Serge-weihao CCNet-Pure-Pytorch issues

Serge-weihao / CCNet-Pure-Pytorch

Criss-Cross Attention (2d&3d) for Semantic Segmentation in pure Pytorch with a faster and more precise implementation.

MIT License

183 stars 21 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

为什么CCNet输入和输出总是一样的呢？

#21 lijing-coder opened 1 year ago
2
About 3D CC Attention module

#20 adchentc opened 1 year ago
0
About 3d version

#19 Limingxing00 closed 3 years ago
2
关于类别一致损失CCL

#18 Asthestarsfalll closed 2 years ago
0
Can you offer the scores you got by the net?

#17 songzijiang closed 3 years ago
1
Num classes

#16 shrutishrestha opened 3 years ago
0
confusion in OhemCrossEntropy loss function

#15 shrutishrestha opened 3 years ago
1
将Pure-Pytorch的RCCA模块应用到视频任务中loss没有完全收敛

#14 Lanezzz opened 3 years ago
1
inplace_abn usage

#13 kisialiou closed 3 years ago
2
padding error

#12 shrutishrestha closed 3 years ago
0
Not an issue - question on maths / paper / relation to this line

#11 8secz-johndpope closed 3 years ago
1
Calculating loss

#10 shrutishrestha opened 3 years ago
2
can this work with cuda toolkit 11.2 + for nvidia 3090 ?

#9 johndpope closed 3 years ago
4
Apex error

#8 shrutishrestha closed 3 years ago
1
rcca compile error

#7 Alva-2020 opened 3 years ago
8
about_CC.py

#6 baba1587 opened 3 years ago
3
why self.gamma =0?

#5 XiXiRuPan opened 3 years ago
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 swjtulinxi opened 4 years ago
1
关于网络最后的return[x, x_dsn]

#3 yearing1017 opened 4 years ago
5
Segmentation fault (core dumped)

#2 YLONl opened 4 years ago
7
precision

#1 swjtulinxi opened 4 years ago
13