bayesian_optimization

Complexity

Total Complexity

4

Size/Duplication

Total Lines	45
Duplicated Lines	0 %

Importance

Changes

0

3 Methods

Rating	Name	Duplication	Size	Complexity
A	GPR.predict()	0	2	1
A	GPR.fit()	0	3	1
A	GPR.__init__()	0	2	1

import GPy
from sklearn.model_selection import cross_val_score
from sklearn.ensemble import GradientBoostingClassifier
from sklearn.datasets import load_breast_cancer
from hyperactive import Hyperactive

data = load_breast_cancer()
X, y = data.data, data.target


def model(para, X, y):
    gbc = GradientBoostingClassifier(
        n_estimators=para["n_estimators"],
        max_depth=para["max_depth"],
        min_samples_split=para["min_samples_split"],
    )
    scores = cross_val_score(gbc, X, y, cv=3)

    return scores.mean()


search_config = {
    model: {
        "n_estimators": range(10, 100, 10),
        "max_depth": range(2, 12),
        "min_samples_split": range(2, 12),
    }
}

class GPR:
    def __init__(self):
        kernel = GPy.kern.RBF(input_dim=1, variance=1., lengthscale=1.)
        
    def fit(self, X, y):
        m = GPy.models.GPRegression(X, y, kernel)

The variable kernel does not seem to be defined.

        m.optimize(messages=True)

    def predict(self, X):
        return m.predict(X)

The variable m does not seem to be defined.

bayes_opt = {"Bayesian": {"gpr": GPR()}}

opt = Hyperactive(X, y)
opt.search(search_config, n_iter=30, optimizer=bayes_opt)