<h1 align="center">
<table border="0">
<td>
<img src="https://raw.githubusercontent.com/numbbo/coco/0ea5f5784c5fa0543261d9c104b490d2d95566f9/logo/coco-pp-300.webp">
</td>
<td>
COmparing Continuous Optimisers (COCO) Post-Processing
</td>
</table>
</h1>
[][paper]
The ([`cocopp`](https://numbbo.github.io/coco-doc/apidocs/cocopp/cocopp.html)) package uses data generated with the [COCO framework](https://numbbo.it) (comparing not only continuous optimisers) and produces output figures and tables in `html` format and for inclusion into `LaTeX` documents. The main documentation page can be found at [getting-started](https://numbbo.it/getting-started#postprocess) and in the [API documentation](https://numbbo.github.io/coco-doc/apidocs/cocopp), but see also [here](https://numbbo.it).
## Installation
To install the latest release from [PyPI](https://pypi.org/project/cocopp):
pip install cocopp
To install the current main branch:
git clone https://github.com/numbbo/coco-postprocess.git
cd coco-postprocess
pip install .
## Usage
The main method of the [`cocopp`](https://numbbo.github.io/coco-doc/apidocs/cocopp/cocopp.html) package is [`main`](https://numbbo.github.io/coco-doc/apidocs/cocopp/cocopp.rungeneric.html#main) (currently aliased to [`cocopp.rungeneric.main`](https://numbbo.github.io/coco-doc/apidocs/cocopp/cocopp.rungeneric.html#main)). The [`main`](https://numbbo.github.io/coco-doc/apidocs/cocopp/cocopp.rungeneric.html#main) method also allows basic use of the post-processing through a shell command-line interface. The recommended use is however from an IPython/Jupyter shell or notebook:
<pre class="py-doctest"><span class="py-prompt">>>></span> <span class="py-keyword">import</span> cocopp
<span class="py-prompt">>>></span> cocopp.main(<span class="py-string">'exdata/my_output another_folder yet_another_or_not'</span>) <span class="py-comment"></span></pre>
postprocesses data from one or several folders, for example data generated with the help from the [`cocoex`](https://numbbo.github.io/coco-doc/apidocs/cocoex) module. Each folder should contain data of a full experiment with a single algorithm. (Within the folder the data can be distributed over subfolders). Results can be explored from the <tt class="rst-docutils literal">ppdata/index.html</tt> file, unless a a different output folder is specified with the <tt class="rst-docutils literal"><span class="pre">-o</span></tt> option. **Comparative data** from over 200 full experiments are archived online and can be listed, filtered, and retrieved from [archives](https://numbbo.github.io/coco-doc/apidocs/cocopp/cocopp.archiving.COCODataArchive.html) which are attributes of [`cocopp.archives`](https://numbbo.github.io/coco-doc/apidocs/cocopp/cocopp.archiving.OfficialArchives.html) and processed alone or together with local data. For example
<pre class="py-doctest"><span class="py-prompt">>>></span> cocopp.archives.bbob(<span class="py-string">'bfgs'</span>) <span class="py-comment"></span>
<span class="py-output">['2009/BFGS_...</span></pre>
lists all data sets run on the `bbob` testbed containing <tt class="rst-docutils literal">'bfgs'</tt> in their name. The first in the list can be postprocessed by
<pre class="py-doctest"><span class="py-prompt">>>></span> cocopp.main(<span class="py-string">'bfgs!'</span>) <span class="py-comment"></span></pre>
All of them can be processed like
<pre class="py-doctest"><span class="py-prompt">>>></span> cocopp.main(<span class="py-string">'bfgs*'</span>) <span class="py-comment"></span></pre>
Only a trailing `*` is accepted and any string containing the substring is matched. The postprocessing result of
<pre class="py-doctest"><span class="py-prompt">>>></span> cocopp.main(<span class="py-string">'bbob/2009/*'</span>) <span class="py-comment"></span></pre>
can be browsed at [https://numbbo.github.io/ppdata-archive/bbob/2009](https://numbbo.github.io/ppdata-archive/bbob/2009). To display algorithms in the background, the [`cocopp.genericsettings.background`](https://numbbo.github.io/coco-doc/apidocs/cocopp/cocopp.genericsettings.html) variable needs to be set:
