SSARCandy DeepCORAL issues

SSARCandy / DeepCORAL

🧠 A PyTorch implementation of 'Deep CORAL: Correlation Alignment for Deep Domain Adaptation.', ECCV 2016

https://ssarcandy.tw/2017/10/31/deep-coral/

226 stars 42 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

The CORAL Loss and Forbenius Norm is defined differently and the results dont match.

#21 ashlytom opened 3 years ago
0
fix errors, now can run and give results arround 56% on target domain

#20 GlenGGG closed 3 years ago
0
I think you have a error.

#19 typhoon1104 opened 4 years ago
2
TypeError: unsupported operand type(s) for /: 'tuple' and 'int'

#18 tjzjp opened 5 years ago
0
CORAL loss is defined differently from the original paper

#17 DenisDsh opened 6 years ago
9
Error in calculation of covariance matrix

#16 harshitbansal05 closed 6 years ago
1
The point of self.target_fc?

#15 georgeadam closed 6 years ago
1
Why it's loss = loss/(4*d*4) ?

#14 jindongwang closed 6 years ago
1
self.target_fc never used

#13 fmcarlucci closed 6 years ago
1
Any update results like Deep CORAL paper ?

#12 deep0learning opened 6 years ago
0
code about testing

#11 cmhungsteve closed 6 years ago
6
A line in main.py().

#10 debasmitdas closed 6 years ago
1
Low Accuracy for SHVN ---> MNIST

#9 deep0learning opened 6 years ago
5
Pretrained model

#8 deep0learning closed 6 years ago
10
The reason for setting _lambda ascending ?

#7 huster-wgm opened 6 years ago
1
Correlation Loss

#5 redhat12345 closed 6 years ago
13
Inputs to the CORAL loss

#4 cmhungsteve closed 6 years ago
3
Graph

#3 redhat12345 closed 6 years ago
3
Test accuracy is low than original test accuracy in the paper.

#2 redhat12345 closed 6 years ago
20
Error !!!

#1 redhat12345 closed 6 years ago
1