SurrogateTrainerOutput

Output for trained surrogate model data.

Overview

Surrogate models, once trained, can be output to a binary file for later use by using the SurrogateTrainerOutput object. All data within the model that was declared using declareTrainerData is automatically stored in the generated file.

Example Input File Syntax

The following snippet includes a surrogate model that is being trained and after the training is complete the model data is output using the SurrogateTrainerOutput object.

[Trainers]
  [poly_chaos]
    type = PolynomialChaosTrainer
    execute_on = timestep_end
    order = 5
    distributions = 'D_dist S_dist'
    sampler = quadrature
    response = storage/data:avg:value
  []
[]

[Outputs]
  [out]
    type = SurrogateTrainerOutput
    trainers = 'poly_chaos'
    execute_on = FINAL
  []
[]

Input Parameters

trainersA list of SurrogateTrainer objects to output.
C++ Type:std::vector<UserObjectName>
Controllable:No
Description:A list of SurrogateTrainer objects to output.

Required Parameters

use_displacedFalseEnable/disable the use of the displaced mesh for outputting
Default:False
C++ Type:bool
Controllable:No
Description:Enable/disable the use of the displaced mesh for outputting

Optional Parameters

additional_execute_onThe list of flag(s) indicating when this object should be executed, the available options include FORWARD, ADJOINT, HOMOGENEOUS_FORWARD, ADJOINT_TIMESTEP_BEGIN, ADJOINT_TIMESTEP_END, NONE, INITIAL, LINEAR, NONLINEAR, POSTCHECK, TIMESTEP_END, TIMESTEP_BEGIN, MULTIAPP_FIXED_POINT_END, MULTIAPP_FIXED_POINT_BEGIN, FINAL, FAILED, CUSTOM.
C++ Type:ExecFlagEnum
Options:FORWARD, ADJOINT, HOMOGENEOUS_FORWARD, ADJOINT_TIMESTEP_BEGIN, ADJOINT_TIMESTEP_END, NONE, INITIAL, LINEAR, NONLINEAR, POSTCHECK, TIMESTEP_END, TIMESTEP_BEGIN, MULTIAPP_FIXED_POINT_END, MULTIAPP_FIXED_POINT_BEGIN, FINAL, FAILED, CUSTOM
Controllable:No
Description:The list of flag(s) indicating when this object should be executed, the available options include FORWARD, ADJOINT, HOMOGENEOUS_FORWARD, ADJOINT_TIMESTEP_BEGIN, ADJOINT_TIMESTEP_END, NONE, INITIAL, LINEAR, NONLINEAR, POSTCHECK, TIMESTEP_END, TIMESTEP_BEGIN, MULTIAPP_FIXED_POINT_END, MULTIAPP_FIXED_POINT_BEGIN, FINAL, FAILED, CUSTOM.
execute_onINITIAL TIMESTEP_ENDThe list of flag(s) indicating when this object should be executed, the available options include FORWARD, ADJOINT, HOMOGENEOUS_FORWARD, ADJOINT_TIMESTEP_BEGIN, ADJOINT_TIMESTEP_END, NONE, INITIAL, LINEAR, NONLINEAR, POSTCHECK, TIMESTEP_END, TIMESTEP_BEGIN, MULTIAPP_FIXED_POINT_END, MULTIAPP_FIXED_POINT_BEGIN, FINAL, FAILED, CUSTOM.
Default:INITIAL TIMESTEP_END
C++ Type:ExecFlagEnum
Options:FORWARD, ADJOINT, HOMOGENEOUS_FORWARD, ADJOINT_TIMESTEP_BEGIN, ADJOINT_TIMESTEP_END, NONE, INITIAL, LINEAR, NONLINEAR, POSTCHECK, TIMESTEP_END, TIMESTEP_BEGIN, MULTIAPP_FIXED_POINT_END, MULTIAPP_FIXED_POINT_BEGIN, FINAL, FAILED, CUSTOM
Controllable:No
Description:The list of flag(s) indicating when this object should be executed, the available options include FORWARD, ADJOINT, HOMOGENEOUS_FORWARD, ADJOINT_TIMESTEP_BEGIN, ADJOINT_TIMESTEP_END, NONE, INITIAL, LINEAR, NONLINEAR, POSTCHECK, TIMESTEP_END, TIMESTEP_BEGIN, MULTIAPP_FIXED_POINT_END, MULTIAPP_FIXED_POINT_BEGIN, FINAL, FAILED, CUSTOM.
output_linearFalseSpecifies whether output occurs on each PETSc linear residual evaluation
Default:False
C++ Type:bool
Controllable:No
Description:Specifies whether output occurs on each PETSc linear residual evaluation
output_nonlinearFalseSpecifies whether output occurs on each PETSc nonlinear residual evaluation
Default:False
C++ Type:bool
Controllable:No
Description:Specifies whether output occurs on each PETSc nonlinear residual evaluation

