sktime
A unified framework for machine learning with time series
What is sktime?
A unified framework for machine learning with time series
Overview of sktime
What it is
sktime is a Python library for time series analysis. It provides a unified interface for multiple time series learning tasks, including forecasting, time series classification, clustering, anomaly/changepoint detection, and other tasks. It includes time series algorithms and scikit-learn compatible tools to build, tune, and validate time series models.
Key features
- Unified interface for distinct but related time series learning tasks.
- Dedicated time series algorithms.
- Tools for composite model building: pipelining, ensembling, tuning, and reduction.
- Interfaces to related libraries such as scikit-learn, statsmodels, tsfresh, PyOD, and fbprophet.
- Modules for forecasting, classification, regression, transformations, detection, parameter fitting, clustering, distances/kernels, alignment, splitters, and distributions/simulation (varying stability levels).
Use cases
- Forecasting time series data.
- Time series classification.
- Time series clustering.
- Anomaly and changepoint detection.
- Time series regression.
- Transformation and parameter fitting.
- Time series alignment and distance computation.
Getting started
Install sktime via pip or conda:
pip install sktime
or with all dependencies:
pip install sktime[all_extras]
For curated dependency sets:
pip install sktime[forecasting]
pip install sktime[forecasting,transformations]
Via conda:
conda install -c conda-forge sktime
or with all extras:
conda install -c conda-forge sktime-all-extras
See the repository README for quickstart code examples.
License and pricing
sktime is open source under the BSD-3-Clause license. No pricing information is provided.
Based on the project's public README and repository metadata.