SIMGA: Selective Gated Mamba for Sequential Recommendation

This is the implementation of the submission "SIGMA: Selective Gated Mamba for Sequential Recommendation".

Configuration of the environment

The hardware and software we used are listed below to facilitate the environment's configuration. The detailed environment setting can be found in the requirements.txt. You can use pip install to reproduce the environment.

• Hardware:
  • GPU: one NVIDIA L4
  • CUDA: 11.8
• Software:
  • Python: 3.10.13
  • Pytorch: 2.1.1 + cu118
• Usage
  • Install Causal Conv1d
    • pip install causal-conv1d==1.1.3.post1
  • Install Recbole
    • pip install recbole==1.2.0
  • Install Mamba
    • pip install mamba-ssm==1.1.4 A detailed configuration process in Colab can be found in the RecMamba.ipynb

Datasets

The procedures for preprocessing the datasets are listed as follows:

• The raw datasets should be preprocessed using the Conversion tool provided by https://github.com/RUCAIBox/RecSysDatasets. After you acquire the atomic files, please put them into dataset/<Amazon_Fashion/Amazon_Sports_and_Outdoors/Amazon_Video_Games/amazon-beauty/ml-1m/yelp>. The Yelp Dataset can be found at https://www.yelp.com/dataset; The Amazon datasets can be found at https://cseweb.ucsd.edu/jmcauley/datasets.html\#amazon_reviews; The MovieLens-1M dataset can be found at https://grouplens.org/datasets/movielens/.
• Or you can directly download the atomic files of these datasets using the Baidu disk link provided by Recbole: https://github.com/RUCAIBox/RecSysDatasets.

For the procedure of filtering the cold-start users and items, please find the corresponding part in the model/config.yaml

Model Training

• You can directly run the model/run.py to reproduce the training procedure.
• For evaluation on grouped users, please refer to model/run_custrainer.py.

Citation

If you found the code and the paper are useful, please kindly cite our paper:

@article{liu2024bidirectional,
  title={Bidirectional gated mamba for sequential recommendation},
  author={Liu, Ziwei and Liu, Qidong and Wang, Yejing and Wang, Wanyu and Jia, Pengyue and Wang, Maolin and Liu, Zitao and Chang, Yi and Zhao, Xiangyu},
  journal={arXiv preprint arXiv:2408.11451},
  year={2024}
}