Ecosyste.ms: Awesome

An open API service indexing awesome lists of open source software.

Awesome Lists | Featured Topics | Projects

https://github.com/scikit-optimize/scikit-optimize

Sequential model-based optimization with a `scipy.optimize` interface
https://github.com/scikit-optimize/scikit-optimize

bayesian-optimization bayesopt binder hacktoberfest hyperparameter hyperparameter-optimization hyperparameter-search hyperparameter-tuning machine-learning optimization scientific-computing scientific-visualization scikit-learn sequential-recommendation visualization

Last synced: 4 months ago
JSON representation

Sequential model-based optimization with a `scipy.optimize` interface

Awesome Lists containing this project

README 

        

|Logo|



|pypi| |conda| |Travis Status| |CircleCI Status| |binder| |gitter| |Zenodo DOI|



Scikit-Optimize

===============



Scikit-Optimize, or ``skopt``, is a simple and efficient library to

minimize (very) expensive and noisy black-box functions. It implements

several methods for sequential model-based optimization. ``skopt`` aims

to be accessible and easy to use in many contexts.



The library is built on top of NumPy, SciPy and Scikit-Learn.



We do not perform gradient-based optimization. For gradient-based

optimization algorithms look at

``scipy.optimize``

`here `_.



.. figure:: https://github.com/scikit-optimize/scikit-optimize/blob/master/media/bo-objective.png

   :alt: Approximated objective



Approximated objective function after 50 iterations of ``gp_minimize``.

Plot made using ``skopt.plots.plot_objective``.



Important links

---------------



-  Static documentation - `Static

   documentation `__

-  Example notebooks - can be found in examples_.

-  Issue tracker -

   https://github.com/scikit-optimize/scikit-optimize/issues

-  Releases - https://pypi.python.org/pypi/scikit-optimize



Install

-------



scikit-optimize requires



* Python >= 3.6

* NumPy (>= 1.13.3)

* SciPy (>= 0.19.1)

* joblib (>= 0.11)

* scikit-learn >= 0.20

* matplotlib >= 2.0.0



You can install the latest release with:

::



    pip install scikit-optimize



This installs an essential version of scikit-optimize. To install scikit-optimize

with plotting functionality, you can instead do:

::



    pip install 'scikit-optimize[plots]'



This will install matplotlib along with scikit-optimize.



In addition there is a `conda-forge `_ package

of scikit-optimize:

::



    conda install -c conda-forge scikit-optimize



Using conda-forge is probably the easiest way to install scikit-optimize on

Windows.



Getting started

---------------



Find the minimum of the noisy function ``f(x)`` over the range

``-2 < x < 2`` with ``skopt``:



.. code:: python



    import numpy as np

    from skopt import gp_minimize



    def f(x):

        return (np.sin(5 * x[0]) * (1 - np.tanh(x[0] ** 2)) +

                np.random.randn() * 0.1)



    res = gp_minimize(f, [(-2.0, 2.0)])



For more control over the optimization loop you can use the ``skopt.Optimizer``

class:



.. code:: python



    from skopt import Optimizer



    opt = Optimizer([(-2.0, 2.0)])



    for i in range(20):

        suggested = opt.ask()

        y = f(suggested)

        opt.tell(suggested, y)

        print('iteration:', i, suggested, y)



Read our `introduction to bayesian

optimization `__

and the other examples_.



Development

-----------



The library is still experimental and under heavy development. Checkout

the `next

milestone `__

for the plans for the next release or look at some `easy

issues `__

to get started contributing.



The development version can be installed through:



::



    git clone https://github.com/scikit-optimize/scikit-optimize.git

    cd scikit-optimize

    pip install -e.



Run all tests by executing ``pytest`` in the top level directory.



To only run the subset of tests with short run time, you can use ``pytest -m 'fast_test'`` (``pytest -m 'slow_test'`` is also possible). To exclude all slow running tests try ``pytest -m 'not slow_test'``.



This is implemented using pytest `attributes `__. If a tests runs longer than 1 second, it is marked as slow, else as fast.



All contributors are welcome!



Making a Release

~~~~~~~~~~~~~~~~



The release procedure is almost completely automated. By tagging a new release

travis will build all required packages and push them to PyPI. To make a release

create a new issue and work through the following checklist:



* update the version tag in ``__init__.py``

* update the version tag mentioned in the README

* check if the dependencies in ``setup.py`` are valid or need unpinning

* check that the ``doc/whats_new/v0.X.rst`` is up to date

* did the last build of master succeed?

* create a `new release `__

* ping `conda-forge `__



Before making a release we usually create a release candidate. If the next

release is v0.X then the release candidate should be tagged v0.Xrc1 in

``__init__.py``. Mark a release candidate as a "pre-release"

on GitHub when you tag it.



Commercial support

------------------



Feel free to `get in touch `_ if you need commercial

support or would like to sponsor development. Resources go towards paying

for additional work by seasoned engineers and researchers.



Made possible by

----------------



The scikit-optimize project was made possible with the support of



.. image:: https://avatars1.githubusercontent.com/u/18165687?v=4&s=128

   :alt: Wild Tree Tech

   :target: http://wildtreetech.com



.. image:: https://i.imgur.com/lgxboT5.jpg

    :alt: NYU Center for Data Science

    :target: https://cds.nyu.edu/



.. image:: https://i.imgur.com/V1VSIvj.jpg

    :alt: NSF

    :target: https://www.nsf.gov



.. image:: https://i.imgur.com/3enQ6S8.jpg

    :alt: Northrop Grumman

    :target: http://www.northropgrumman.com/Pages/default.aspx



If your employer allows you to work on scikit-optimize during the day and would like

recognition, feel free to add them to the "Made possible by" list.



.. |pypi| image:: https://img.shields.io/pypi/v/scikit-optimize.svg

   :target: https://pypi.python.org/pypi/scikit-optimize

.. |conda| image:: https://anaconda.org/conda-forge/scikit-optimize/badges/version.svg

   :target: https://anaconda.org/conda-forge/scikit-optimize

.. |Travis Status| image:: https://travis-ci.org/scikit-optimize/scikit-optimize.svg?branch=master

   :target: https://travis-ci.org/scikit-optimize/scikit-optimize

.. |CircleCI Status| image:: https://circleci.com/gh/scikit-optimize/scikit-optimize/tree/master.svg?style=shield&circle-token=:circle-token

   :target: https://circleci.com/gh/scikit-optimize/scikit-optimize

.. |Logo| image:: https://avatars2.githubusercontent.com/u/18578550?v=4&s=80

.. |binder| image:: https://mybinder.org/badge.svg

   :target: https://mybinder.org/v2/gh/scikit-optimize/scikit-optimize/master?filepath=examples

.. |gitter| image:: https://badges.gitter.im/scikit-optimize/scikit-optimize.svg

   :target: https://gitter.im/scikit-optimize/Lobby

.. |Zenodo DOI| image:: https://zenodo.org/badge/54340642.svg

   :target: https://zenodo.org/badge/latestdoi/54340642

.. _examples: https://scikit-optimize.github.io/stable/auto_examples/index.html