Project details for pyGPs

Manage

Project List
Login to submit a new project

Details

Version:
1.3.1
Submitter:
mn
Author(s):
Marion Neumann , Daniel Marthaler , Shan Huang , Kristian Kersting
Posted:
1 December 2014
Last Updated:
1 December 2014
Open Source License:
bsd
Programming Language(s):
python

RSS Feed RSS Feed for "pyGPs"

Screenshot pyGPs 1.3.1

by mn - December 1, 2014, 17:36:32 CET [ Project Homepage BibTeX Download ]

view ( today), download ( today ), 0 subscriptions

Description:

pyGPs is a Python project for Gaussian process (GP) regression and classification for machine learning.

pyGPs is an object-oriented implemetation of GP regression and classificaion additionally supporting useful routines for the practical use of GPs, such as cross validation functionalities for evaluation as well as basic routines for iterative restarts for the GP hyperparameter optimization.

Note, there is also a procedural implementation of GPs (pyGP_PR) which follows structure and functionality of the gpml matlab implementaion by Carl Edward Rasmussen and Hannes Nickisch (Copyright (c) by Carl Edward Rasmussen and Hannes Nickisch, 2013-01-21). This version can be downloaded via this link: https://github.com/marionmari/pyGP_PR/archive/v1.1.tar.gz.

Future extensions will be designed for pyGPs. pyGP_PR will be maintained as it is now.

Changes to previous version:

Changelog pyGPs v1.3.1

November 25th 2014

structural updates:

  • full inline documentation with input parameter and output specified

  • check for the inputs and provide diagnostic messages for some inputs

  • consistant naming in inline and online documentation

  • string representation for dnlZStruct and postStruct. Now you can do sth like:

nlZ, dnlZ, post = model.getPosterior(x,y)

print post

  • instead of a python object, we provide now a more informative description.

  • add optimization into unit test routines. Also add checking for cholesky decomposition and checking positive-definite property of kernel matrix.

  • add jitter to the digonal of linear, linARD, and poly covariance for numerical stability.

  • fix several minor problems in unit test framework

  • hierachically rearranged for online documentation

  • add several supplementary instruction in online documentation

BibTeX Entry: Download
Supported Operating Systems: Platform Independent
Data Formats: Numpy
Tags: Classification, Regression, Gaussian Processes
Archive: download here

Comments

No one has posted any comments yet. Perhaps you'd like to be the first?

Leave a comment

You must be logged in to post comments.