irfanICMLL structure_knowledge_distillation issues

irfanICMLL / structure_knowledge_distillation

The official code for the paper 'Structured Knowledge Distillation for Semantic Segmentation'. (CVPR 2019 ORAL) and extension to other tasks.

BSD 2-Clause "Simplified" License

708 stars 103 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

R.I.P.

#71 yajieC opened 1 year ago
8
training log

#70 songyang86 opened 1 year ago
1
How to apply KD for two input images?

#69 Tejuwi opened 1 year ago
0
Training from scratch reached the performance similar with that when initializing from ImageNet

#68 Shawn207 opened 2 years ago
1
/data/ckj/SKD-master/libs/_ext/__ext.so: failed to map segment from shared object

#67 asiguan opened 2 years ago
0
code error

#66 Liuhao-128 opened 2 years ago
0
concat RGB images and seg maps as input of discriminator

#65 Lyan-ing opened 2 years ago
0
FileNotFoundError: [Errno 2] No such file or directory: './ckpt/CS_scenes_39326_0.75.pth'

#64 GeneralJing opened 3 years ago
0
How can I run build.py in pytorch1.9?

#63 BBBBchan closed 3 years ago
10
Holistic loss

#62 rose-jinyang opened 3 years ago
2
__ext.so error: undefined symbol: __cudaRegisterFatBinaryEnd

#61 RonDen opened 3 years ago
1
softmax should be calculated along the last dimension(-1), not dim 1

#60 zoucheng1991 closed 9 months ago
1
About the input of discriminator

#59 blessingforever opened 3 years ago
0
Pretrained student?

#58 mbahri closed 3 years ago
2
How to apply pair-wise distillation on depth prediction task

#57 mzy97 opened 3 years ago
1
the pixelwise loss never decreased during training

#56 DHuiTnut opened 3 years ago
4
Train script

#55 GeneralJing closed 3 years ago
1
about the code

#54 15757170756 opened 4 years ago
2
checkpoint

#53 NickChang97 closed 4 years ago
0
Puzzled about this loop

#52 NickChang97 closed 3 years ago
1
Why the patch size turn out to be half of the original feature map size in CriterionPairWiseforWholeFeatAfterPool?

#51 Ssssseason closed 3 years ago
3
Why structure knowledge distillation is effective?

#50 xiaozhimabing opened 4 years ago
2
How to run on my own dataset?

#49 eskeron-yxgg opened 4 years ago
4
ESPNet training using 'master' & 'cvpr2019' branch

#48 betogulliver opened 4 years ago
1
Example training method for Depth Estimation VNL network

#47 atrah22 closed 3 years ago
2
The paper use average pooling to aggregate β × C features in one node to be 1 × C, but the code shows max pooling.

#46 aye0804 opened 4 years ago
1
Pixel-wise loss

#45 duanxuesong closed 4 years ago
5
TypeError: add_() received an invalid combination of arguments - got (complex, Tensor)

#44 lmunan closed 4 years ago
5
training fails with "RuntimeError: cuda runtime error (11) : invalid argument at THCGeneral.cpp:405"

#43 betogulliver opened 4 years ago
2
retrain teacher net

#42 wl082013 closed 4 years ago
2
CUDA Error encountered in <function CompiledLib.bn_forward_cuda at 0x7fef222d0400>

#41 zhLawliet closed 4 years ago
1
no models for detection and depth estimation

#40 wl082013 closed 4 years ago
3
Can you provide the ResNet(1.0) baseline train log?

#39 aye0804 closed 3 years ago
1
Hello, I also cannot reproduce the results

#38 DanceInDark closed 3 years ago
6
Build error

#37 shidilrzf opened 4 years ago
1
build error

#36 wl082013 closed 4 years ago
2
Could you provide your training log?

#35 happycoding1996 closed 4 years ago
3
pixel-wise loss dominates generator training

#34 ArsenLuca closed 4 years ago
3
codes for PAMI version

#33 ArsenLuca closed 4 years ago
2
ESPNet training

#32 sungsooo closed 4 years ago
2
Question about the pair-wise loss

#31 colorjam closed 4 years ago
1
The source code of FCOS

#30 miaoshuyu closed 4 years ago
1
I can't get your mIoU on cityscapes dataset.

#29 mailhoho closed 4 years ago
3
D_loss becomes NAN during training

#28 zhouzg closed 4 years ago
2
about pair-wise and pixel-wise distillation for FCOS

#27 Haiyang21 closed 4 years ago
4
run_train_val.sh: --pi: not found

#26 YBIO closed 4 years ago
2
nan loss value

#25 sungsooo closed 4 years ago
4
Where is Discriminator weights

#24 swordlidev closed 4 years ago
1
TypeError: add_() received an invalid combination of arguments

#23 w11m closed 4 years ago
2
Hi, I cannot reproduce the result.

#22 aachenhang closed 4 years ago
4