Execution Scheduling Parameters

append_dateFalseWhen true the date and time are appended to the output filename.
Default:False
C++ Type:bool
Controllable:No
Description:When true the date and time are appended to the output filename.
append_date_formatThe format of the date/time to append, if not given UTC format is used (see http://www.cplusplus.com/reference/ctime/strftime).
C++ Type:std::string
Controllable:No
Description:The format of the date/time to append, if not given UTC format is used (see http://www.cplusplus.com/reference/ctime/strftime).
file_baseThe desired solution output name without an extension. If not provided, MOOSE sets it with Outputs/file_base when available. Otherwise, MOOSE uses input file name and this object name for a master input or uses master file_base, the subapp name and this object name for a subapp input to set it.
C++ Type:std::string
Controllable:No
Description:The desired solution output name without an extension. If not provided, MOOSE sets it with Outputs/file_base when available. Otherwise, MOOSE uses input file name and this object name for a master input or uses master file_base, the subapp name and this object name for a subapp input to set it.
output_if_base_containsIf this is supplied then output will only be done in the case that the output base contains one of these strings. This is helpful in outputting only a subset of outputs when using MultiApps.
C++ Type:std::vector<std::string>
Controllable:No
Description:If this is supplied then output will only be done in the case that the output base contains one of these strings. This is helpful in outputting only a subset of outputs when using MultiApps.
padding4The number of digits for the extension suffix (e.g., out.e-s002)
Default:4
C++ Type:unsigned int
Controllable:No
Description:The number of digits for the extension suffix (e.g., out.e-s002)

File Name Customization Parameters

control_tagsAdds user-defined labels for accessing object parameters via control logic.
C++ Type:std::vector<std::string>
Controllable:No
Description:Adds user-defined labels for accessing object parameters via control logic.
enableTrueSet the enabled status of the MooseObject.
Default:True
C++ Type:bool
Controllable:Yes
Description:Set the enabled status of the MooseObject.

Advanced Parameters

end_stepTime step at which this output object stop operating
C++ Type:int
Controllable:No
Description:Time step at which this output object stop operating
end_timeTime at which this output object stop operating
C++ Type:double
Controllable:No
Description:Time at which this output object stop operating
min_simulation_time_interval0The minimum simulation time between output steps
Default:0
C++ Type:double
Controllable:No
Description:The minimum simulation time between output steps
simulation_time_interval1.79769e+308The target simulation time interval (in seconds) at which to output
Default:1.79769e+308
C++ Type:double
Controllable:No
Description:The target simulation time interval (in seconds) at which to output
start_stepTime step at which this output object begins to operate
C++ Type:int
Controllable:No
Description:Time step at which this output object begins to operate
start_timeTime at which this output object begins to operate
C++ Type:double
Controllable:No
Description:Time at which this output object begins to operate
sync_onlyFalseOnly export results at sync times
Default:False
C++ Type:bool
Controllable:No
Description:Only export results at sync times
sync_timesTimes at which the output and solution is forced to occur
C++ Type:std::vector<double>
Controllable:No
Description:Times at which the output and solution is forced to occur
sync_times_objectTimes object providing the times at which the output and solution is forced to occur
C++ Type:TimesName
Controllable:No
Description:Times object providing the times at which the output and solution is forced to occur
time_step_interval1The interval (number of time steps) at which output occurs. Unless explicitly set, the default value of this parameter is set to infinity if the wall_time_interval is explicitly set.
Default:1
C++ Type:unsigned int
Controllable:No
Description:The interval (number of time steps) at which output occurs. Unless explicitly set, the default value of this parameter is set to infinity if the wall_time_interval is explicitly set.
time_tolerance1e-14Time tolerance utilized checking start and end times
Default:1e-14
C++ Type:double
Controllable:No
Description:Time tolerance utilized checking start and end times
wall_time_interval1.79769e+308The target wall time interval (in seconds) at which to output
Default:1.79769e+308
C++ Type:double
Controllable:No
Description:The target wall time interval (in seconds) at which to output

Timing And Frequency Of Output Parameters

linear_residual_dt_divisor1000Number of divisions applied to time step when outputting linear residuals
Default:1000
C++ Type:double
Controllable:No
Description:Number of divisions applied to time step when outputting linear residuals
linear_residual_end_timeSpecifies an end time to begin output on each linear residual evaluation
C++ Type:double
Controllable:No
Description:Specifies an end time to begin output on each linear residual evaluation
linear_residual_start_timeSpecifies a start time to begin output on each linear residual evaluation
C++ Type:double
Controllable:No
Description:Specifies a start time to begin output on each linear residual evaluation
nonlinear_residual_dt_divisor1000Number of divisions applied to time step when outputting non-linear residuals
Default:1000
C++ Type:double
Controllable:No
Description:Number of divisions applied to time step when outputting non-linear residuals
nonlinear_residual_end_timeSpecifies an end time to begin output on each nonlinear residual evaluation
C++ Type:double
Controllable:No
Description:Specifies an end time to begin output on each nonlinear residual evaluation
nonlinear_residual_start_timeSpecifies a start time to begin output on each nonlinear residual evaluation
C++ Type:double
Controllable:No
Description:Specifies a start time to begin output on each nonlinear residual evaluation

