Skip to content

Implement our enhanced loss base on CIKM2020-S3Rec repo, and provide tutorials to implement enhanced loss at Aprec repo.

Notifications You must be signed in to change notification settings

Li-fAngyU/sequential_rec

 
 

Repository files navigation

Our code forked from https://github.com/aHuiWang/CIKM2020-S3Rec

We provide the replication code for all experimental results. The specific steps are as follows:

Replicating Results of Enhanced Loss Function on CIKM2020-S3Rec

Preparation

pip install -r requirements.txt

Replicating Results

Introduction img result.

introduction We provide a simple script file for one-step reproduction of the experimental results presented in the introduction section.

python run_introduction_experiment.py

Result save at ./output/introduction/All_result.txt, and the training log save at ./output/introduction/.*

Experiment table result.

Table2

We provide a convenient way to reproduce the experimental results in the table:

sh table2_results.sh

Result will save at output/ folder, and our training log already save at our_output/ folder.

Replicating Results of Enhanced Loss Function on Aprec repo.

We did not fork the Aprec repo at this repo. But we provide code and tutorials to implement our enhanced loss in Aprec repo.

Note

If you have any question please leave message at ISSUE.

Cite

If you find the our codes and datasets useful for your research or development, please cite our paper:

@misc{li2023improving,
      title={Improving Sequential Recommendation Models with an Enhanced Loss Function},
      author={Fangyu Li and Shenbao Yu and Feng Zeng and Fang Yang},
      year={2023},
      eprint={2301.00979},
      archivePrefix={arXiv},
      primaryClass={cs.IR}
}

About

Implement our enhanced loss base on CIKM2020-S3Rec repo, and provide tutorials to implement enhanced loss at Aprec repo.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 99.8%
  • Shell 0.2%