Reproducible Logistic Regression on Fisher's Iris Flower Database

This bundle contains the code I used to generate the tables on the paper:

@inproceedings{bob2012,
  author = {John Doe},
  title = {A Simple Solution to Iris Flower Classification},
  year = {2015},
  month = jun,
  booktitle = {Reproducible Research Conference, Rio de Janeiro, 2015},
  url = {http://example.com/path/to/my/article.pdf},
}

We appreciate your citation in case you use results obtained directly or indirectly via this software package.

Methods are implemented using Bob's Iris Flower database API and the Logistic Regression trainer and Linear Machines available on that framework. Authors of Bob as you cite this paper if you make use of their software:

@inproceedings{bob2012,
  author = {A. Anjos AND L. El Shafey AND R. Wallace AND
  M. G\"unther AND C. McCool AND S. Marcel},
  title = {Bob: a free signal processing and machine learning
  toolbox for researchers},
  year = {2012},
  month = oct,
  booktitle = {20th ACM Conference on Multimedia Systems
  (ACMMM), Nara, Japan},
  publisher = {ACM Press},
  url =
  {http://publications.idiap.ch/downloads/papers/2012/Anjos_Bob_ACMMM12.pdf},
}

Installation

Run:

$ python bootstrap-buildout.py
$ ./bin/buildout

The tests on my paper were executed on a machine running Ubuntu 12.04, with Python versions 2.6, 2.7, 3.3 and 3.4, providing the same results.

Running

I have created a script that can run the source code reproducing all tables from the above paper. Run it like so:

$ ./bin/paper.py

The contents of each table in the paper should be printed one after the other.

Troubleshooting

You can run unit tests I have prepared like this:

$ ./bin/nosetests

In case of problems, please get in touch with me by e-mail.

Licensing

This work is licensed under the GPLv3.