https://github.com/midas-research/sthgcn-icdm
Code for Spatiotemporal Hypergraph Convolution Network for Stock Movement Forecasting
https://github.com/midas-research/sthgcn-icdm
hypergraph spatio-temporal
Last synced: 10 months ago
JSON representation
Code for Spatiotemporal Hypergraph Convolution Network for Stock Movement Forecasting
- Host: GitHub
- URL: https://github.com/midas-research/sthgcn-icdm
- Owner: midas-research
- Created: 2021-02-07T14:03:41.000Z (over 5 years ago)
- Default Branch: main
- Last Pushed: 2021-06-13T15:41:25.000Z (about 5 years ago)
- Last Synced: 2025-04-23T08:11:25.958Z (over 1 year ago)
- Topics: hypergraph, spatio-temporal
- Language: Python
- Homepage:
- Size: 14.6 KB
- Stars: 41
- Watchers: 8
- Forks: 5
- Open Issues: 3
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Spatiotemporal Hypergraph Convolution Network for Stock Movement Forecasting
This codebase contains the python scripts for STHGCN, the model for the ICDM 2020 paper [link](https://ieeexplore.ieee.org/document/9338303).
## Environment & Installation Steps
Python 3.6, Pytorch, Pytorch-Geometric and networkx.
## Dataset and Preprocessing
Download the dataset and follow preprocessing steps from [here](https://github.com/dmis-lab/hats).
```bash
bash download.sh
```
## Run
Execute the following python command to train STHGCN:
```bash
make test_phase=1 save_dir=save
```
test_phase : phase that you want to test
## Cite
Consider citing our work if you use our codebase
```c
@INPROCEEDINGS{9338303, author={Sawhney, Ramit and Agarwal, Shivam and Wadhwa, Arnav and Shah, Rajiv Ratn}, booktitle={2020 IEEE International Conference on Data Mining (ICDM)}, title={Spatiotemporal Hypergraph Convolution Network for Stock Movement Forecasting}, year={2020}, volume={}, number={}, pages={482-491}, doi={10.1109/ICDM50108.2020.00057}}
```