A unified and flexible and comprehensive traffic prediction library
LibTraffic（阡陌）
LibTraffic is a unified, flexible and comprehensive traffic prediction library, which provides researchers with a credibly experimental tool and a convenient development framework. Our library is implemented based on PyTorch, and includes all the necessary steps or components related to traffic prediction into a systematic pipeline.
LibTraffic currently supports the following tasks:
- Traffic State Prediction
- Traffic Flow Prediction
- Traffic Speed Prediction
- On-Demand Service Prediction
- Trajectory Next-Location Prediction
Features
- Unified: LibTraffic builds a systematic pipeline to implement, use and evaluate traffic prediction models in a unified platform. We design basic spatial-temporal data storage, unified model instantiation interfaces, and standardized evaluation procedure.
- Comprehensive: 42 models covering four traffic prediction tasks have been reproduced to form a comprehensive model warehouse. Meanwhile, LibTraffic collects 29 commonly used