def optimizer_UCB(optimizer: BaseLearner,
X: modALinput,
beta: float = 1) -> np.ndarray:
"""
Upper confidence bound acquisition function for Bayesian optimization.
Args:
optimizer: The :class:`~modAL.models.BayesianOptimizer` object for which the utility is to be calculated.
X: The samples for which the upper confidence bound is to be calculated.
beta: Value controlling the beta parameter.
Returns:
Upper confidence bound utility score.
"""
try:
mean, std = optimizer.predict(X, return_std=True)
std = std.reshape(-1, 1)
except NotFittedError:
mean, std = np.zeros(shape=(X.shape[0], 1)), np.ones(shape=(X.shape[0],
1))
return UCB(mean, std, beta)
def optimizer_EI(optimizer: BaseLearner,
X: modALinput,
tradeoff: float = 0) -> np.ndarray:
"""
Expected improvement acquisition function for Bayesian optimization.
Args:
optimizer: The :class:`~modAL.models.BayesianOptimizer` object for which the utility is to be calculated.
X: The samples for which the expected improvement is to be calculated.
tradeoff: Value controlling the tradeoff parameter.
Returns:
Expected improvement utility score.
"""
try:
mean, std = optimizer.predict(X, return_std=True)
std = std.reshape(-1, 1)
except NotFittedError:
mean, std = np.zeros(shape=(X.shape[0], 1)), np.ones(shape=(X.shape[0],
1))
return EI(mean, std, optimizer.y_max, tradeoff)