https://github.com/dfm/pcp

A Python implementation of the Principal Component Pursuit algorithm from arXiv:0912.3599
https://github.com/dfm/pcp

Last synced: about 1 year ago
JSON representation

A Python implementation of the Principal Component Pursuit algorithm from arXiv:0912.3599

• Host: GitHub
• URL: https://github.com/dfm/pcp
• Owner: dfm
• License: mit
• Created: 2015-04-13T20:10:22.000Z (about 11 years ago)
• Default Branch: main
• Last Pushed: 2020-06-12T18:24:35.000Z (almost 6 years ago)
• Last Synced: 2025-03-27T10:21:35.206Z (about 1 year ago)
• Language: Python
• Size: 98.6 KB
• Stars: 64
• Watchers: 6
• Forks: 18
• Open Issues: 1
• Metadata Files:
  • Readme: README.rst
  • License: LICENSE

Awesome Lists containing this project

README

          
Principal Component Pursuit in Python

=====================================

This is a Python implementation of the `Principal Component

Pursuit `_ algorithm for robust PCA.

This implementation uses the `fbpca `_

implementation of approximate partial SVD for speed so you'll need to install

that first.

Usage

-----

TODO

Demo

----

Applied to `the 'Escalator' dataset

`_ (using the code

in the ``demo.py`` script, this algorithm produces a video with frames that

look like:

.. image:: https://raw.githubusercontent.com/dfm/pcp/master/demo.png

Author & License

----------------

Copyright 2015 Daniel Foreman-Mackey

This is open source software written by Dan Foreman-Mackey and released under

the terms of the MIT license (see LICENSE).