Contrastive Self-supervised Sequential Recommendation with Robust Augmentation (CoSeRec)
Source code for paper: Contrastive Self-supervised Sequential Recommendation with Robust Augmentation
Model architecture:
Data Augmentations:
Please cite our paper if you use this code.
@article{liu2021contrastive,
title={Contrastive self-supervised sequential recommendation with robust augmentation},
author={Liu, Zhiwei and Chen, Yongjun and Li, Jia and Yu, Philip S and McAuley, Julian and Xiong, Caiming},
journal={arXiv preprint arXiv:2108.06479},
year={2021}
}
Python >= 3.7
Pytorch >= 1.2.0
tqdm == 4.26.0
Four prepared datasets are included in data
folder.
To train CoSeRec on Sports_and_Outdoors
dataset, change to the src
folder and run following command:
python main.py --data_name Sports_and_Outdoors
The script will automatically train CoSeRec and save the best model found in validation set, and then evaluate on test set. You are expected to get following results after training:
'HIT@5': '0.0287', 'NDCG@5': '0.0194', 'HIT@10': '0.0437', 'NDCG@10': '0.0242', 'HIT@20': '0.0635', 'NDCG@20': '0.0292'
You can directly evaluate a trained model on test set by running:
python main.py --data_name Sports_and_Outdoors --model_idx 0 --do_eval
We provide a model that trained on Sports_and_Games dataset in ./src/output
folder. Please feel free to test is out.
- Transformer and training pipeline are implemented based on S3-Rec. Thanks them for providing efficient implementation.