Skip to main content

Python pacakge to flatten and fold parameter data structures.

Project description

https://travis-ci.org/rgiordan/paragami.svg?branch=master https://codecov.io/gh/rgiordan/paragami/branch/master/graph/badge.svg

Description.

This is a library (very much still in development) intended to make sensitivity analysis easier for optimization problems. The core functionality consists of tools for “folding” and “flattening” collections of parameters – i.e., for converting data structures of constrained parameters to and from vectors of unconstrained parameters.

For background and motivation, see the following papers:

Covariances, Robustness, and Variational Bayes
Ryan Giordano, Tamara Broderick, Michael I. Jordan

A Swiss Army Infinitesimal Jackknife
Ryan Giordano, Will Stephenson, Runjing Liu, Michael I. Jordan, Tamara Broderick

Evaluating Sensitivity to the Stick Breaking Prior in Bayesian Nonparametrics
Runjing Liu, Ryan Giordano, Michael I. Jordan, Tamara Broderick

Using the package.

We welcome new users! However, please be aware that the package is still in development. We encourage users to contact the author (github user rgiordan) for advice, bugs, or if you’re using the package for something important.

Installation.

The package is not (yet) registered on pypi. To install, run the following command, pointing to the root of the git repo:

sudo -H pip3 install --user -e paragami.

Documentation and Examples.

For API documentation, see readthedocs.

Alternatively, check out the repo and run make html in docs/.

For motivating and expository Jupyter notebooks, see docs/example_notebooks/. A good place to start is docs/example_notebooks/front_page_example.ipynb.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

paragami-0.1.tar.gz (45.4 kB view hashes)

Uploaded Source

Built Distribution

paragami-0.1-py3-none-any.whl (22.2 kB view hashes)

Uploaded Python 3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page