yuanli2333 Teacher-free-Knowledge-Distillation issues

yuanli2333 / Teacher-free-Knowledge-Distillation

Knowledge Distillation: CVPR2020 Oral, Revisiting Knowledge Distillation via Label Smoothing Regularization

MIT License

580 stars 67 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

I trained the resnet50 baseline model for 500 rounds, but the accuracy obtained was only 71%

#40 panshoudeng opened 1 month ago
0
Train your own model

#39 panshoudeng opened 2 months ago
0
Bump werkzeug from 0.15.4 to 2.2.3

#38 dependabot[bot] opened 1 year ago
0
Bump werkzeug from 0.15.4 to 0.15.5

#37 dependabot[bot] closed 1 year ago
1
Bump certifi from 2018.8.24 to 2022.12.7

#36 dependabot[bot] opened 1 year ago
0
Does this method work on the detection tasks？

#35 fmaaf opened 2 years ago
0
Bump protobuf from 3.15.0 to 3.18.3

#34 dependabot[bot] opened 2 years ago
0
KD loss is zero

#33 minato1000 opened 2 years ago
0
Bump numpy from 1.21.0 to 1.22.0

#32 dependabot[bot] opened 2 years ago
0
Does this work for dataset with only two classes

#31 wugh opened 2 years ago
0
Bump numpy from 1.15.2 to 1.21.0

#30 dependabot[bot] closed 2 years ago
0
Bump protobuf from 3.6.0 to 3.15.0

#29 dependabot[bot] closed 2 years ago
0
Torch Vision Version

#28 Amik-TJ opened 3 years ago
0
Working with larger image size

#27 sri9s opened 3 years ago
0
Bump urllib3 from 1.25.8 to 1.26.5

#26 dependabot[bot] closed 2 years ago
0
Bump urllib3 from 1.25.3 to 1.25.8

#25 dependabot[bot] closed 3 years ago
0
Question about the loss function of Tf-reg KD

#24 HowieMa opened 3 years ago
1
Data augmentation for Tiny-ImageNet

#23 aryanasadianuoit opened 3 years ago
0
Difference between L_REG and LSR

#22 real-brilliant opened 3 years ago
1
Implementation doesn't have loss_soft_regularization and loss_fn_kd for ImageNet dataset

#21 sainatarajan opened 4 years ago
0
The baseline of ResNet18 on CIFAR100 is relatively lower

#20 JosephChenHub opened 4 years ago
3
Mismatch between Eq.9 in the paper and the code

#19 MingSun-Tse opened 4 years ago
4
What is the difference between Born Again Network and your self-training KD method?

#18 JiyueWang closed 4 years ago
2
Have you ever try on deeper network?

#17 JiyueWang closed 4 years ago
3
How to search the best temperature and alpha

#16 TimeBear opened 4 years ago
5
Pretrained model for student network

#15 he-y opened 4 years ago
2
TFselftraining parameters in the paper ?

#14 Shiro-LK closed 4 years ago
1
Resnet architectures is differnet from the original networks in the paper

#13 peipei-pig closed 4 years ago
2
It just feels like "炼丹"

#12 ykk648 closed 4 years ago
1
do you have email? I have some trouble with your code.

#11 TimeBear opened 4 years ago
2
where's the paper?

#10 vraivon closed 4 years ago
2
Create LICENSE

#9 yuanli2333 closed 4 years ago
0
a question about mobilenetv2

#8 lansss closed 4 years ago
0
Many Bugs

#7 SunCherry closed 4 years ago
0
why there is a 'multiplier' param in the loss funtion?

#6 luanyunteng closed 4 years ago
2
Questions about KD loss

#5 Paper99 closed 4 years ago
5
Can't download the pre-trained model

#4 SunCherry opened 5 years ago
3
Question about KD Regularization in code

#3 GengZ closed 5 years ago
3
questions about The two Tf-KD methods

#2 pecanjk closed 4 years ago
6
wonder if it work on a weak and small student network

#1 pecanjk closed 4 years ago
1