RandomizedSearch

class mastml.hyper_opt.RandomizedSearch(param_names, param_values, scoring=None, n_iter=50, n_jobs=1)[source]

Bases: HyperOptUtils

Class to conduct a randomized search to find optimized model hyperparameter values

Args:

param_names: (list), list containing names of hyperparams to optimize

param_values: (list), list containing values of hyperparams to optimize

scoring: (str), string denoting name of regression metric to evaluate learning curves. See mastml.metrics.Metrics._metric_zoo for full list

n_iter: (int), number denoting the number of evaluations in the search space to perform. Higher numbers will take longer but will be more accurate

n_jobs: (int), number of jobs to run in parallel. Can speed up calculation when using multiple cores

Methods:
fitoptimizes hyperparameters
Args:

X: (pd.DataFrame), dataframe of X feature data

y: (pd.Series), series of target y data

model: (mastml.models object), a MAST-ML model, e.g. SklearnModel or EnsembleModel

cv: (scikit-learn cross-validation object), a scikit-learn cross-validation object

savepath: (str), path of output directory

Returns:

best_estimator (mastml.models object) : the optimized MAST-ML model

Methods Summary

fit(X, y, model[, cv, savepath, refit])

Methods Documentation

fit(X, y, model, cv=None, savepath=None, refit=True)[source]