houqb SPNet issues - Githubissues

houqb / SPNet

Code for our CVPR2020 paper "Strip Pooling: Rethinking Spatial Pooling for Scene Parsing"

MIT License

387 stars 57 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

About MPM and SPM

#23 Haus226 opened 2 months ago
0
What is the visualization method in the article？

#22 litao-debug opened 1 year ago
0
Get nan loss when training with StripPooling

#21 DISAPPEARED13 opened 2 years ago
0
can't reduce the value of “MainLoss ”

#20 zhk779 opened 2 years ago
0
What data are the pretrained Resnet 50 and 101 models trained on?

#19 maopal opened 2 years ago
0
Need your suggestions ?

#18 haideralimughal opened 3 years ago
0
When reproducing experimental results on Cityscapes, should i use the CrossEntropy loss or OHEM-CrossEntropy version

#17 Bio-shine opened 3 years ago
0
我没有足够的GPU，batchsize只能设置4，请问对结果影响大吗

#16 swjtulinxi opened 3 years ago
2
Why is there only 19 categories in cityscapes_colors.txt and no background categories?

#15 Dumbledore-bit opened 4 years ago
0
download pretrained

#14 liushuang2140 opened 4 years ago
0
how to reproduce the result for cityscapes

#13 SHMCU opened 4 years ago
11
AssertionError: Non-existent key: ade20k

#12 adwxy opened 4 years ago
0
bias=False for Conv2d when using BatchNorm

#11 ShoufaChen closed 4 years ago
0
which part is the MPM module?

#10 xuxiuzhi2627 opened 4 years ago
4
Why the pool_size is set to (20, 12) in the SPHead?

#9 yiminglin-ai opened 4 years ago
3
pretrained models download failed

#8 jinhaijian closed 4 years ago
2
About the SPM

#7 yexiguafuqihao closed 4 years ago
3
About SPM structure, source code is different from the paper

#6 weiyu-zeng closed 4 years ago
2
About SPM module

#5 anti199994 closed 4 years ago
1
about strip pooling

#4 swjtulinxi closed 4 years ago
8
Hello, thank you for your excellent contributions. I have a question. I want to ask, I always get the error of object has no attribute '_up_kwargs' when I add the strippooling module to my own network. What is the reason? My code is as follows,

#3 houqian2180320171 closed 4 years ago
3
Hello, thank you for your excellent contributions. I have a question. I want to ask, I always get the error of object has no attribute '_up_kwargs' when I add the strippooling module to my own network. What is the reason? My code is as follows, def __init __ (self, nDenselayer, channels, growth, norm_layer = nn.BatchNorm2d, ** kwargs): super (RDB1, self) .__ init __ () modules = [] channels_buffer = channels for i in range (nDenselayer): modules.append (make_dense1 (channels_buffer, growth, 3, norm_layer, self._up_kwargs))

#2 houqian2180320171 closed 4 years ago
0
About the state of the art.

#1 MandyMo closed 4 years ago
0