The dataset refers to the traffic speed data in San Francisco Bay Area, containing 307 sensors on 29 roads. The time span of the dataset is January-February in 2018. It is a popular benchmark for traffic forecasting.
Variants: PeMSD4
This dataset is used in 1 benchmark:
| Task | Model | Paper | Date |
|---|---|---|---|
| Traffic Prediction | FasterSTS | FasterSTS: A Faster Spatio-Temporal Synchronous … | 2025-01-01 |
| Traffic Prediction | PM-DMNet(R) | Pattern-Matching Dynamic Memory Network for … | 2024-08-12 |
| Traffic Prediction | PM-DMNet(P) | Pattern-Matching Dynamic Memory Network for … | 2024-08-12 |
| Traffic Prediction | HTVGNN | A novel hybrid time-varying graph … | 2024-01-17 |
| Traffic Prediction | STD-MAE | Spatial-Temporal-Decoupled Masked Pre-training for Spatiotemporal … | 2023-12-01 |
| Traffic Prediction | STG-NRDE | Graph Neural Rough Differential Equations … | 2023-03-20 |
| Traffic Prediction | PDFormer | PDFormer: Propagation Delay-Aware Dynamic Long-Range … | 2023-01-19 |
| Traffic Prediction | HAGCN | HAGCN : Network Decentralization Attention … | 2022-09-05 |
| Traffic Prediction | Hierarchical-Attention-LSTM (HierAttnLSTM) | Network Level Spatial Temporal Traffic … | 2022-01-15 |
| Traffic Prediction | STG-NCDE | Graph Neural Controlled Differential Equations … | 2021-12-07 |
Recent papers with results on this dataset: