LoadCovarianceDataAction
This action operates on existing GaussianProcess
objects contained within the [Surrogates]
block. If the model provides a filename (as shown below), a [Covariance]
object equivalent to the function used in the training phase is reconstructed for use in model evaluation.
Example Input File Syntax
In the training input file we setup a GaussianProcessTrainer, with a SquaredExponential covariance function.
[Trainers]
[GP_avg_trainer]
type = GaussianProcessTrainer
execute_on = timestep_end
covariance_function = 'covar' #Choose a squared exponential for the kernel
standardize_params = 'true' #Center and scale the training params
standardize_data = 'true' #Center and scale the training data
sampler = train_sample
response = results/data:avg:value
[]
[]
[Covariance]
[covar]
type = SquaredExponentialCovariance
signal_variance = 1 #Use a signal variance of 1 in the kernel
noise_variance = 1e-6 #A small amount of noise can help with numerical stability
length_factor = '0.38971 0.38971' #Select a length factor for each parameter (k and q)
[]
[]
(contrib/moose/modules/stochastic_tools/test/tests/surrogates/gaussian_process/GP_squared_exponential_training.i)In the surrogate input file, the GaussianProcess surrogate recreates the covariance function used in training and links to it.
[Surrogates]
[GP_avg]
type = GaussianProcessSurrogate
filename = 'gauss_process_training_GP_avg_trainer.rd'
[]
[]
(contrib/moose/modules/stochastic_tools/test/tests/surrogates/gaussian_process/GP_squared_exponential_testing.i)