An official source code for paper Interpolation-based Contrastive Learning for Few-Label Semi-Supervised Learning, accepted by IEEE TNNLS 2022. Any communications or issues are welcomed. Please contact xihong_edu@163.com. If you find this repository useful to your research or work, it is really appreciate to star this repository. ❤️

Illustration of Interpolation-based Contrastive Learning Semi-Supervised Learning (ICL-SSL) mechanism.

The proposed ICL_SSL is implemented with python 3.8.8 on a NVIDIA 1080 Ti GPU.

Python package information is summarized in requirements.txt:

• torch==1.8.0
• tqdm==4.61.2
• numpy==1.21.0
• tensorboard==2.8.0

python train.py 

If you use code or datasets in this repository for your research, please cite our paper.

@article{yang2022interpolation,
  title={Interpolation-based contrastive learning for few-label semi-supervised learning},
  author={Yang, Xihong and Hu, Xiaochang and Zhou, Sihang and Liu, Xinwang and Zhu, En},
  journal={IEEE Transactions on Neural Networks and Learning Systems},
  volume={35},
  number={2},
  pages={2054--2065},
  year={2022},
  publisher={IEEE}
}