SELFRec is a Python framework for self-supervised recommendation (SSR) which integrates commonly used datasets and metrics, and implements 10+ state-of-the-art SSR models. SELFRec has a lightweight architecture and provides user-friendly interfaces. It can facilitate model implementation and evaluation.
Founder and principal contributor: @Coder-Yu @xiaxin1998
Supported by: @AIhongzhi (A/Prof. Hongzhi Yin, UQ)
- Fast execution: SELFRec is developed with Python 3.7+, Tensorflow 1.14+ and Pytorch 1.7+. All models run on GPUs. Particularly, we optimize the time-consuming procedure of item ranking, drastically reducing the ranking time to seconds (less than 10 seconds for the scale of 10,000×50,000).
- Easy configuration: SELFRec provides a set of simple and high-level interfaces, by which new SSR models can be easily added in a plug-and-play fashion.
- Highly Modularized: SELFRec is divided into multiple discrete and independent modules/layers. This design decouples the model design from other procedures. For users of SELFRec, they just need to focus on the logic of their method, which streamlines the development.
- SSR-Specific: SELFRec is designed for SSR. For the data augmentation and self-supervised tasks, it provides specific modules and interfaces for rapid development.