Scalable Bayesian Tensor Ring Factorization for Multiway Data Analysis

Description

This is the official implementation for the following paper presented in ICONIP 2023.

@inproceedings{tao2023scalable_btr,
  title={Scalable Bayesian Tensor Ring Factorization for Multiway Data Analysis},
  author={Tao, Zerui and Tanaka, Toshihisa and Zhao, Qibin},
  booktitle={International Conference on Neural Information Processing},
  year={2023},
  publisher={Springer International Publishing}
}

The code is mainly based on Python 3.9 and PyTorch 1.12.1.

Run pip install -r requirements.txt to install all required libraries.

Examples

1. To reproduce the experiment on rank estimation, run ./simulation_rank.py file.

2. Example of running continuous data completion is in ./ushcn.py.

3. Example of running binary data completion is in ./enron.py.