Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/dmis-lab/hats
HATS: A Hierarchical Graph Attention Network for Stock Movement Prediction
https://github.com/dmis-lab/hats
Last synced: about 1 month ago
JSON representation
HATS: A Hierarchical Graph Attention Network for Stock Movement Prediction
- Host: GitHub
- URL: https://github.com/dmis-lab/hats
- Owner: dmis-lab
- Created: 2019-11-12T02:23:39.000Z (about 5 years ago)
- Default Branch: master
- Last Pushed: 2019-12-02T02:01:27.000Z (about 5 years ago)
- Last Synced: 2024-05-14T00:23:18.855Z (7 months ago)
- Language: Python
- Size: 239 KB
- Stars: 139
- Watchers: 7
- Forks: 63
- Open Issues: 8
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# HATS
![overview_model](figs/HATS.PNG)
This repository contains source codes of HATS, A Hierarchical Graph Attention Network for Stock Movement Prediction. As we conducted experiments on two different tasks, node classification and graph classification, we provide two different version of codes for each tasks. Please refer to our paper [HATS: A Hierarchical Graph Attention Network for Stock Movement Prediction](https://arxiv.org/abs/1908.07999) for further details.## Requirements
Numpy 1.15.1
Tensorflow 1.11.0## Dataset
Price-realted data and corporate relation data is used for HATS. We gathered both data for S&P 500 listed companies from 2013/02/08 to
2019/06/17 (1174 trading days in total). Price data are gathered from [Yahoo Finance](https://finance.yahoo.com/) and corporate relation data are collected based on the information on [Wikidata](https://www.wikidata.org/wiki/Wikidata:Main_Page). Both datasets can be downloaded with the command below.## Usage
### Download Data
```
bash download.sh
```
### Excute model with makefile
You need to pass some arguments.
test_phase : phase that you want to test
save_dir : name of saving directories
data_type (only in graph_classification) : choose among ['S5CONS', 'S5ENRS', 'S5UTIL', 'S5FINL', 'S5INFT']
```
e.g.
make test_phase=1 save_dir=save data_type='S5CONS'
```