This repository provides the implementation of benchmark functions for Bayesian optimization. The details of benchmark functions can be found in these notes.
We recommend installing it with virtualenv
.
You can choose one of three installation options.
- Using PyPI repository (for user installation)
To install the released version in PyPI repository, command it.
$ pip install bayeso-benchmarks
- Using source code (for developer installation)
To install bayeso-benchmarks
from source code, command
$ pip install .
in the bayeso-benchmarks
root.
- Using source code (for editable development mode)
To use editable development mode, command
$ pip install -r requirements.txt
$ python setup.py develop
in the bayeso-benchmarks
root.
- Uninstallation
If you would like to uninstall bayeso-benchmarks
, command it.
$ pip uninstall bayeso-benchmarks
Mandatory pacakges are inlcuded in requirements.txt
.
The following requirements
files include the package list, the purpose of which is described as follows.
requirements-dev.txt
: It is for developing thebayeso-benchmarks
package.
A simple example on Branin function is shown below.
from bayeso_benchmarks import Branin
obj_fun = Branin()
bounds = obj_fun.get_bounds()
X = obj_fun.sample_uniform(100)
Y = obj_fun.output(X)
Y_noise = obj_fun.output_gaussian_noise(X)