jeongwhanchoi / one-shot-optimization

This is the repository of our accepted CIKM 2022 paper "Prediction-based One-shot Dynamic Parking Pricing".

Prediction-based One-shot Dynamic Parking Pricing

Introduction

This is the repository of our accepted CIKM 2022 paper "Prediction-based One-shot Dynamic Parking Pricing". Paper is available on arxiv. You can also download data here.

Citation

If you find this code useful, you may cite us as:

@article{hong2022prediction,
  title={Prediction-based One-shot Dynamic Parking Pricing},
  author={Hong, Seoyoung and Shin, Heejoo and Choi, Jeongwhan and Park, Noseong},
  journal={arXiv preprint arXiv:2208.14231},
  year={2022}
}

Setup an environment

$ conda env create -f requirements.yaml 

Usage

i) train parking occupancy rate prediction model

- Run run_*.sh to train the prediction model or just pass as we uploaded the pre-trained model.

ii) optimize parking price with pre-trained prediction model

- Run optimize_*.sh to optimize the price with the pre-trained prediction model.