for testing and deploying your application
for finding and fixing issues
for empowering human code reviews
# Author: Simon Blanke
# Email: [email protected]
# License: MIT License
import pytest
import numpy as np
from gradient_free_optimizers import BayesianOptimizer
def parabola_function(para):
loss = para["x"] * para["x"] + para["y"] * para["y"]
return -loss
search_space = {
"x": np.arange(-1, 1, 1),
"y": np.arange(-1, 1, 1),
}
def test_replacement_0():
opt = BayesianOptimizer(search_space, replacement=True)
opt.search(parabola_function, n_iter=15)
with pytest.raises(ValueError):
opt = BayesianOptimizer(search_space, replacement=False)
SimonBlanke /
Gradient-Free-Optimizers
