tslearn-team/tslearn

The machine learning toolkit for time series analysis in Python

What it solves

It provides a comprehensive toolkit for performing machine learning on time series data, simplifying the process of formatting, preprocessing, and analyzing sequential data that often requires specialized distance metrics and algorithms.

How it works

The library expects time series datasets to be formatted as 3D numpy arrays (number of series, number of measurements, number of dimensions). It integrates seamlessly with scikit-learn, following the same API for model training and hyper-parameter tuning. It provides specialized utilities for handling variable-length time series and preprocessing steps like scaling and resampling.

Who it’s for

Data scientists and researchers who need to perform classification, clustering, and regression on time series data using Python.

Highlights

  • Scikit-learn Compatibility: Fully compatible with scikit-learn pipelines and utilities.
  • Diverse Task Support: Includes tools for classification (KNN, SVC), clustering (K-Means, KShape), and regression (SVR, MLP).
  • Specialized Metrics: Implements advanced distance metrics such as Dynamic Time Warping (DTW) and Global Alignment Kernel (GAK).
  • Data Utilities: Built-in support for UCR datasets, synthetic data generators, and conversion tools for other Python toolkits.

Related

  • Project
    aeon-toolkit/aeon

    A scikit-learn compatible Python library for time series machine learning, providing state-of-the-art algorithms for classification, forecasting, and clustering.

  • Project
    nilearn/nilearn

    Nilearn is a Python library that combines neuro‑imaging data handling with scikit‑learn machine‑learning tools, enabling statistical and predictive analyses of brain volumes and surfaces. Install with `pip install nilearn`; community support includes weekly drop‑in hours and a Discord server.

  • Project
    sktime/sktime

    A Python library providing a unified interface for time series analysis, including forecasting, classification, and clustering, compatible with scikit-learn.

  • Project
    Nixtla/mlforecast

    A scalable framework for time series forecasting that allows any scikit-learn compatible machine learning model to be used for forecasting millions of series.

  • Project
    fraunhoferportugal/tsfel

    A Python library for time series analysis that provides a centralized set of over 65 features across statistical, temporal, spectral, and fractal domains for machine learning preparation.