Petsc Linear/Nonlinear Output Parameters

Input Files

(modules/combined/examples/stochastic/poly_chaos_train_uniform.i)
(modules/stochastic_tools/test/tests/surrogates/load_store/train.i)
(modules/stochastic_tools/examples/surrogates/poly_chaos_normal_quad.i)
(modules/stochastic_tools/test/tests/surrogates/gaussian_process/GP_squared_exponential_training.i)
(modules/stochastic_tools/examples/surrogates/combined/trans_diff_2d/trans_diff_trainer.i)
(modules/stochastic_tools/examples/surrogates/pod_rb/2d_multireg/trainer.i)
(modules/stochastic_tools/examples/surrogates/nearest_point_training.i)
(modules/stochastic_tools/test/tests/auxkernels/surrogate_aux/surrogate.i)
(modules/stochastic_tools/examples/surrogates/poly_chaos_normal_mc.i)
(modules/stochastic_tools/test/tests/surrogates/pod_rb/errors/trainer.i)
(modules/stochastic_tools/examples/surrogates/poly_chaos_uniform_mc.i)
(modules/stochastic_tools/examples/surrogates/poly_chaos_uniform_quad.i)
(modules/stochastic_tools/test/tests/surrogates/nearest_point/train.i)
(modules/stochastic_tools/examples/surrogates/polynomial_regression/normal_train.i)
(modules/stochastic_tools/test/tests/surrogates/polynomial_regression/train.i)
(modules/stochastic_tools/test/tests/surrogates/libtorch_nn/train.i)
(modules/stochastic_tools/test/tests/userobjects/inverse_mapping/create_mapping_main.i)
(modules/stochastic_tools/examples/surrogates/polynomial_regression/uniform_train.i)
(modules/stochastic_tools/examples/surrogates/gaussian_process/GP_normal_mc.i)
(modules/stochastic_tools/test/tests/surrogates/pod_rb/boundary/trainer.i)
(modules/stochastic_tools/test/tests/surrogates/pod_rb/internal/trainer.i)

(modules/stochastic_tools/test/tests/surrogates/load_store/train.i)

[StochasticTools]
[]

[Distributions]
  [D_dist]
    type = Uniform
    lower_bound = 2.5
    upper_bound = 7.5
  []
  [S_dist]
    type = Uniform
    lower_bound = 2.5
    upper_bound = 7.5
  []
[]

[Samplers]
  [quadrature]
    type = Quadrature
    distributions = 'D_dist S_dist'
    execute_on = INITIAL
    order = 5
  []
[]

[MultiApps]
  [quad_sub]
    type = SamplerFullSolveMultiApp
    input_files = sub.i
    sampler = quadrature
    mode = batch-restore
  []
[]

[Transfers]
  [quad]
    type = SamplerParameterTransfer
    to_multi_app = quad_sub
    sampler = quadrature
    parameters = 'Materials/diffusivity/prop_values Materials/xs/prop_values'
  []
  [data]
    type = SamplerReporterTransfer
    from_multi_app = quad_sub
    sampler = quadrature
    stochastic_reporter = storage
    from_reporter = avg/value
  []
[]

[Reporters]
  [storage]
    type = StochasticReporter
    parallel_type = ROOT
  []
[]

[Trainers]
  [poly_chaos]
    type = PolynomialChaosTrainer
    execute_on = timestep_end
    order = 5
    distributions = 'D_dist S_dist'
    sampler = quadrature
    response = storage/data:avg:value
  []
[]

[Outputs]
  [out]
    type = SurrogateTrainerOutput
    trainers = 'poly_chaos'
    execute_on = FINAL
  []
[]

(modules/combined/examples/stochastic/poly_chaos_train_uniform.i)

[StochasticTools]
[]

[Distributions]
  [cond_inner]
    type = Uniform
    lower_bound = 20
    upper_bound = 30
  []
  [cond_outer]
    type = Uniform
    lower_bound = 90
    upper_bound = 110
  []
  [heat_source]
    type = Uniform
    lower_bound = 9000
    upper_bound = 11000
  []
  [alpha_inner]
    type = Uniform
    lower_bound = 1e-6
    upper_bound = 3e-6
  []
  [alpha_outer]
    type = Uniform
    lower_bound = 5e-7
    upper_bound = 1.5e-6
  []
  [ymod_inner]
    type = Uniform
    lower_bound = 2e5
    upper_bound = 2.2e5
  []
  [ymod_outer]
    type = Uniform
    lower_bound = 3e5
    upper_bound = 3.2e5
  []
  [prat_inner]
    type = Uniform
    lower_bound = 0.29
    upper_bound = 0.31
  []
  [prat_outer]
    type = Uniform
    lower_bound = 0.19
    upper_bound = 0.21
  []
[]

[GlobalParams]
  distributions = 'cond_inner cond_outer heat_source alpha_inner alpha_outer ymod_inner ymod_outer prat_inner prat_outer'
[]

