bayesian-optimization

To perform Bayesian optimization, you must define a function to maximize and a dictionary of parameter bounds (pbounds).

1. Define the function: The function f must take a known set of parameters and return a real number.
2. Define bounds: Provide a dictionary where keys are parameter names and values are tuples representing the (min, max) range.
3. Instantiate BayesianOptimization: Pass the function, bounds, and an optional random_state.
4. Call maximize: Use the maximize method to run the optimization process.

Key maximize parameters:

• init_points: Number of initial random exploration steps.
• n_iter: Number of steps of bayesian optimization to perform.

from bayes_opt import BayesianOptimization

# 1. Define the function to be optimized
def black_box_function(x, y):
    return -x ** 2 - (y - 1) ** 2 + 1

# 2. Define the bounded region of parameter space
pbounds = {'x': (2, 4), 'y': (-3, 3)}

# 3. Instantiate the optimizer
optimizer = BayesianOptimization(
    f=black_box_function,
    pbounds=pbounds,
    random_state=1,
)

# 4. Run the optimization
optimizer.maximize(
    init_points=2,
    n_iter=3,
)