physbo.search.score module¶
-
physbo.search.score.EHVI(fmean, fstd, pareto)[ソース]¶ Calculate Expected Hyper-Volume Improvement (EHVI).
Reference: (Couckuyt et al., 2014) Fast calculation of multiobjective probability of improvement and expected improvement criteria for Pareto optimization
-
physbo.search.score.EI(predictor, training, test, fmax=None)[ソース]¶ Maximum expected improvement.
- パラメータ
predictor (predictor object) -- Base class is defined in physbo.predictor.
training (physbo.variable) -- Training dataset. If already trained, the model does not use this.
test (physbo.variable) -- Inputs
fmax (float) -- Max value of posterior probability distribution. If not set fmax, the max value of posterior mean of weights is set.
- 戻り値
score
- 戻り値の型
numpy.ndarray
-
physbo.search.score.HVPI(fmean, fstd, pareto)[ソース]¶ Calculate Hypervolume-based Probability of Improvement (HVPI).
Reference: (Couckuyt et al., 2014) Fast calculation of multiobjective probability of improvement and expected improvement criteria for Pareto optimization
-
physbo.search.score.PI(predictor, training, test, fmax=None)[ソース]¶ Maximum probability of improvement.
- パラメータ
predictor (predictor object) -- Base class is defined in physbo.predictor.
training (physbo.variable) -- Training dataset. If already trained, the model does not use this.
test (physbo.variable) -- Inputs
fmax (float) -- Max value of posterior probability distribution. If not set fmax, the max value of posterior mean of weights is set.
- 戻り値
score
- 戻り値の型
numpy.ndarray
-
physbo.search.score.TS(predictor, training, test, alpha=1)[ソース]¶ Thompson sampling (See Sec. 2.1 in Materials Discovery Volume 4, June 2016, Pages 18-21)
- パラメータ
predictor (predictor object) -- Base class is defined in physbo.predictor.
training (physbo.variable) -- Training dataset. If already trained, the model does not use this.
test (physbo.variable) -- Inputs
alpha (float) -- noise for sampling source (default: 1.0)
- 戻り値
score
- 戻り値の型
numpy.ndarray