Automated feature construction for multiple time series data
Project description
TSFuse
Python package for automatically constructing features from multi-view time series data.
Installation
TSFuse requires Python 3 and the following packages:
- Cython>=0.28.5
- numpy>=1.16.1
These packages can be installed using pip:
pip install "cython>=0.28.5" "numpy>=1.16.1"
To install the latest unreleased version of TSFuse from GitHub:
pip install git+https://github.com/arnedb/tsfuse#egg=tsfuse
Documentation
The documentation is available on https://arnedb.github.io/tsfuse/
Examples on how to use TSFuse are shown in the getting started page and the synthetic sine waves demo notebook.
Paper
To learn more about TSFuse's feature construction method, read the following paper:
Arne De Brabandere, Pieter Robberechts, Tim Op De Beéck and Jesse Davis. Automating Feature Construction for Multi-View Time Series Data. ECML/PKDD Workshop on Automating Data Science 2019.
Project details
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.