Welcome to sktime#

A unified framework for machine learning with time series.

Mission#

Features#

  • unified API for ML/AI with time series, for model building, fitting, application, and validation

  • support for a variety of learning tasks including forecasting, time series classification, regression, clustering.

  • composite model building, including pipelines with transformations, ensembles, tuning, reduction

  • interactive user experience with scikit-learn like interface conventions

Technical specification#

  • In-memory computation of a single machine, no distributed computing

  • Medium-sized data in pandas and NumPy based containers

  • Modular, principled and object-oriented API

  • Using interactive Python interpreter, no command-line interface or graphical user interface

Contents#

Get Started

Get started using sktime quickly.

Documentation

User documentation.

Installation

Installation Guide.

Estimator Overview

Overview of all estimators in sktime.

API Reference

Understand sktime’s API.

Get Involved

Find out how you can contribute.

Developers

Information for developers.

About

Learn more about sktime.