<pre class="py-doctest"><span class="py-prompt">>>></span> cocopp.genericsettings.background = {<span class="py-builtin">None</span>: cocopp.archives.bbob.get_all(<span class="py-string">'bfgs'</span>)} <span class="py-comment"></span></pre>
where [`None`](http://docs.python.org/library/constants.html#None) invokes the default color (grey) and line style (solid) [`cocopp.genericsettings.background_default_style`](https://numbbo.github.io/coco-doc/apidocs/cocopp/cocopp.genericsettings.html). Now we could compare our own data with the first <tt class="rst-docutils literal">'bfgs'</tt>-matching archived algorithm where all other archived BFGS data are shown in the background with the command
<pre class="py-doctest"><span class="py-prompt">>>></span> cocopp.main(<span class="py-string">'exdata/my_output bfgs!'</span>) <span class="py-comment"></span></pre>
[paper]: https://doi.org/10.5281/zenodo.2594848
Raw data
{
"_id": null,
"home_page": null,
"name": "cocopp",
"maintainer": null,
"docs_url": null,
"requires_python": ">=3.9",
"maintainer_email": null,
"keywords": null,
"author": "Umut Batu, Dimo Brockhoff, Paul Dufoss\u00e9, Tobias Glasmachers, Nikolaus Hansen, Filip Matzner, Olaf Mersmann, Raymond Ros, Dejan Tu\u0161ar",
"author_email": null,
"download_url": "https://files.pythonhosted.org/packages/01/cf/706122cc19a13737b3b4f29bbe5fc7a4fea4bc4e11deb0c82672f399a099/cocopp-2.7.1.tar.gz",
"platform": null,
"description": "<h1 align=\"center\">\n <table border=\"0\">\n <td>\n <img src=\"https://raw.githubusercontent.com/numbbo/coco/0ea5f5784c5fa0543261d9c104b490d2d95566f9/logo/coco-pp-300.webp\"> \n </td>\n <td>\n COmparing Continuous Optimisers (COCO) Post-Processing \n</td>\n</table>\n</h1>\n\n[][paper] \nThe ([`cocopp`](https://numbbo.github.io/coco-doc/apidocs/cocopp/cocopp.html)) package uses data generated with the [COCO framework](https://numbbo.it) (comparing not only continuous optimisers) and produces output figures and tables in `html` format and for inclusion into `LaTeX` documents. The main documentation page can be found at [getting-started](https://numbbo.it/getting-started#postprocess) and in the [API documentation](https://numbbo.github.io/coco-doc/apidocs/cocopp), but see also [here](https://numbbo.it).\n\n## Installation\nTo install the latest release from [PyPI](https://pypi.org/project/cocopp):\n\n pip install cocopp\n\nTo install the current main branch:\n\n git clone https://github.com/numbbo/coco-postprocess.git\n cd coco-postprocess\n pip install .\n\n## Usage\n\nThe main method of the [`cocopp`](https://numbbo.github.io/coco-doc/apidocs/cocopp/cocopp.html) package is [`main`](https://numbbo.github.io/coco-doc/apidocs/cocopp/cocopp.rungeneric.html#main) (currently aliased to [`cocopp.rungeneric.main`](https://numbbo.github.io/coco-doc/apidocs/cocopp/cocopp.rungeneric.html#main)). The [`main`](https://numbbo.github.io/coco-doc/apidocs/cocopp/cocopp.rungeneric.html#main) method also allows basic use of the post-processing through a shell command-line interface. The recommended use is however from an IPython/Jupyter shell or notebook:\n\n<pre class=\"py-doctest\"><span class=\"py-prompt\">>>></span> <span class=\"py-keyword\">import</span> cocopp\n<span class=\"py-prompt\">>>></span> cocopp.main(<span class=\"py-string\">'exdata/my_output another_folder yet_another_or_not'</span>) <span class=\"py-comment\"></span></pre>\n\npostprocesses data from one or several folders, for example data generated with the help from the [`cocoex`](https://numbbo.github.io/coco-doc/apidocs/cocoex) module. Each folder should contain data of a full experiment with a single algorithm. (Within the folder the data can be distributed over subfolders). Results can be explored from the <tt class=\"rst-docutils literal\">ppdata/index.html</tt> file, unless a a different output folder is specified with the <tt class=\"rst-docutils literal\"><span class=\"pre\">-o</span></tt> option. **Comparative