DS3L

This is the code for paper "Safe Deep Semi-Supervised Learning for Unseen-Class Unlabeled Data" published in ICML 2020.

Setups

The code is implemented with Python and Pytorch.

Running D3SL for benchmark datasets

Here is an example:

python train.py --dataset MNIST --ratio 0.6 --n_labels 60 --iterations 200000

Acknowledgements

We thank the Pytorch implementation on Meta-Net (https://github.com/xjtushujun/meta-weight-ne) and learning-to-reweight-examples(https://github.com/danieltan07/learning-to-reweight-examples).

Contact

If you have any questions, please contact Lan-Zhe Guo ([email protected]).