LargeST: A Benchmark Dataset for Large-Scale Traffic Forecasting
In this work, we propose LargeST as a new benchmark dataset (see Figure 1), with the goal of facilitating the development of accurate and efficient methods in the context of large-scale traffic forecasting. The distinguishing characteristic of LargeST lies not only in its extensive graph size, encompassing a total of 8,600 sensors in California, but also in its substantial temporal coverage and rich node information – each sensor contains 5 years of data and comprehensive metadata.
LargeST comprises four sub-datasets, each characterized by a different number of sensors. The biggest one is California (CA), including a total number of 8,600 sensors. We also construct three subsets of CA by selecting three representative areas within CA and forming the sub-datasets of Greater Los Angeles (GLA), Greater Bay Area (GBA), and San Diego (SD).
Variants: LargeST
This dataset is used in 1 benchmark:
| Task | Model | Paper | Date |
|---|---|---|---|
| Traffic Prediction | PatchSTG | Efficient Large-Scale Traffic Forecasting with … | 2024-12-13 |
| Traffic Prediction | RPMixer | RPMixer: Shaking Up Time Series … | 2024-02-16 |
| Traffic Prediction | STID | Spatial-Temporal Identity: A Simple yet … | 2022-08-10 |
| Traffic Prediction | STGODE | Spatial-Temporal Graph ODE Networks for … | 2021-06-24 |
| Traffic Prediction | GWNET | Graph WaveNet for Deep Spatial-Temporal … | 2019-05-31 |
Recent papers with results on this dataset: