This is an open-source library for deep learning researchers, especially for spatial-temporal forecasting.
We provide a neat code base to evaluate advanced spatial-temporal models or develop your model.
| Model | Journals/Conferences | Year |
|---|---|---|
| STGCN | IJCAI | 2018 |
| DCRNN | ICLR | 2018 |
| ASTGCN | AAAI | 2019 |
| T-GCN | IEEE Transactions on Intelligent Transportation Systems | 2019 |
| GWNET | IJCAI | 2019 |
| MSTGCN | AAAI | 2019 |
| AGCRN | NIPS | 2020 |
| GMAN | AAAI | 2020 |
| DFKN | SIGSPATIAL | 2020 |
| STTNS | Arxiv | 2020 |
| HGCN | AAAI | 2021 |
| SANN | Neural Computing and Applications | 2021 |
| ST-Norm | CIKM | 2021 |
| STGODE | SIGKDD | 2021 |
| DGCN | IEEE Transactions on Intelligent Transportation Systems | 2022 |
| D2STGNN | VLDB | 2022 |
| STID | CIKM | 2022 |
| PDFormer | AAAI | 2023 |
| AFDGCN | ECAI | 2023 |
| STWave | ICDE | 2023 |
| MegaCRN | AAAI | 2023 |
| PGCN | IEEE Transactions on Intelligent Transportation Systems | 2024 |
| PMC-GCN | IEEE Transactions on Intelligent Transportation Systems | 2024 |
| STIDGCN | IEEE Transactions on Intelligent Transportation Systems | 2024 |
| TESTAM | ICLR | 2024 |
| WAVGCRN | Arxiv | 2024 |
📝Install dependecies [Back to Top]
Install the required packages
pip install -r requirements.txt👉Data Preparation[Back to Top]
The Los Angeles traffic speed files (METR-LA) and the Bay Area traffic speed files (PEMS-BAY), as well as the Los Angeles traffic flow files (PEMS04 and PEMS08), can be accessed and downloaded from Google Drive. Please place these files in the datasets/ folder. The tree structure of the files is as follows:
\datasets
├─METR-LA
│
├─PEMS04
│
├─PEMS08
│
├─PeMS-Bay
│
└─cache
🚀Run Experiment[Back to Top]
We have provided all the experimental scripts for the benchmarks in the ./scripts folder, which covers all the benchmarking experiments. To reproduce the results, you can run the following shell code.
./scripts/METR-LA.sh
./scripts/PEMS04.sh
./scripts/PEMS08.sh
./scripts/PeMS-Bay.shFor any questions or feedback, feel free to contact Liwei Deng or Junhao Tan.
If you find this code useful in your research or applications, please consider giving our repositories a star.
We express our gratitude to the following members for their contributions to the project, completed under the guidance of Professor Hao Wang: