hyperparameter-tuning

A minimal framework for running hyperparameter tuning


Keywords
ml, optimization, hyperparameter, tuning, fairness, easy-to-use, hyperparameter-optimization, hyperparameter-tuning, machine-learning, optuna
License
MIT
Install
pip install hyperparameter-tuning==0.3.1

Documentation

hpt

This repository is under construction 🚧

badge for tests status badge for PyPI publishing status

A minimal hyperparameter tuning framework to help you train hundreds of models.

It's essentially a set of helpful wrappers over optuna.

Install

pip install hpt

Getting started

from hpt.tuner import ObjectiveFunction, OptunaTuner

obj_func = ObjectiveFunction(
    X_train, y_train, X_test, y_test,
    hyperparameter_space=HYPERPARAM_SPACE_PATH,
    eval_metric='accuracy',
    s_train=s_train,
    s_val=s_test,
    threshold=0.50,
)

tuner = OptunaTuner(obj_func) # NOTE: can pass other useful study kwargs here (e.g. storage)

TODO: finish readme.