Semi-supervised Domain Adaptation via Minimax Entropy (ICCV 2019)


pip install -r requirements.txt

The code is written for Pytorch 0.4.0, but should work for other version with some modifications.

Data preparation (DomainNet)

To get data, run


The images will be stored in the following way.



The dataset split files are stored as follows,




At the moment (8/18/2019), we do not publish all data of DomainNet because we hold a competition and some domains are used there.

With regard to office and office home dataset, store the image files in the following ways,

./data/office/amazon/category_name, ./data/office_home/Real/category_name,

We provide the split of office and office-home.


To run training using alexnet,

sh gpu_id method alexnet

where, gpu_id = 0,1,2,3…., method=[MME,ENT,S+T].


This repository is contributed by and Donghyun Kim If you consider using this code or its derivatives, please consider citing:

  title={Semi-supervised Domain Adaptation via Minimax Entropy},
  author={Saito, Kuniaki and Kim, Donghyun and Sclaroff, Stan and Darrell, Trevor and Saenko, Kate},