EA-Multi-Head-GCN

Implementation of paper: Building Pattern Recognition by Using an Edge-attention Multi-head Graph Convolutional Network

Github • Figshare • Paper

Setup

conda create -n eamhgcn python=3.10
conda activate eamhgcn
conda install pytorch torchvision torchaudio pytorch-cuda=11.7 -c pytorch -c nvidia
pip install tqdm scikit-learn

Train

python CODE/main/multi_train.py --model GCN_A_H_S --model_num 5

Test

 python CODE/main/test.py --model GCN_A_H_S --model_num 5

Make Dataset[Optional]

If you want to make your own dataset, you can use the following command to make it.

• Install the python packages:

pip install pyshp matplotlib scipy

• Install the package of shapely by your self.
• Run the code CODE/predata/main.py to make the dataset.

Directory Structure

├── [dir]CODE
│   ├── [dir]main
│   │   ├── global config file
│   │   ├── test.py
│   │   └── train.py
│   │   ├── [dir]model
│   │   │   ├── [dir]models
│   │   │   │   ├── config.py
│   │   │   │   ├── graph.py
│   │   │   │   ├── layers.py
│   │   │   │   └── model.py
│   │   ├── [dir]utils
│   │   │   └── global util files
│   ├── predata
│   │   ├── BuildingDataLoader.py
│   │   ├── Skeleton.py
│   │   └── main.py
├── [dir]DATA
│   ├── [dir]dataset
│   └── [dir]split_data #shape file

Citation

If you find this repository useful in your research, please cite our paper:

@article{doi:10.1080/13658816.2024.2427853,
author = {Wang, Haitao and Xu, Yongyang and Hu, Anna and Xie, Xuejing and Chen, Siqiong and Xie, Zhong},
title = {Building pattern recognition by using an edge-attention multi-head graph convolutional network},
journal = {International Journal of Geographical Information Science},
volume = {0},
number = {0},
pages = {1--26},
year = {2024},
publisher = {Taylor \& Francis},
doi = {10.1080/13658816.2024.2427853}}

✨ Happy research! ✨