feijoa 0.1.6

Hyperparameter's optimization framework

Feijoa is a Python framework for hyperparameter's optimization.

The Feijoa API is very easy to use, effective for optimizing machine learning algorithms and various software. Feijoa contains many different use cases.

Compatibility

Feijoa works with Linux and OS X. Requires Python 3.8 or later.

Feijoa works with Jupyter notebooks with no additional configuration required.

Installing

Install with pip or your favourite PyPI package manager.

python -m pip install feijoa

Code example

from feijoa import create_job, SearchSpace, Real
from math import sin


def objective(experiment):
    x = experiment.params.get('x')
    y = experiment.params.get('y')

    return sin(x * y)
    
space = SearchSpace()
space.insert(Real('x', low=0.0, high=2.0))
space.insert(Real('y', low=0.0, high=2.0))

job = create_job(search_space=space)
job.do(objective)