Skip to main content

MetaOpt is a library that optimizes black-box functions using a limited amount of time and utilizing multiple processors. The main focus of MetaOpt is the parameter tuning for machine learning and heuristic optimization.

Project description

MetaOpt is a library that optimizes black-box functions using a limited amount of time and utilizing multiple processors. The main focus of MetaOpt is the parameter tuning for machine learning and heuristic optimization.

MetaOpt has been developed by the Computational Intelligence Group at the University of Oldenburg. The authors are: Renke Grunwald, Bengt Lüers, Jendrik Poloczek, Justin Heinermann, Oliver Kramer.

For a user guide, see the documentation. The following describes basic operations with the repository.

Status

PyPIn

PyPIn downloads

PyPIn version

PyPIn Download Format

PyPIn License

GitHub

repo size

releases

tags

master

Build Status

Code Health

Coverage Master

develop

Build Status

Code Health

Coverage Develop

Download

MetaOpt is obtainable via archives of past releases, but you can also get the sources by cloning the repository.

To get a working copy of the MetaOpt repository:

$ git clone https://github.com/cigroup-ol/metaopt.git

Installation

MetaOpt is available on PyPI, but you can also install it from source.

To install MetaOpt from PyPI using pip:

$ sudo pip install metaopt

To install MetaOpt from a working copy:

$ cd metaopt
$ sudo python setup.py install

To verify MetaOpt was installed correctly:

$ python metaopt --version
metaopt 0.1.0.0

Tests

MetaOpt has automated online tests, but you can also run them locally.

To run MetaOpt’s test suite from a working copy:

$ cd metaopt
$ sudo pip install -r requirements_test.txt
$ make tests-all

Coverage

MetaOpt has automated online test coverage reports, but you can also create them locally.

To create MetaOpt’s coverage reports from a working copy:

$ cd metaopt
$ sudo pip install -r requirements_coverage.txt
$ make coverage

Examples

MetaOpt comes with examples, which you can view in the docs, but you can also run them locally.

To install MetaOpt’s example dependecies and run an example from a working copy:

$ cd metaopt
$ sudo pip install -r requirements_examples.txt
$ PYTHONPATH=. python examples/svm_saes_global_timeout.py

Documentation

MetaOpt has automatically generated online documentation, but you can build yourself a local copy.

To build MetaOpt’s html documentation from a working copy:

$ cd metaopt
$ sudo pip install -r requirements_docs.txt
$ make docs

0.1.0 – initial release

Download files

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

Source Distribution

metaopt-0.1.0.tar.gz (271.3 kB view hashes)

Uploaded Source

Built Distribution

metaopt-0.1.0-py2.py3-none-any.whl (117.5 kB view hashes)

Uploaded Python 2 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