[Samplers]
  [sample]
    type = Quadrature
    sparse_grid = smolyak
    order = 5
    execute_on = INITIAL
  []
[]

[MultiApps]
  [sub]
    type = SamplerFullSolveMultiApp
    input_files = graphite_ring_thermomechanics.i
    sampler = sample
    mode = batch-reset
  []
[]

[Transfers]
  [sub]
    type = SamplerParameterTransfer
    to_multi_app = sub
    sampler = sample
    parameters = 'Materials/cond_inner/prop_values Materials/cond_outer/prop_values
                  Postprocessors/heat_source/scale_factor
                  Materials/thermal_strain_inner/thermal_expansion_coeff Materials/thermal_strain_outer/thermal_expansion_coeff
                  Materials/elasticity_tensor_inner/youngs_modulus Materials/elasticity_tensor_outer/youngs_modulus
                  Materials/elasticity_tensor_inner/poissons_ratio Materials/elasticity_tensor_outer/poissons_ratio'
    check_multiapp_execute_on = false
  []
  [data]
    type = SamplerReporterTransfer
    from_multi_app = sub
    sampler = sample
    stochastic_reporter = storage
    from_reporter = 'temp_center_inner/value  temp_center_outer/value  temp_end_inner/value  temp_end_outer/value
                     dispx_center_inner/value dispx_center_outer/value dispx_end_inner/value dispx_end_outer/value
                     dispz_inner/value dispz_outer/value'
  []
[]

[Reporters]
  [storage]
    type = StochasticReporter
  []
[]

[Trainers]
  [temp_center_inner]
    type = PolynomialChaosTrainer
    execute_on = timestep_end
    order = 4
    sampler = sample
    response = storage/data:temp_center_inner:value
  []
  [temp_center_outer]
    type = PolynomialChaosTrainer
    execute_on = timestep_end
    order = 4
    sampler = sample
    response = storage/data:temp_center_outer:value
  []
  [temp_end_inner]
    type = PolynomialChaosTrainer
    execute_on = timestep_end
    order = 4
    sampler = sample
    response = storage/data:temp_end_inner:value
  []
  [temp_end_outer]
    type = PolynomialChaosTrainer
    execute_on = timestep_end
    order = 4
    sampler = sample
    response = storage/data:temp_end_outer:value
  []
  [dispx_center_inner]
    type = PolynomialChaosTrainer
    execute_on = timestep_end
    order = 4
    sampler = sample
    response = storage/data:dispx_center_inner:value
  []
  [dispx_center_outer]
    type = PolynomialChaosTrainer
    execute_on = timestep_end
    order = 4
    sampler = sample
    response = storage/data:dispx_center_outer:value
  []
  [dispx_end_inner]
    type = PolynomialChaosTrainer
    execute_on = timestep_end
    order = 4
    sampler = sample
    response = storage/data:dispx_end_inner:value
  []
  [dispx_end_outer]
    type = PolynomialChaosTrainer
    execute_on = timestep_end
    order = 4
    sampler = sample
    response = storage/data:dispx_end_outer:value
  []
  [dispz_inner]
    type = PolynomialChaosTrainer
    execute_on = timestep_end
    order = 4
    sampler = sample
    response = storage/data:dispz_inner:value
  []
  [dispz_outer]
    type = PolynomialChaosTrainer
    execute_on = timestep_end
    order = 4
    sampler = sample
    response = storage/data:dispz_outer:value
  []
[]

[Outputs]
  [out]
    type = SurrogateTrainerOutput
    trainers = 'temp_center_inner  temp_center_outer  temp_end_inner  temp_end_outer
                dispx_center_inner dispx_center_outer dispx_end_inner dispx_end_outer
                dispz_inner dispz_outer'
    execute_on = FINAL
  []
[]

(modules/stochastic_tools/test/tests/surrogates/load_store/train.i)

[StochasticTools]
[]

[Distributions]
  [D_dist]
    type = Uniform
    lower_bound = 2.5
    upper_bound = 7.5
  []
  [S_dist]
    type = Uniform
    lower_bound = 2.5
    upper_bound = 7.5
  []
[]

[Samplers]
  [quadrature]
    type = Quadrature
    distributions = 'D_dist S_dist'
    execute_on = INITIAL
    order = 5
  []
[]

[MultiApps]
  [quad_sub]
    type = SamplerFullSolveMultiApp
    input_files = sub.i
    sampler = quadrature
    mode = batch-restore
  []
[]

[Transfers]
  [quad]
    type = SamplerParameterTransfer
    to_multi_app = quad_sub
    sampler = quadrature
    parameters = 'Materials/diffusivity/prop_values Materials/xs/prop_values'
  []
  [data]
    type = SamplerReporterTransfer
    from_multi_app = quad_sub
    sampler = quadrature
    stochastic_reporter = storage
    from_reporter = avg/value
  []
[]

[Reporters]
  [storage]
    type = StochasticReporter
    parallel_type = ROOT
  []
[]

