bayesvalidrox.surrogate_models.sequential_design.SequentialDesign

class bayesvalidrox.surrogate_models.sequential_design.SequentialDesign(MetaMod, Model, ExpDes, engine, parallel=False)

Bases: object

Contains options for choosing the next training sample iteratively.

__init__(MetaMod, Model, ExpDes, engine, parallel=False)

Methods

__init__(MetaMod, Model, ExpDes, engine[, ...])
choose_next_sample([sigma2, n_candidates, var])	Runs optimal sequential design.
dual_annealing(method, Bounds, sigma2Dict, ...)	Exploration algorithm to find the optimum parameter space.
run_util_func(method, candidates, index[, ...])	Runs the utility function based on the given method.
start_seqdesign()	Do all the preparations that need to be run before the actual training
tradeoff_weights(tradeoff_scheme, old_EDX, ...)	Calculates weights for exploration scores based on the requested scheme: None, equal, epsilon-decreasing and adaptive.
util_AlphOptDesign(candidates[, var])	Enriches the Experimental design with the requested alphabetic criterion based on exploring the space with number of sampling points.
util_BayesianActiveDesign(y_hat, std, sigma2Dict)	Computes scores based on Bayesian active design criterion (var).
util_BayesianDesign(X_can, X_MC, sigma2Dict)	Computes scores based on Bayesian sequential design criterion (var).
util_VarBasedDesign(X_can, index[, util_func])	Computes the exploitation scores based on: active learning MacKay(ALM) and active learning Cohn (ALC) Paper: Sequential Design with Mutual Information for Computer Experiments (MICE): Emulation of a Tsunami Model by Beck and Guillas (2016)

choose_next_sample(sigma2=None, n_candidates=5, var='DKL')

Runs optimal sequential design.

Parameters

sigma2dict, optional

A dictionary containing the measurement errors (sigma^2). The default is None.

n_candidatesint, optional

Number of candidate samples. The default is 5.

varstring, optional

Utility function. The default is None. # TODO: default is set to DKL, not none

Raises

NameError

Wrong utility function.

Returns

Xnewarray (n_samples, n_params)

Selected new training point(s).

dual_annealing(method, Bounds, sigma2Dict, var, Run_No, verbose=False)

Exploration algorithm to find the optimum parameter space.

Parameters

methodstring

Exploitation method: VarOptDesign, BayesActDesign and BayesOptDesign.

# TODO: BayesActDesign has no corresponding function call in this function! Bounds : list of tuples

List of lower and upper boundaries of parameters.

sigma2Dictdict

A dictionary containing the measurement errors (sigma^2).

var : unknown Run_No : int

Run number.

verbosebool, optional

Print out a summary. The default is False.

Returns

Run_Noint

Run number.

array

Optimial candidate.

run_util_func(method, candidates, index, sigma2Dict=None, var=None, X_MC=None)

Runs the utility function based on the given method.

Parameters

methodstring

Exploitation method: VarOptDesign, BayesActDesign and BayesOptDesign.

candidatesarray of shape (n_samples, n_params)

All candidate parameter sets.

indexint

ExpDesign index.

sigma2Dictdict, optional

A dictionary containing the measurement errors (sigma^2). The default is None.

varstring, optional

Utility function. The default is None.

X_MCTYPE, optional

DESCRIPTION. The default is None.

Returns

indexTYPE

DESCRIPTION.

List

Scores.

start_seqdesign() → None

Do all the preparations that need to be run before the actual training

Returns

None

tradeoff_weights(tradeoff_scheme, old_EDX, old_EDY)

Calculates weights for exploration scores based on the requested scheme: None, equal, epsilon-decreasing and adaptive.

None: No exploration. equal: Same weights for exploration and exploitation scores. epsilon-decreasing: Start with more exploration and increase the

influence of exploitation along the way with an exponential decay function

adaptive: An adaptive method based on:

Liu, Haitao, Jianfei Cai, and Yew-Soon Ong. “An adaptive sampling approach for Kriging metamodeling by maximizing expected prediction error.” Computers & Chemical Engineering 106 (2017): 171-182.

Parameters

tradeoff_schemestring

Trade-off scheme for exloration and exploitation scores.

old_EDXarray (n_samples, n_params)

Old experimental design (training points).

old_EDYdict

Old model responses (targets).

Returns

exploration_weightfloat

Exploration weight.

exploitation_weight: float

Exploitation weight.

util_AlphOptDesign(candidates, var='D-Opt')

Enriches the Experimental design with the requested alphabetic criterion based on exploring the space with number of sampling points.

Ref: Hadigol, M., & Doostan, A. (2018). Least squares polynomial chaos expansion: A review of sampling strategies., Computer Methods in Applied Mechanics and Engineering, 332, 382-407.

Arguments

candidatesint?

Number of candidate points to be searched

varstring

Alphabetic optimality criterion

Returns

X_newarray of shape (1, n_params)

The new sampling location in the input space.

util_BayesianActiveDesign(y_hat, std, sigma2Dict, var='DKL')

Computes scores based on Bayesian active design criterion (var).

It is based on the following paper: Oladyshkin, Sergey, Farid Mohammadi, Ilja Kroeker, and Wolfgang Nowak. “Bayesian3 active learning for the gaussian process emulator using information theory.” Entropy 22, no. 8 (2020): 890.

Parameters

y_hat : unknown std : unknown sigma2Dict : dict

A dictionary containing the measurement errors (sigma^2).

varstring, optional

BAL design criterion. The default is ‘DKL’.

Returns

float

Score.

util_BayesianDesign(X_can, X_MC, sigma2Dict, var='DKL')

Computes scores based on Bayesian sequential design criterion (var).

Parameters

X_canarray of shape (n_samples, n_params)

Candidate samples.

X_MC : unknown sigma2Dict : dict

A dictionary containing the measurement errors (sigma^2).

varstring, optional

Bayesian design criterion. The default is ‘DKL’.

Returns

float

Score.

util_VarBasedDesign(X_can, index, util_func='Entropy')

Computes the exploitation scores based on: active learning MacKay(ALM) and active learning Cohn (ALC) Paper: Sequential Design with Mutual Information for Computer Experiments (MICE): Emulation of a Tsunami Model by Beck and Guillas (2016)

Parameters

X_canarray of shape (n_samples, n_params)

Candidate samples.

indexint

Model output index.

util_funcstring, optional

Exploitation utility function. The default is ‘Entropy’.

Returns

float

Score.