A unified Python framework for machine learning with time series data.
Build, tune, and validate time series models using scikit-learn compatible tools.
sktime is a Python library providing a unified interface for time series analysis tasks like forecasting, classification, and anomaly detection. It suits data scientists and supply chain analysts building predictive models.
It connects distinct time series learning tasks through dedicated algorithms and composite model building tools like pipelining and ensembling. It also integrates with libraries such as statsmodels, tsfresh, PyOD, and fbprophet.
It suits Python developers and data scientists building complex supply chain forecasting or anomaly detection pipelines. It does not suit non-technical users who require point-and-click software interfaces.
Provides a consistent API for multiple forecasting algorithms and models.
Supports pipelining, ensembling, tuning, and reduction for time series tasks.
Connects with scikit-learn, statsmodels, tsfresh, PyOD, and fbprophet.
Includes maturing modules for anomaly and changepoint detection.
Free under the BSD-3-Clause license. The only cost is running it yourself in your own Python environment.
873 open remote AI-training roles, updated November 15, 2026; most pay $20–$170 an hour. Zekai lists them free; the application goes to the employer’s platform.
It supports forecasting, time series classification, regression, transformations, clustering, anomaly detection, and changepoint detection.
Yes, it uses scikit-learn compatible tools and design patterns to build, tune, and validate models.
See Zekai first in Google