[Trainers]
  [poly_chaos]
    type = PolynomialChaosTrainer
    execute_on = timestep_end
    order = 5
    distributions = 'D_dist S_dist'
    sampler = quadrature
    response = storage/data:avg:value
  []
[]

[Outputs]
  [out]
    type = SurrogateTrainerOutput
    trainers = 'poly_chaos'
    execute_on = FINAL
  []
[]

(modules/stochastic_tools/examples/surrogates/poly_chaos_normal_quad.i)

[StochasticTools]
[]

[Distributions]
  [k_dist]
    type = Normal
    mean = 5
    standard_deviation = 2
  []
  [q_dist]
    type = Normal
    mean = 10000
    standard_deviation = 500
  []
  [L_dist]
    type = Normal
    mean = 0.03
    standard_deviation = 0.01
  []
  [Tinf_dist]
    type = Normal
    mean = 300
    standard_deviation = 10
  []
[]

[Samplers]
  [sample]
    type = Quadrature
    order = 10
    distributions = 'k_dist q_dist L_dist Tinf_dist'
    execute_on = PRE_MULTIAPP_SETUP
  []
[]

[MultiApps]
  [sub]
    type = SamplerFullSolveMultiApp
    input_files = sub.i
    sampler = sample
  []
[]

[Controls]
  [cmdline]
    type = MultiAppSamplerControl
    multi_app = sub
    sampler = sample
    param_names = 'Materials/conductivity/prop_values Kernels/source/value Mesh/xmax BCs/right/value'
  []
[]

[Transfers]
  [data]
    type = SamplerReporterTransfer
    from_multi_app = sub
    sampler = sample
    stochastic_reporter = results
    from_reporter = 'avg/value max/value'
  []
[]

[Reporters]
  [results]
    type = StochasticReporter
  []
[]

[Trainers]
  [poly_chaos_avg]
    type = PolynomialChaosTrainer
    execute_on = timestep_end
    order = 10
    distributions = 'k_dist q_dist L_dist Tinf_dist'
    sampler = sample
    response = results/data:avg:value
  []
  [poly_chaos_max]
    type = PolynomialChaosTrainer
    execute_on = timestep_end
    order = 10
    distributions = 'k_dist q_dist L_dist Tinf_dist'
    sampler = sample
    response = results/data:max:value
  []
[]

[Outputs]
  file_base = poly_chaos_training
  [out]
    type = SurrogateTrainerOutput
    trainers = 'poly_chaos_avg poly_chaos_max'
    execute_on = FINAL
  []
[]

(modules/stochastic_tools/test/tests/surrogates/gaussian_process/GP_squared_exponential_training.i)

[StochasticTools]
[]

[Distributions]
  [k_dist]
    type = Uniform
    lower_bound = 1
    upper_bound = 10
  []
  [q_dist]
    type = Uniform
    lower_bound = 9000
    upper_bound = 11000
  []
[]

[Samplers]
  [train_sample]
    type = MonteCarlo
    num_rows = 10
    distributions = 'k_dist q_dist'
    execute_on = PRE_MULTIAPP_SETUP
  []
[]

[MultiApps]
  [sub]
    type = SamplerFullSolveMultiApp
    input_files = sub.i
    sampler = train_sample
  []
[]

[Controls]
  [cmdline]
    type = MultiAppSamplerControl
    multi_app = sub
    sampler = train_sample
    param_names = 'Materials/conductivity/prop_values Kernels/source/value'
  []
[]

[Transfers]
  [data]
    type = SamplerReporterTransfer
    from_multi_app = sub
    sampler = train_sample
    stochastic_reporter = results
    from_reporter = 'avg/value'
  []
[]

[Reporters]
  [results]
    type = StochasticReporter
    parallel_type = ROOT
  []
[]

[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)
  []
[]

[Outputs]
  file_base = gauss_process_training
  [out]
    type = SurrogateTrainerOutput
    trainers = 'GP_avg_trainer'
    execute_on = FINAL
  []
[]

(modules/stochastic_tools/examples/surrogates/combined/trans_diff_2d/trans_diff_trainer.i)

[StochasticTools]
[]

[Distributions]
  [C_dist]
    type = Uniform
    lower_bound = 0.01
    upper_bound = 0.02
  []
  [f_dist]
    type = Uniform
    lower_bound = 15
    upper_bound = 25
  []
  [init_dist]
    type = Uniform
    lower_bound = 270
    upper_bound = 330
  []
[]

[Samplers]
  [sample]
    type = Quadrature
    order = 5
    distributions = 'C_dist f_dist init_dist'
    execute_on = PRE_MULTIAPP_SETUP
  []
[]

[MultiApps]
  [runner]
    type = SamplerFullSolveMultiApp
    input_files = 'trans_diff_sub.i'
    sampler = sample
  []
[]

[Controls]
  [cmdline]
    type = MultiAppSamplerControl
    multi_app = runner
    sampler = sample
    param_names = 'Materials/diff_coeff/constant_expressions Functions/src_func/vals Variables/T/initial_condition'
  []
[]

