johannfaouzi/pyts

Health v2 · maintenance only

Adoption not part of the score

1876 stars · 179 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

pyts is a Python package for time series classification that provides preprocessing, transformation, and utility tools along with implementations of state-of-the-art algorithms. It follows scikit-learn conventions and uses Numba for performance optimization.

Use cases

• classify time series data
• transform time series for machine learning
• run state-of-the-art time series classification algorithms
• preprocess sensor or signal data for classification
• benchmark time series classifiers on UCR datasets
• integrate time series classification into scikit-learn pipelines

When to choose

• you need a scikit-learn-compatible library for time series classification
• you want access to many classic and modern TSC algorithms in one package
• you work in Python with NumPy/SciPy and want easy pip/conda installation
• you need well-documented tools for time series transformations like SAX, PAA, or BOSS

When to avoid

• you need time series forecasting or regression rather than classification
• you need deep-learning-based time series models (consider sktime, tsai, or PyTorch)
• you need streaming or online time series classification
• you work outside the Python ecosystem

Facets

library · maturity stable

machine-learning data-science etl machine-learning data-science time-series python cross-platform time-series-classification scikit-learn-compatible numba signal-processing

Member repositories

For agents

markdown · JSON · MCP: product_card(name="johannfaouzi/pyts")