Skip to content

Code release for Discriminative Adversarial Domain Adaptation (AAAI2020).

Notifications You must be signed in to change notification settings

Gorilla-Lab-SCUT/DADA-AAAI2020

Repository files navigation

DADA-AAAI2020

Code release for Discriminative Adversarial Domain Adaptation (AAAI2020).

The paper is avaliable here.

Requirements

  • Python 3.6.8
  • Pytorch 1.0.0

Dataset

The structure of the dataset should be like

VisDA
|_ visda_train
|  |_ aeroplane
|     |_ <im-1-name>.jpg
|     |_ ...
|     |_ <im-N-name>.jpg
|  |_ bicycle
|     |_ <im-1-name>.jpg
|     |_ ...
|     |_ <im-N-name>.jpg
|  |_ ... (omit 9 classes)
|  |_ truck
|     |_ <im-1-name>.jpg
|     |_ ...
|     |_ <im-N-name>.jpg
|_ visda_validation_first6classes
|  |_ aeroplane
|     |_ <im-1-name>.jpg
|     |_ ...
|     |_ <im-N-name>.jpg
|  |_ bicycle
|     |_ <im-1-name>.jpg
|     |_ ...
|     |_ <im-N-name>.jpg
|  |_ ... (omit 3 classes)
|  |_ knife
|     |_ <im-1-name>.jpg
|     |_ ...
|     |_ <im-N-name>.jpg
|_ ...

Training

Replace paths and domains in run_office31_partial.sh with those in your own system.

Replace paths and domains in run_visda_partial.sh with those in your own system.

Citation

@InProceedings{dada,   
  title={Discriminative Adversarial Domain Adaptation},   
  author={Hui Tang and Kui Jia},   
  booktitle={Association for the Advancement of Artificial Intelligence (AAAI)},   
  year={2020},
}

About

Code release for Discriminative Adversarial Domain Adaptation (AAAI2020).

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published