[Transfers]
  [results]
    type = SamplerReporterTransfer
    from_multi_app = runner
    sampler = sample
    stochastic_reporter = trainer_results
    from_reporter = 'time_max/value time_min/value'
  []
[]

[Reporters]
  [trainer_results]
    type = StochasticReporter
  []
[]

[Trainers]
  [pc_max]
    type = PolynomialChaosTrainer
    execute_on = final
    order = 5
    distributions = 'C_dist f_dist init_dist'
    sampler = sample
    response = trainer_results/results:time_max:value
  []
  [pc_min]
    type = PolynomialChaosTrainer
    execute_on = final
    order = 5
    distributions = 'C_dist f_dist init_dist'
    sampler = sample
    response = trainer_results/results:time_min:value
  []
  [np_max]
    type = NearestPointTrainer
    execute_on = final
    sampler = sample
    response = trainer_results/results:time_max:value
  []
  [np_min]
    type = NearestPointTrainer
    execute_on = final
    sampler = sample
    response = trainer_results/results:time_min:value
  []
  [pr_max]
    type = PolynomialRegressionTrainer
    regression_type = "ols"
    execute_on = final
    max_degree = 4
    sampler = sample
    response = trainer_results/results:time_max:value
  []
  [pr_min]
    type = PolynomialRegressionTrainer
    regression_type = "ols"
    execute_on = final
    max_degree = 4
    sampler = sample
    response = trainer_results/results:time_min:value
  []
[]

[Outputs]
  [out]
    type = SurrogateTrainerOutput
    trainers = 'pc_max pc_min np_max np_min pr_max pr_min'
    execute_on = FINAL
  []
[]

(modules/stochastic_tools/examples/surrogates/pod_rb/2d_multireg/trainer.i)

[StochasticTools]
[]

[Distributions]
  [D012_dist]
    type = Uniform
    lower_bound = 0.2
    upper_bound = 0.8
  []
  [D1_dist]
    type = Uniform
    lower_bound = 0.2
    upper_bound = 0.8
  []
  [D2_dist]
    type = Uniform
    lower_bound = 0.2
    upper_bound = 0.8
  []
  [D3_dist]
    type = Uniform
    lower_bound = 0.15
    upper_bound = 0.6
  []
  [absxs0_dist]
    type = Uniform
    lower_bound = 0.0425
    upper_bound = 0.17
  []
  [absxs1_dist]
    type = Uniform
    lower_bound = 0.065
    upper_bound = 0.26
  []
  [absxs2_dist]
    type = Uniform
    lower_bound = 0.04
    upper_bound = 0.16
  []
  [absxs3_dist]
    type = Uniform
    lower_bound = 0.005
    upper_bound = 0.02
  []
  [src_dist]
    type = Uniform
    lower_bound = 5
    upper_bound = 20
  []
[]

[Samplers]
  [sample]
    type = LatinHypercube
    distributions = 'D012_dist D012_dist D012_dist D3_dist
                     absxs0_dist absxs1_dist absxs2_dist absxs3_dist
                     src_dist src_dist src_dist'
    num_rows = 100
    execute_on = PRE_MULTIAPP_SETUP
    max_procs_per_row = 1
  []
[]

[MultiApps]
  [sub]
    type = PODFullSolveMultiApp
    input_files = sub.i
    sampler = sample
    trainer_name = 'pod_rb'
    execute_on = 'timestep_begin final'
    max_procs_per_app = 1
  []
[]

[Transfers]
  [param]
    type = SamplerParameterTransfer
    to_multi_app = sub
    sampler = sample
    parameters = 'Materials/D0/prop_values
                  Materials/D1/prop_values
                  Materials/D2/prop_values
                  Materials/D3/prop_values
                  Materials/absxs0/prop_values
                  Materials/absxs1/prop_values
                  Materials/absxs2/prop_values
                  Materials/absxs3/prop_values
                  Kernels/src0/value
                  Kernels/src1/value
                  Kernels/src2/value'
    execute_on = 'timestep_begin'
    check_multiapp_execute_on = false
  []
  [data]
    type = PODSamplerSolutionTransfer
    from_multi_app = sub
    sampler = sample
    trainer_name = 'pod_rb'
    execute_on = 'timestep_begin'
    check_multiapp_execute_on = false
  []
  [mode]
    type = PODSamplerSolutionTransfer
    to_multi_app = sub
    sampler = sample
    trainer_name = 'pod_rb'
    execute_on = 'final'
    check_multiapp_execute_on = false
  []
  [res]
    type = PODResidualTransfer
    from_multi_app = sub
    sampler = sample
    trainer_name = 'pod_rb'
    execute_on = 'final'
    check_multiapp_execute_on = false
  []
[]

[Trainers]
  [pod_rb]
    type = PODReducedBasisTrainer
    var_names = 'psi'
    error_res = '1e-9'
    tag_names = 'diff0 diff1 diff2 diff3 abs0 abs1 abs2 abs3 src0 src1 src2'
    tag_types = 'op op op op op op op op src src src'
    execute_on = 'timestep_begin final'
  []
[]