data** from over 200 full experiments are archived online and can be listed, filtered, and retrieved from [archives](https://numbbo.github.io/coco-doc/apidocs/cocopp/cocopp.archiving.COCODataArchive.html) which are attributes of [`cocopp.archives`](https://numbbo.github.io/coco-doc/apidocs/cocopp/cocopp.archiving.OfficialArchives.html) and processed alone or together with local data. For example\n\n<pre class=\"py-doctest\"><span class=\"py-prompt\">>>></span> cocopp.archives.bbob(<span class=\"py-string\">'bfgs'</span>) <span class=\"py-comment\"></span>\n<span class=\"py-output\">['2009/BFGS_...</span></pre>\n\nlists all data sets run on the `bbob` testbed containing <tt class=\"rst-docutils literal\">'bfgs'</tt> in their name. The first in the list can be postprocessed by\n\n<pre class=\"py-doctest\"><span class=\"py-prompt\">>>></span> cocopp.main(<span class=\"py-string\">'bfgs!'</span>) <span class=\"py-comment\"></span></pre>\n\nAll of them can be processed like\n\n<pre class=\"py-doctest\"><span class=\"py-prompt\">>>></span> cocopp.main(<span class=\"py-string\">'bfgs*'</span>) <span class=\"py-comment\"></span></pre>\n\nOnly a trailing `*` is accepted and any string containing the substring is matched. The postprocessing result of\n\n<pre class=\"py-doctest\"><span class=\"py-prompt\">>>></span> cocopp.main(<span class=\"py-string\">'bbob/2009/*'</span>) <span class=\"py-comment\"></span></pre>\n\ncan be browsed at [https://numbbo.github.io/ppdata-archive/bbob/2009](https://numbbo.github.io/ppdata-archive/bbob/2009). To display algorithms in the background, the [`cocopp.genericsettings.background`](https://numbbo.github.io/coco-doc/apidocs/cocopp/cocopp.genericsettings.html) variable needs to be set:\n\n<pre class=\"py-doctest\"><span class=\"py-prompt\">>>></span> cocopp.genericsettings.background = {<span class=\"py-builtin\">None</span>: cocopp.archives.bbob.get_all(<span class=\"py-string\">'bfgs'</span>)} <span class=\"py-comment\"></span></pre>\n\nwhere [`None`](http://docs.python.org/library/constants.html#None) invokes the default color (grey) and line style (solid) [`cocopp.genericsettings.background_default_style`](https://numbbo.github.io/coco-doc/apidocs/cocopp/cocopp.genericsettings.html). Now we could compare our own data with the first <tt class=\"rst-docutils literal\">'bfgs'</tt>-matching archived algorithm where all other archived BFGS data are shown in the background with the command\n\n<pre class=\"py-doctest\"><span class=\"py-prompt\">>>></span> cocopp.main(<span class=\"py-string\">'exdata/my_output bfgs!'</span>) <span class=\"py-comment\"></span></pre>\n\n[paper]: https://doi.org/10.5281/zenodo.2594848\n\n",
"bugtrack_url": null,
"license": "BSD-3-Clause",
"summary": "Benchmarking framework for all types of black-box optimization algorithms, postprocessing.",
"version": "2.7.1",
"project_urls": {
"Homepage": "https://coco-platform.org",
"Issues": "https://github.com/numbbo/coco-postprocess/issues",
"Repository": "https://github.com/numbbo/coco-postprocess"
},
"split_keywords": [],
"urls": [
{
"comment_text": null,
"digests": {
"blake2b_256": "01cf706122cc19a13737b3b4f29bbe5fc7a4fea4bc4e11deb0c82672f399a099",
"md5": "0af007aa01c19807a890d4504feacf24",
"sha256": "9ba9e908f4038e6de8704543f2fce86f435fd7dcfe18d0ee8770992a34777aac"
},
"downloads": -1,
"filename": "cocopp-2.7.1.tar.gz",
"has_sig": false,
"md5_digest": "0af007aa01c19807a890d4504feacf24",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.9",
"size": 8620934,
"upload_time": "2025-02-06T00:05:00",
"upload_time_iso_8601": "2025-02-06T00:05:00.512014Z",
"url": "https://files.pythonhosted.org/packages/01/cf/706122cc19a13737b3b4f29bbe5fc7a4fea4bc4e11deb0c82672f399a099/cocopp-2.7.1.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2025-02-06 00:05:00",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "numbbo",
"github_project": "coco-postprocess",
"travis_ci": false,
"coveralls": false,
"github_actions": true,
"lcname": "cocopp"
}
JSON
