winedarksea/AutoTS

Automated Time Series Forecasting

What it solves

AutoTS 旨在快速部署大规模高精度的时间序列预测。它自动化选择最佳预测模型、预处理步骤和集成技术的过程，去除时间序列分析中常见的手动反复试验。

How it works

该库提供多种预测模型（朴素、统计、机器学习和深度学习）以及超过 30 种时间序列专用转换。它使用遗传算法执行 AutoML 特征搜索，自动找到数据集的最佳模型与预处理组合。支持多变量输出、概率预测，并能扩展到数十万条输入序列。

Who it’s for

需要在大规模时间序列数据上部署高精度预测且不想手动调参的 数据科学家 和 分析师。

Highlights

  • Automated Model Selection: Uses genetic algorithms to find the best models and preprocessing steps.
  • Scalability: Capable of handling tens to hundreds of thousands of input series.
  • Flexible Data Formats: Works directly with Pandas Dataframes in both long and wide formats.
  • Ensembling: Features horizontal and mosaic style ensembles to maximize accuracy while maintaining scalability.
  • Probabilistic Forecasting: Supports upper and lower bound forecasts for uncertainty estimation.