[Outputs]
  [out]
    type = SurrogateTrainerOutput
    trainers = 'pod_rb'
    execute_on = FINAL
  []
[]

(modules/stochastic_tools/examples/surrogates/nearest_point_training.i)

[StochasticTools]
[]

[Samplers]
  [grid]
    type = CartesianProduct
    execute_on = PRE_MULTIAPP_SETUP
    # Grid spacing:
    # k:    1.45  2.35  3.25  4.15  5.05  5.95  6.85  7.75  8.65  9.55
    # q:    9100  9300  9500  9700  9900  10100 10300 10500 10700 10900
    # L:    0.012 0.016 0.020 0.024 0.028 0.032 0.036 0.040 0.044 0.048
    # Tinf: 291   293   295   297   299   301   303   305   307   309
    linear_space_items = '1.45  0.9   10
                          9100  200   10
                          0.012 0.004 10
                          291   2     10'
  []
[]

[MultiApps]
  [sub]
    type = SamplerFullSolveMultiApp
    input_files = sub.i
    sampler = grid
  []
[]

[Controls]
  [cmdline]
    type = MultiAppSamplerControl
    multi_app = sub
    sampler = grid
    param_names = 'Materials/conductivity/prop_values Kernels/source/value Mesh/xmax BCs/right/value'
  []
[]

[Transfers]
  [data]
    type = SamplerReporterTransfer
    from_multi_app = sub
    sampler = grid
    stochastic_reporter = results
    from_reporter = 'avg/value max/value'
  []
[]

[Reporters]
  [results]
    type = StochasticMatrix
    sampler = grid
    outputs = none
  []
[]

[Trainers]
  [nearest_point_avg]
    type = NearestPointTrainer
    execute_on = timestep_end
    sampler = grid
    predictors = 'results/grid_0'
    predictor_cols = '1 2 3'
    response = results/data:avg:value
  []
  [nearest_point_max]
    type = NearestPointTrainer
    execute_on = timestep_end
    sampler = grid
    response = results/data:max:value
  []
[]

[Outputs]
  [out]
    type = SurrogateTrainerOutput
    trainers = 'nearest_point_avg nearest_point_max'
    execute_on = FINAL
  []
[]

(modules/stochastic_tools/test/tests/auxkernels/surrogate_aux/surrogate.i)

[StochasticTools]
[]

[Distributions]
  [uniform]
    type = Uniform
    lower_bound = 0
    upper_bound = 1
  []
[]

[Samplers]
  [mc]
    type = MonteCarlo
    distributions = 'uniform uniform uniform uniform'
    num_rows = 50
  []
[]

[GlobalParams]
  sampler = mc
[]

[MultiApps]
  [model]
    type = SamplerFullSolveMultiApp
    input_files = model.i
    mode = batch-restore
    no_backup_and_restore = true
  []
[]

[Transfers]
  [param]
    type = SamplerParameterTransfer
    to_multi_app = model
    parameters ='Postprocessors/x1/value Postprocessors/x2/value Postprocessors/x3/value Postprocessors/x4/value'
  []
  [data]
    type = SamplerReporterTransfer
    from_multi_app = model
    stochastic_reporter = storage
    from_reporter = 'val/value'
  []
[]

[Reporters]
  [storage]
    type = StochasticReporter
  []
[]

[Trainers]
  [poly_regression]
    type = PolynomialRegressionTrainer
    regression_type = ols
    max_degree = 2
    response = storage/data:val:value
  []
[]

[Outputs]
  [trainer]
    type = SurrogateTrainerOutput
    trainers = poly_regression
  []
[]

(modules/stochastic_tools/examples/surrogates/poly_chaos_normal_mc.i)

[StochasticTools]
[]

[Distributions]
  [k_dist]
    type = Normal
    mean = 5
    standard_deviation = 2
  []
  [q_dist]
    type = Normal
    mean = 10000
    standard_deviation = 500
  []
  [L_dist]
    type = Normal
    mean = 0.03
    standard_deviation = 0.01
  []
  [Tinf_dist]
    type = Normal
    mean = 300
    standard_deviation = 10
  []
[]

[Samplers]
  [sample]
    type = MonteCarlo
    num_rows = 10000
    distributions = 'k_dist q_dist L_dist Tinf_dist'
    execute_on = PRE_MULTIAPP_SETUP
  []
[]

[MultiApps]
  [sub]
    type = SamplerFullSolveMultiApp
    input_files = sub.i
    sampler = sample
  []
[]

[Controls]
  [cmdline]
    type = MultiAppSamplerControl
    multi_app = sub
    sampler = sample
    param_names = 'Materials/conductivity/prop_values Kernels/source/value Mesh/xmax BCs/right/value'
  []
[]

[Transfers]
  [data]
    type = SamplerReporterTransfer
    from_multi_app = sub
    sampler = sample
    stochastic_reporter = results
    from_reporter = 'avg/value max/value'
  []
[]

