mashaan14 / adda-toy

A pytorch implementation of ADDA paper published in CVPR2017 with 2D toy example.

Hi there 👋

I'm Mashaan (مشعان)

Education

Sydney Uni ---> PhD

KFUPM -------> MS

Check out my YouTube channel

adda-toy's Issues

Low Adaptation Accuracy

Sometimes I'm getting a low adaptation accuracy. I think what causes the low adaptation accuracy is that the class labels are swapped by the target encoder. This makes sense because it is an unsupervised task and the target encoder didn't see the class labels.

You can see in the attached image that the target encoder separates the classes well. But the class labels were swapped.

If you can think of a way how to align the class labels for the target encoder, please let me know.