This is the repository of the implementtion for the presented method in our paper: "Extending Interpolation Consistency Training for Unsupervised Domain Adaptation" by Shayan gharib and Arto Klami.
Our paper is accepted for publication at the International Joint Conference on Neural Networks (IJCNN) 2023.
To run an experiment, use the file main.py:
python main.py --index [experiment_index]
To change the settings of each experiment, use the setting yaml file: settings.yml.
To run the experiment for MNIST --> MNIST-M setup, the MNIST-M dataset needs to be manually downloaded from this link, and be placed in "datasets/MNISTM" directory. The other two datasets (i.e. MNIST and USPS) are automatically downloaded and processed.