Dynamic Spatial-Temporal Model for Carbon Emission Forecasting

This repository contains the implementation of the model Dynamic Spatial-Temporal Graph Convolutional Recurrent Network (DSTGCRN) presented in the manuscript "Dynamic Spatial-Temporal Model for Carbon Emission Forecasting".

Table of Contents

• Description
• Requirements
• Model Training
• Contact

Description

This project introduces a dynamic spatial-temporal modeling approach to forecast carbon emissions. The model leverages spatial correlations and temporal dynamics to provide more accurate predictions. The repository includes the source code and datasets.

Requirements

The following packages are required to run the code provided in this repository:

• Python: 3.8.18
• Hydra: 2.5
• Hydra Core: 1.3.2
• PyTorch: 2.0.1 or above
• NumPy: 1.24.4

These are the core requirements; however, additional packages may be needed as you work through the project. Please install any other necessary packages as required.

Model Training

To train the model, you will need to run the script located at src/model/DSTGCRN/run.py. By default, the script is configured to train on a dataset from China. The relative path src/model/DSTGCRN/run.py is based on the scenario where DSTGCRN is the folder you navigate to as the starting point.

You can also specify different datasets by using the dataset argument when running the script.

python src/model/DSTGCRN/run.py dataset=US
python src/model/DSTGCRN/run.py dataset=EU

Contact

For any inquiries or further discussions, feel free to reach out at [email protected].

Citation

If you find this work useful, please cite the following paper:

@article{gong2024dynamic,
  title={Dynamic spatial-temporal model for carbon emission forecasting},
  author={Gong, Mingze and Zhang, Yongqi and Li, Jia and Chen, Lei},
  journal={Journal of Cleaner Production},
  pages={142581},
  year={2024},
  publisher={Elsevier}
}

This paper is available at https://www.sciencedirect.com/science/article/pii/S0959652624020298