facebookresearch/Kats

Kats, a kit to analyze time series data, a lightweight, easy-to-use, generalizable, and extendable framework to perform time series analysis, from understanding the key statistics and characteristics, detecting change points and anomalies, to forecasting future trends.

What it solves

Kats is a comprehensive toolkit designed to simplify the analysis of time series data. It provides a unified framework for common data science tasks such as forecasting future trends, detecting anomalies or change points, and extracting meaningful features from time series, reducing the need to switch between multiple disparate tools.

How it works

Kats provides a lightweight, generalizable framework that integrates various algorithms and models. It includes support for forecasting models like Prophet, a neural network-based global model, and detection algorithms such as CUSUM. The toolkit also includes a self-supervised learning framework for fast and scalable hyper-parameter tuning to optimize forecasting models at scale.

Who it’s for

This toolkit is designed for data scientists and engineers who need to perform time series analysis in industrial settings to understand statistics, detect regressions, and forecast trends.

Highlights

  • One-stop shop: Integrates detection, forecasting, feature extraction, and multivariate analysis in one library.
  • Scalable Tuning: Uses a self-supervised learning framework to automatically optimize hyper-parameters for forecasting models.
  • Diverse Detection: Supports anomaly and changepoint detection with tools like CUSUM, ProphetTrendDetector, and Dynamic Time Warping.
  • Flexible Installation: Offers a minimal installation option to reduce dependency overhead.

Related

  • Project
    unit8co/darts

    A Python library for user-friendly time series forecasting and anomaly detection, offering a unified API for models ranging from ARIMA to deep neural networks.

  • Project
    WenjieDu/PyPOTS

    A Python toolbox for machine learning on partially-observed time series, providing unified APIs for imputation, forecasting, and anomaly detection on data with missing values.

  • Project
    sktime/sktime

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

  • Project
    AIStream-Peelout/flow-forecast

    An open-source deep learning framework for time series forecasting, classification, and anomaly detection featuring a wide array of state-of-the-art models.

  • Project
    LongxingTan/Time-series-prediction

    A TensorFlow-based time series package that provides a unified interface for prediction, classification, and anomaly detection using state-of-the-art deep learning models.