corenel pytorch-adda issues

corenel / pytorch-adda

A PyTorch implementation for Adversarial Discriminative Domain Adaptation

MIT License

473 stars 138 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

How to adapt when the dataset of the target domain is much smaller than source domain

#30 universe27king opened 9 months ago
0
Minimal Code Changes to Support Latest PyTorch and Bug Fixed for Extremely Low Adaptation Accuracy

#29 yuhui-zh15 opened 2 years ago
2
why in adapt.py line 91 label_tgt = make_variable(torch.ones(feat_tgt.size(0)).long())? I think label_tgt = make_variable(torch.zeros(feat_tgt.size(0)).long())

#28 zzzpc opened 3 years ago
2
Could you verify the implementation? Acc = 11% after domain adaptation

#27 lindagaw opened 3 years ago
5
RNN network invalid

#26 zhangqingqing24630 opened 3 years ago
1
Question about the USPS dataset

#25 GaochangWu opened 3 years ago
2
something about eval

#24 mrsempress opened 4 years ago
0
The adaptation does not work

#23 ZhangC2 opened 5 years ago
1
d_loss and accuracy

#22 Dr-Zhou opened 5 years ago
8
error about mnist data shape

#21 SpiritBear000 opened 5 years ago
7
ERROR!!!

#20 jiebiandeshitou opened 5 years ago
1
Questions about zero_grad()

#19 omg777 opened 5 years ago
0
I can't download

#18 OilGao opened 5 years ago
0
I can't download

#17 OilGao opened 5 years ago
0
Question about the loss

#16 taylover-pei opened 5 years ago
1
Adaptation leads to lower precision.

#15 TGISer opened 5 years ago
7
Two feature maps output from discriminator?

#14 aasharma90 opened 5 years ago
0
Anybody visualized feature adaptation?

#13 aasharma90 opened 5 years ago
0
Pretrain+Finetune results?

#12 jinfengr opened 6 years ago
0
Criterion problem

#11 chihhuiho opened 6 years ago
2
0% accuracy with pytorch >= 0.4.0

#10 williford opened 6 years ago
4
adaptation question, discriminator acc=0.5 representing best adaptation?

#9 dupanfei1 opened 6 years ago
2
accuracy just 13% after adaptation?

#8 dupanfei1 opened 6 years ago
8
Thanks ! + testing the "src_only" baseline ...

#7 emmanuelrouxfr opened 6 years ago
3
Update main.py

#6 emmanuelrouxfr closed 6 years ago
0
update USPS download url.

#5 wogong closed 6 years ago
0
Can't load usps_28x28.pkl

#4 IssamLaradji closed 6 years ago
1
USPS to MNIST

#3 redhat12345 opened 6 years ago
2
Losses

#2 redhat12345 opened 6 years ago
4
How to tune the training parameters to rise the accuracy ?

#1 yxyuni opened 6 years ago
0