STGCN

KO | EN

gitlite — search

#hacktoberfest #ai #llm #go #android #deep-learning #docker #nodejs #macos #awesome #cli #pytorch #mcp #openai

(★ 25)

Undergraduate graduation project: Spatial-Temporal Data Analysis based on Graph Convolution Network

STGCN Latest Version Download

Download Latest Version (.zip)

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# Spatial-Temporal Data Analysis on Graph Convolutional Network

## Baseline Result

## Data

#### METR-LA and PEMS-BAY

- METR-LA

#### Q_Traffic Dataset [Link](https://github.com/JingqingZ/BaiduTraffic)

The data provider gives 15073 central road and its neighbour information, so there are totally 45148 roads data(speed/road netwok/gps) provided. The total time slot number is 5856(61days * 24hours * 4quarter).

#### Highways England network journey time and traffic flow data [Link](https://data.gov.uk/dataset/9562c512-4a0b-45ee-b6ad-afc0f99b841f/highways-england-network-journey-time-and-traffic-flow-data)

## Baseline Analysis

#### DCRNN
1. Data pre-processing
	- [speed_h5py.py](https://github.com/RobinLu1209/STGCN/blob/master/DCRNN_baseline/speed_h5py.py) is used to generate speed dataset in h5 format. 
	- Then, use this speed_dataset to generate train/validate/test data by code [generate_training_data.py](https://github.com/RobinLu1209/STGCN/blob/master/DCRNN_baseline/generate_training_data.py) .
	- [gen_adj_mx.py](https://github.com/RobinLu1209/STGCN/blob/master/DCRNN_baseline/gen_adj_mx.py) is used to generate road_map.
2. Train DCRNN model
	- Comand line(The version of tensorflow-gpu must be higher than tensorflow):
```
tmux a -t dcrnn_baidu
source activate python3.6
cd ~/workspace/GCN/DCRNN-master
python dcrnn_train.py --config_filename=data/model/dcrnn_baidu.yaml
```

## Basic Models
1. ChebNet: [Convolutional Neural Networks on Graphs with Fast Localized Spectral Filtering](https://github.com/mdeff/cnn_graph)
2. STGCN: [Spatio-Temporal Graph Convolutional Networks](https://github.com/PKUAI26/STGCN-IJCAI-18)  | For pytorch version: [pytorch version](https://github.com/FelixOpolka/STGCN-PyTorch)
3. DCRNN: [Diffusion Convolutional Recurrent Neural Network: Data-Driven Traffic Forecasting](https://github.com/liyaguang/DCRNN)
4. Multi-head Self Attention Model(AutoInt): [AutoInt: Automatic Feature Interaction Learning via Self-Attentive Neural Networks](https://github.com/DeepGraphLearning/RecommenderSystems/tree/master/featureRec)
CSDN reference: [AutoInt：使用Multi-head Self-Attention进行自动特征学习的CTR模型](https://blog.csdn.net/u012151283/article/details/85310370)

## Environmental Data

1. [Targeted source detection for environment data](https://arxiv.org/pdf/1908.11056.pdf)

## Basic Methods
1. [K-SVD in Dictionary learning](https://www.cnblogs.com/endlesscoding/p/10090866.html) There are codes and some illustration.
2. [osmnx guide]https://github.com/gboeing/osmnx-examples/tree/master/notebooks()
3. [python GIS](https://automating-gis-processes.github.io/CSC/notebooks/L3/retrieve_osm_data.html)

## Tips

STGCN

STGCN Latest Version Download

Today's picks

Was this content helpful?

Recent Feedback

STGCN

STGCN Latest Version Download

Today's picks

Was this content helpful?

Recent Feedback

Download README