LargeST
收藏资源简介:
LargeST数据集是由新加坡国立大学团队开发的大规模交通预测基准数据集,包含加利福尼亚州8600个传感器在五年时间内的详细交通数据和全面元数据。数据集旨在解决现有公共数据集在规模、时间覆盖和元数据丰富性方面的不足,以支持更准确、高效的交通预测模型开发。通过深入的数据分析和基准测试,LargeST为研究者提供了研究长期交通模式和模型可扩展性的平台,同时也揭示了未来研究中的挑战和机遇。
The LargeST dataset is a large-scale traffic forecasting benchmark dataset developed by a team from the National University of Singapore. It contains detailed traffic data and comprehensive metadata from 8,600 sensors across California over a five-year period. This dataset aims to address the shortcomings of existing public datasets in terms of scale, temporal coverage, and metadata richness, to support the development of more accurate and efficient traffic forecasting models. Through in-depth data analysis and benchmark testing, LargeST provides researchers with a platform for studying long-term traffic patterns and model scalability, while also revealing challenges and opportunities for future research.

- 1LargeST: A Benchmark Dataset for Large-Scale Traffic Forecasting新加坡国立大学 · 2023年



