct CUstream_st *)" (?preprocess_kernel_img@@YAXPEAEHHPEAMHHPEAUCUstream_st@@@Z)，函数 main 中引用了该符号 yolov5 D:\yanyi\project_process\Completed_projects\yolo-deploy\tensorrtx-master\yolov5\build\yolov5.obj 1

songjiahao-wq

SWIN transformer pretrained model issue

2

## About this repo - commit - https://github.com/wang-xinyu/tensorrtx/commit/0f944b89750c2cd2e6be7933f84d61eb4e34ec83? - model - SWIN-Transformer ## Your problem ``` python gen_wts.py upernet_swin_tiny_patch4_window7_512x512.pth Traceback (most recent call last): File "gen_wts.py", line 8, in model...

abhiagwl4262

about yolov7 with wts and engine

2

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。。。  

QIANXUNZDL123

请问我对YOLOv5结构做了一些调整比如加了注意力模块，Ghost卷积这种，是不是没有办法用这个repo了

6

如题

15050188022

how about https://github.com/WongKinYiu/yolov7 trt convert?

2

yolov7 just conming.this project will suport pt->wts->trt?

futureflsl

Cannot deserialize plugin since corresponding IPluginCreator not found in Plugin Registry

2

## Env Jetson NX + Cuda 10.2 + TRT 8.2.1.8 ## About this repo SlowFast （RoI Head） ## Your problem When I run the code: doInference_slowfast(*context_action, data_fast, data_slow, proposals, prob_data,...

Cuzny

Question about yolov7

2

Thank you for your awesome implementation of yolo models. I wonder are you planning to integrate yolov7 to tensorrtx?

spacewalk01

Yolov5 v4.0

2

支持tensorRT 8

greenkarson

docker里面编译yolov5-4.0 make报错 版本yolov5-4.0 tensorrt 6.0.1.8 容器cuda10.2 主机cuda11.2

2

具体报错如下： [root@ecs-7d42 build]# cmake .. -- The C compiler identification is GNU 4.8.5 -- The CXX compiler identification is GNU 4.8.5 -- Check for working C compiler: /usr/bin/cc -- Check...

yinhai86924

int8转换问题，能转成，但检测不到目标

3

## Env ## Env - GPU,RTX1080Ti. - Ubuntu18.04 - Cuda 10.2 - TensorRT 7.0.0 ## About this repo yolov5l/yolov5m/yolov5s ## Your problem yolov5l 使用int8，make/转换都可以，推理时间明显减少，10ms即可，但是检测不到目标，阈值设置很低了都不行。 yolov5s/yolov5m,make/转换都可以，推理时间没有明显变化，有的甚至增加了，但是能检测到目标。 还在找问题

laomaup