Serge-weihao/CCNet-Pure-Pytorch
Criss-Cross Attention (2d&3d) for Semantic Segmentation in pure Pytorch with a faster and more precise implementation.
PythonMIT
Issues
- 2
为什么CCNet输入和输出总是一样的呢?
#21 opened by lijing-coder - 0
About 3D CC Attention module
#20 opened by adchentc - 0
关于类别一致损失CCL
#18 opened by Asthestarsfalll - 2
About 3d version
#19 opened by Limingxing00 - 1
Can you offer the scores you got by the net?
#17 opened by songzijiang - 0
Num classes
#16 opened by shrutishrestha - 1
将Pure-Pytorch的RCCA模块应用到视频任务中loss没有完全收敛
#14 opened by Lanezzz - 1
- 2
inplace_abn usage
#13 opened by kisialiou - 1
Apex error
#8 opened by shrutishrestha - 0
padding error
#12 opened by shrutishrestha - 1
- 2
Calculating loss
#10 opened by shrutishrestha - 4
- 8
rcca compile error
#7 opened by Alva-2020 - 3
about_CC.py
#6 opened by baba1587 - 1
why self.gamma =0?
#5 opened by XiXiRuPan - 1
关于评估代码与原始代码有些不同,请大佬帮忙解释一下疑惑?
#4 opened by swjtulinxi - 7
Segmentation fault (core dumped)
#2 opened by YLONl - 5
关于网络最后的return[x, x_dsn]
#3 opened by yearing1017 - 13
precision
#1 opened by swjtulinxi