sktime and skforecast
These are complementary tools: skforecast specializes in univariate and multivariate forecasting with a scikit-learn-compatible API, while sktime provides a broader unified framework for time series tasks (classification, regression, forecasting) that can incorporate skforecast's models as estimators within its pipelines.
About sktime
sktime/sktime
A unified framework for machine learning with time series
Provides dedicated algorithms for forecasting, classification, regression, clustering, and anomaly detection alongside scikit-learn-compatible tools for pipelining and ensembling. The unified API enables task reduction—applying algorithms designed for one time series task to another—while offering adapters to statsmodels, tsfresh, PyOD, and fbprophet for ecosystem interoperability.
About skforecast
skforecast/skforecast
Time series forecasting with machine learning models
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work