#127127th of 168 in ML frameworks and libraries
tsfresh
Python library that extracts and filters features from time series
- Stars
- 9.5k
- License
- MIT
- Last commit
- Jul 2026
- Last release
- May 2026
- Language
- Jupyter Notebook
9,134 on 21 Apr 20269,491 on 10 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
tsfresh is a Python package that computes hundreds of statistical, signal-processing and nonlinear-dynamics features from time series. It then filters them with a multiple hypothesis-testing procedure (FRESH) that controls the share of irrelevant features for a given regression or classification target. It works with pandas, numpy and scikit-learn, and can run locally or on a cluster.
Who it is for: Data scientists building ML models on time-series or sensor data
Strengths
- Extracts hundreds of features per series, such as peak counts and time reversal symmetry
- Built-in feature selection based on hypothesis tests with multiple-test control
- Compatible with pandas, numpy and scikit-learn pipelines
- Custom feature calculators can be added; documented on Read the Docs
Weaknesses
- Matrixprofile feature calculators require a Python 3.8 environment
- Feature selection is described only against a regression or classification target
- Classical feature engineering for tabular ML, not an LLM or generative AI tool
- Cluster execution details are only in the external docs, not the README
What it needs
- no GPU
- Docker
- Needs pandas, numpy, scikit-learn
Quick answers
Is tsfresh open source?
Yes. Its code is on GitHub under the MIT license.
Can you run tsfresh with Docker?
Yes. Its repository has a Docker setup: a Dockerfile, a Compose file or a published image.
Does tsfresh need a GPU?
No, it runs without one.
Also in ML frameworks and libraries
See all 168| Rank | Project | Score |
|---|---|---|
| 1 | transformersPython library defining open models for text, vision, audio and multimodal tasks | 94 out of 100 |
| 2 | pytorchPython tensor library with GPU support and tape-based autograd | 93 out of 100 |
| 3 | ultralyticsPython library and CLI for training and running YOLO vision models | 93 out of 100 |
| #4 | pytorch-image-modelsPyTorch library of image models, pretrained weights and training scripts | 93 out of 100 |
| #5 | supervisionPython toolkit for loading, annotating and counting computer vision detections | 92 out of 100 |