pyGPGO.acquisition module¶
-
class
pyGPGO.acquisition.
Acquisition
(mode, eps=1e-06, **params)[source]¶ Bases:
object
Acquisition function class.
Parameters: - mode (str) – Defines the behaviour of the acquisition strategy. Currently supported values are ExpectedImprovement, IntegratedExpectedÌmprovement, ProbabilityImprovement, IntegratedProbabilityImprovement, UCB, IntegratedUCB, Entropy, tExpectedImprovement, and tIntegratedExpectedImprovement. Integrated improvement functions are only to be used with MCMC surrogates.
- eps (float) – Small floating value to avoid np.sqrt or zero-division warnings.
- params (float) – Extra parameters needed for certain acquisition functions, e.g. UCB needs to be supplied with beta.
-
Entropy
(tau, mean, std, sigman=1.0)[source]¶ Predictive entropy acquisition function
Parameters: Returns: Predictive entropy.
Return type:
-
ExpectedImprovement
(tau, mean, std)[source]¶ Expected Improvement acquisition function.
Parameters: Returns: Expected improvement.
Return type:
-
IntegratedExpectedImprovement
(tau, meanmcmc, stdmcmc)[source]¶ Integrated expected improvement. Can only be used with GaussianProcessMCMC instance.
Parameters: - tau (float) – Best observed function evaluation
- meanmcmc (array-like) – Means of posterior predictive distributions after sampling.
- stdmcmc – Standard deviations of posterior predictive distributions after sampling.
Returns: Integrated Expected Improvement
Return type:
-
IntegratedProbabilityImprovement
(tau, meanmcmc, stdmcmc)[source]¶ Integrated probability of improvement. Can only be used with GaussianProcessMCMC instance.
Parameters: - tau (float) – Best observed function evaluation
- meanmcmc (array-like) – Means of posterior predictive distributions after sampling.
- stdmcmc – Standard deviations of posterior predictive distributions after sampling.
Returns: Integrated Probability of Improvement
Return type:
-
IntegratedUCB
(tau, meanmcmc, stdmcmc, beta=1.5)[source]¶ Integrated probability of improvement. Can only be used with GaussianProcessMCMC instance.
Parameters: Returns: Integrated UCB.
Return type:
-
ProbabilityImprovement
(tau, mean, std)[source]¶ Probability of Improvement acquisition function.
Parameters: Returns: Probability of improvement.
Return type:
-
UCB
(tau, mean, std, beta=1.5)[source]¶ Upper-confidence bound acquisition function.
Parameters: Returns: Upper confidence bound.
Return type:
-
__init__
(mode, eps=1e-06, **params)[source]¶ Acquisition function class.
Parameters: - mode (str) – Defines the behaviour of the acquisition strategy. Currently supported values are ExpectedImprovement, IntegratedExpectedÌmprovement, ProbabilityImprovement, IntegratedProbabilityImprovement, UCB, IntegratedUCB, Entropy, tExpectedImprovement, and tIntegratedExpectedImprovement. Integrated improvement functions are only to be used with MCMC surrogates.
- eps (float) – Small floating value to avoid np.sqrt or zero-division warnings.
- params (float) – Extra parameters needed for certain acquisition functions, e.g. UCB needs to be supplied with beta.
-
eval
(tau, mean, std)[source]¶ Evaluates selected acquisition function.
Parameters: Returns: Acquisition function value.
Return type:
-
tExpectedImprovement
(tau, mean, std, nu=3.0)[source]¶ Expected Improvement acquisition function. Only to be used with tStudentProcess surrogate.
Parameters: Returns: Expected improvement.
Return type:
-
tIntegratedExpectedImprovement
(tau, meanmcmc, stdmcmc, nu=3.0)[source]¶ Integrated expected improvement. Can only be used with tStudentProcessMCMC instance.
Parameters: - tau (float) – Best observed function evaluation
- meanmcmc (array-like) – Means of posterior predictive distributions after sampling.
- stdmcmc – Standard deviations of posterior predictive distributions after sampling.
- nu – Degrees of freedom.
Returns: Integrated Expected Improvement
Return type: