optimization routines for hyperparameter tuning
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
bin
docs
logo
notebooks
optunity
wrappers typo in error message of auc_roc Oct 13, 2015
.gitignore
.travis.yml
LICENSE.txt
MANIFEST.in
Makefile
README.rst
setup.py exponential search space in first example on main page Jul 23, 2015

README.rst

Optunity

https://travis-ci.org/claesenm/optunity.svg?branch=master Documentation Status

Optunity is a library containing various optimizers for hyperparameter tuning. Hyperparameter tuning is a recurrent problem in many machine learning tasks, both supervised and unsupervised. Tuning examples include optimizing regularization or kernel parameters.

From an optimization point of view, the tuning problem can be considered as follows: the objective function is non-convex, non-differentiable and typically expensive to evaluate.

This package provides several distinct approaches to solve such problems including some helpful facilities such as cross-validation and a plethora of score functions.

The Optunity library is implemented in Python and allows straightforward integration in other machine learning environments, including R and MATLAB.

If you have any comments, suggestions you can get in touch with us at gitter:

Join the chat at https://gitter.im/claesenm/optunity

To get started with Optunity on Linux, issue the following commands:

git clone https://github.com/claesenm/optunity.git
echo "export PYTHONPATH=$PYTHONPATH:$(pwd)/optunity" >> ~/.bashrc

Afterwards, importing optunity should work in Python:

#!/usr/bin/env python
import optunity

Optunity is developed at the STADIUS lab of the dept. of electrical engineering at KU Leuven (ESAT). Optunity is free software, using a BSD license.

For more information, please refer to the following pages: http://www.optunity.net

Contributors

The main contributors to Optunity are:

  • Marc Claesen: framework design & implementation, communication infrastructure, MATLAB wrapper and all solvers.
  • Jaak Simm: R wrapper.
  • Vilen Jumutc: Julia wrapper.