Enable GPMSA-internal normalization

Specification

Alias: none

Argument(s): none

Description

Normalize variables and data (surrogate model build and experiment) when constructing the underlying Gaussian Process emulator. When specified, this option will normalize calibration parameters and configuration (scenario) variables based on the min/max of the build data. It will normalize simulation and experiment data by the mean and variance of the simulation build data.

Default Behavior No GPMSA-specific scaling used.

Examples

method
  bayes_calibration gpmsa
    chain_samples 1000
    seed 2460
    ...
    gpmsa_normalize