Multi-objective continuous Acquisition Function

PyALAF.acfn_continuous_multi.EI_multi(x, model, aggregation_function, opt, alpha=0.5, max=True, *args, **kwargs)

PyALAF.acfn_continuous_multi.GSx_multi(x, x_sample)

PyALAF.acfn_continuous_multi.GSy_multi(x, y_sample, model, aggregation_function, poly_x, *args, **kwargs)

PyALAF.acfn_continuous_multi.IDEAL_multi(x, x_samples, model, y_true, lim, aggregation_function, alpha=1, poly_x=None, tol=1e-08, *args, **kwargs)

PyALAF.acfn_continuous_multi.NIPV_multi(x, model, aggregation_function, X_int, *args, **kwargs)

Negative Integrated Posterior Variance acquisition function.

Parameters:

• x (ndarray) – Candidate point(s) to be evaluated.

• model (list of fitted sklearn GaussianProcessRegressor) – Ensemble of GP models (one per output / task).

• aggregation_function (callable) – Same signature as used by EI_multi: aggregates the per-model scores into a single value per candidate.

• X_int (ndarray) – Integration / reference points over which the posterior variance is integrated (e.g. a Monte-Carlo sample of the input domain).

PyALAF.acfn_continuous_multi.POI_multi(x, model, aggregation_function, opt, alpha, max=True, *args, **kwargs)

PyALAF.acfn_continuous_multi.QBC_multi(x, models, aggregation_function, poly_x, *args, **kwargs)

PyALAF.acfn_continuous_multi.SGSx_multi(x, x_sample, model, aggregation_function, alpha=0.5, *args, **kwargs)

PyALAF.acfn_continuous_multi.UCB_multi(x, model, aggregation_function, alpha=0.5, *args, **kwargs)

PyALAF.acfn_continuous_multi.UIDAL_multi(x, x_samples, model, aggregation_function, alpha=1, *args, **kwargs)

PyALAF.acfn_continuous_multi.iGS_multi(x, x_sample, y_sample, model, aggregation_function, poly_x=None, *args, **kwargs)

PyALAF.acfn_continuous_multi.max_multi(x, model, aggregation_function, *args, **kwargs)

PyALAF.acfn_continuous_multi.std_multi(x, model, aggregation_function, *args, **kwargs)