Skip to main content

Module related to Bayesian approach to inverse problems

Project description

The Bayesian inversion (edwin) package provides algorithm developed during my scientific research work in numerical computation for inverse problems (signal, image processing). Feel free to use them as you want. Any comments and contributions are welcome.

The name edwin is in reference to Edwin T. Jaynes, a great Bayesian Analysis scientific.

Acknowledgements

The use of edwin software package should be explicitly acknowledged in publications in the following form:

  1. an acknowledgment statement: “Some of the results in this paper have been derived using some of the edwin package algorithms From F. Orieux et al. published in citations.

  2. at the first reference, a footnote placed in the main body of the paper referring to the edwin web site, currently http://bitbucket.org/forieux/edwin

The citations are mentioned in documentation, References section of this file and are available in bibtex file.

Info

Contents

improcessing.py

A module that implement the algorithm described in [2] for unsupervised myopic image deconvolution. However the myopic part is not actually available.

inversion.py

A module that implement the algorithm described in [1] and use in [3-4] and other papers. It’s implement an unsupervised general inverse problem algorithm estimation, based on MCMC algorithm.

sampling.py

Implementation of stochastic sampling algorithm, specially [1].

optim.py

A module that implement classical optimisation algorithm for use of other module. They are design for very large system resolution (dim > 1e6).

Requirements

This package depends on my free otb package (utility functions).

  • Numpy version >= 1.4.1

  • otb version >= 0.2.1

Installation

The pip version:

pip install edwin

If you have not pip, download the archive, decompress it and to install in your user path, run in a command line:

python setup.py install --user

or for the system path, run as root:

python setup.py install

Development

This package follow the Semantic Versionning convention http://semver.org/. To get the development version you can clone the mercurial repository available here http://bitbucket.org/forieux/edwin

The ongoing development depends on my research activity but is open. I try to fix bugs.

License

edwin is free software distributed under the MIT license, see LICENSE.txt

References

A bibtex file is provided in the archive.

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

edwin-0.2.0.tar.gz (4.0 kB view details)

Uploaded Source

Built Distribution

edwin-0.2.0-py2-none-any.whl (19.9 kB view details)

Uploaded Python 2

File details

Details for the file edwin-0.2.0.tar.gz.

File metadata

  • Download URL: edwin-0.2.0.tar.gz
  • Upload date:
  • Size: 4.0 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No

File hashes

Hashes for edwin-0.2.0.tar.gz
Algorithm Hash digest
SHA256 96318855268e9c01c8930d07f57e116968e9c50ea2033982ff6bd5c1ac08fb01
MD5 5adf9ff5f13662d112cbb0c1591c4814
BLAKE2b-256 caaf4222d679ef0338997529d8441ab7700b4f0f2cea81ba6a2772682510bc22

See more details on using hashes here.

File details

Details for the file edwin-0.2.0-py2-none-any.whl.

File metadata

File hashes

Hashes for edwin-0.2.0-py2-none-any.whl
Algorithm Hash digest
SHA256 dbbbcff8ce5e11a56305774fb78a55748f58084d6c5bd2dad4d0f3aeefd4317e
MD5 032f8cd9402c1b119fac4337605465b5
BLAKE2b-256 6183c6ffcc2de71954c771381171aa5a5fa27f1a8bdcfd5586012664693b1e40

See more details on using hashes here.

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