Bobo-y/flexible-yolov5
More readable and flexible yolov5 with more backbone(gcn, resnet, shufflenet, moblienet, efficientnet, hrnet, swin-transformer, etc) and (cbam,dcn and so on), and tensorrt
PythonGPL-3.0
Pinned issues
Issues
- 1
Quantization error
#159 opened by ratom - 0
SE and CBAM module implementation error
#160 opened by ratom - 0
- 30
- 5
repvgg做为backbone时,保存的pt模型是多分支结构还是单路结构?
#156 opened by wjli-debug - 3
Error help please
#154 opened by ratom - 1
Drop layer
#153 opened by ratom - 1
- 11
error during compile
#151 opened by PraveenMNaik - 5
ONNX: export failure: Expected all tensors to be on the same device, but found at least two devices, cpu and cuda:0! (when checking argument for argument other in method wrapper__equal)
#150 opened by taoxunqiang - 5
使用mobilenetv3作为主干网络,无法导出onnx,并且没有错误提示。
#147 opened by LaoJiuMoYang - 2
- 2
Add True negative data to training dataset
#148 opened by a227799770055 - 3
请问模型的flops应该在哪输出
#144 opened by cds221 - 1
About the Resume
#146 opened by zhaoxixixi - 2
作者大大是否有跟进YOLOv5 V7.0版本实例分割功能的打算?
#145 opened by pengcheng4181452 - 2
0 gradients 问题
#138 opened by huangzizheng01 - 4
请问yaml文件中有关head的网络结构
#143 opened by cds221 - 3
- 1
更改backbone
#141 opened by AustinzzZ-1997 - 1
- 1
- 2
- 4
MobilenetV3主干网络中scale的作用是什么?
#134 opened by LaoJiuMoYang - 1
YOLOv5中验证代码存在一定问题,需要更改
#132 opened by Huster-Hq - 2
我用HRNet32作Backbone时,训练完出现了RuntimeError: Expected all tensors to be on the same device, but found at least two devices, cuda:0 and cpu!这个问题
#131 opened by ChenShuai-CA - 1
想咨询一下如何在验证图片的时候显示置信度呢
#129 opened by jiezai18 - 1
请问v2分支是基于yolov5 6.1版本吗?
#130 opened by cds221 - 3
关于模型搭建和初始化权重时的若干问题
#128 opened by danchaofan-git - 12
- 2
u版现在yolov5s的map为37.4,但本项目只有35
#122 opened by nobody-cheng - 2
@Bobo-y
#123 opened by farzadimanpour - 4
Tips on training using Swin-T on YOLO?
#125 opened by sarmientoj24 - 3
- 9
采用swin transoforme的预训练权重问题
#121 opened by cds221 - 4
- 2
如何添加FPN层以实现更好的小目标检出
#97 opened by cydiachen - 14
- 0
- 1
RuntimeError: Expected all tensors to be on the same device, but found at least two devices, cuda:0 and cpu!
#111 opened by Samuelmicer - 7
- 7
Squeeze and Excitation Conv vs Linear
#117 opened by seermer - 1
GNN is Graph Neural Network?
#116 opened by daiyuanm - 2
cls_loss is always 0
#113 opened by a227799770055 - 3
Can you share some weights to run the network?
#112 opened by qianmin - 3
windows下:模型和数据集的yaml文件找不到
#110 opened by StandGreat - 3
- 35
c++ sample code for trt inference
#98 opened by kaishijeng - 14
似乎因为模型细微错误导致精度低2个点!?
#96 opened by LinB203 - 1
un-reproducible result
#101 opened by wykhan