Research code that accompanies the paper Improving Stock Trend Prediction with Multi-granularity Denoising Contrastive Learning.
The basic version of the article has been accepted by IJCNN2023, and the extended version for the journal has been accepted to Knowledge and Information Systems.
In memory.py, I provided the code for obtaining positive and negative samples based on the memory module. At the same time, in tool.py, I attached the all loss functions used.
Please cite the following paper if you use this code in your work.
@inproceedings{wang2023improving,
title={Improving Stock Trend Prediction with Multi-granularity Denoising Contrastive Learning},
author={Wang, Mingjie and Chen, Feng and Guo, Jianxiong and Jia, Weijia},
booktitle={International Joint Conference on Neural Networks (IJCNN)},
year={2023}
}
@article{wang2023improving,
title={Improving stock trend prediction with pretrain multi-granularity denoising contrastive learning},
author={Wang, Mingjie and Wang, Siyuan and Guo, Jianxiong and Jia, Weijia},
journal={Knowledge and Information Systems},
pages={1--28},
year={2023},
publisher={Springer}
}