Skip to content
A Python package to preprocess time series
Python
Branch: master
Clone or download

Latest commit

Latest commit 08695c2 Aug 24, 2017

Files

Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
docs Initial commit Jul 19, 2017
tests
tspreprocess Replace "time" with "bin" Aug 24, 2017
.coveragerc Initial commit Jul 19, 2017
.gitignore added dsstore to .gitignore Jul 20, 2017
AUTHORS.rst
CHANGES.rst Initial commit Jul 19, 2017
LICENSE.txt added MIT licence Jul 20, 2017
README.rst Update README.rst Jul 27, 2017
requirements.txt fixed tsfresh version to 0.9.0 Aug 7, 2017
setup.cfg Initial commit Jul 19, 2017
setup.py Initial commit Jul 19, 2017
test-requirements.txt Initial commit Jul 19, 2017

README.rst

Disclaimer: This package is WIP. Do not take any APIs for granted.

tspreprocess

Time series can contain noise, may be sampled under a non fitting rate or just need to be compressed. tspreprocess is a library for such preprocessing tasks. It contains tools to transform and clean time series data for better analyses.

In detail, we are planning to add methods to do

  • Denoising
  • Compression
  • Resampling
  • ...

Our goal is to make this the most comprehensive time series preprocessing library.

Installation

Clone the repo, cd into it and install it with pip locally

git clone https://github.com/MaxBenChrist/tspreprocess
cd tspreprocess
pip install -e .

You can run the test suite by

python setup.py test

Relation to tsfresh

This package will based on the data formats from the python feature extraction pacakge tsfresh (https://github.com/blue-yonder/tsfresh), allowing a seamless integration between both packages.

You can’t perform that action at this time.