[Reporters]
  [results]
    type = StochasticReporter
  []
[]

[Trainers]
  [poly_chaos_avg]
    type = PolynomialChaosTrainer
    execute_on = timestep_end
    order = 10
    regression_type = integration
    distributions = 'k_dist q_dist L_dist Tinf_dist'
    sampler = sample
    response = results/data:avg:value
  []
  [poly_chaos_max]
    type = PolynomialChaosTrainer
    execute_on = timestep_end
    order = 10
    regression_type = integration
    distributions = 'k_dist q_dist L_dist Tinf_dist'
    sampler = sample
    response = results/data:max:value
  []
[]

[Outputs]
  file_base = poly_chaos_training
  [out]
    type = SurrogateTrainerOutput
    trainers = 'poly_chaos_avg poly_chaos_max'
    execute_on = FINAL
  []
[]

(modules/stochastic_tools/test/tests/surrogates/pod_rb/errors/trainer.i)

[StochasticTools]
[]

[Distributions]
  [k_dist]
    type = Uniform
    lower_bound = 2.5
    upper_bound = 7.5
  []
  [alpha_dist]
    type = Uniform
    lower_bound = 2.5
    upper_bound = 7.5
  []
  [S_dist]
    type = Uniform
    lower_bound = 2.5
    upper_bound = 7.5
  []
[]

[Samplers]
  [sample]
    type = LatinHypercube
    distributions = 'k_dist alpha_dist S_dist'
    num_rows = 3
    execute_on = PRE_MULTIAPP_SETUP
    max_procs_per_row = 1
  []
[]

[MultiApps]
  [sub]
    type = PODFullSolveMultiApp
    input_files = sub.i
    sampler = sample
    trainer_name = 'pod_rb'
    execute_on = 'timestep_begin final'
    max_procs_per_app = 1
  []
[]

[Transfers]
  [param]
    type = SamplerParameterTransfer
    to_multi_app = sub
    sampler = sample
    parameters = 'Materials/k/prop_values Materials/alpha/prop_values Kernels/source/value'
    execute_on = 'timestep_begin'
    check_multiapp_execute_on = false
  []
  [snapshot]
    type = PODSamplerSolutionTransfer
    from_multi_app = sub
    sampler = sample
    trainer_name = 'pod_rb'
    execute_on = 'timestep_begin'
    check_multiapp_execute_on = false
  []
  [mode]
    type = PODSamplerSolutionTransfer
    to_multi_app = sub
    sampler = sample
    trainer_name = 'pod_rb'
    execute_on = 'final'
    check_multiapp_execute_on = false
  []
  [res]
    type = PODResidualTransfer
    from_multi_app = sub
    sampler = sample
    trainer_name = "pod_rb"
    execute_on = 'final'
    check_multiapp_execute_on = false
  []
[]

[Trainers]
  [pod_rb]
    type = PODReducedBasisTrainer
    var_names = 'u'
    error_res = '1e-9'
    tag_names = 'diff react bodyf'
    tag_types = 'op op src'
    execute_on = 'timestep_begin final'
  []
[]

[Outputs]
  [out]
    type = SurrogateTrainerOutput
    trainers = 'pod_rb'
    execute_on = FINAL
  []
[]

(modules/stochastic_tools/examples/surrogates/poly_chaos_uniform_mc.i)

[StochasticTools]
[]

[Distributions]
  [k_dist]
    type = Uniform
    lower_bound = 1
    upper_bound = 10
  []
  [q_dist]
    type = Uniform
    lower_bound = 9000
    upper_bound = 11000
  []
  [L_dist]
    type = Uniform
    lower_bound = 0.01
    upper_bound = 0.05
  []
  [Tinf_dist]
    type = Uniform
    lower_bound = 290
    upper_bound = 310
  []
[]

[Samplers]
  [sample]
    type = MonteCarlo
    num_rows = 10000
    distributions = 'k_dist q_dist L_dist Tinf_dist'
    execute_on = PRE_MULTIAPP_SETUP
  []
[]

[MultiApps]
  [sub]
    type = SamplerFullSolveMultiApp
    input_files = sub.i
    sampler = sample
  []
[]

[Controls]
  [cmdline]
    type = MultiAppSamplerControl
    multi_app = sub
    sampler = sample
    param_names = 'Materials/conductivity/prop_values Kernels/source/value Mesh/xmax BCs/right/value'
  []
[]

[Transfers]
  [data]
    type = SamplerReporterTransfer
    from_multi_app = sub
    sampler = sample
    stochastic_reporter = results
    from_reporter = 'avg/value max/value'
  []
[]

[Reporters]
  [results]
    type = StochasticReporter
  []
[]

[Trainers]
  [poly_chaos_avg]
    type = PolynomialChaosTrainer
    execute_on = timestep_end
    order = 10
    regression_type = integration
    distributions = 'k_dist q_dist L_dist Tinf_dist'
    sampler = sample
    response = results/data:avg:value
  []
  [poly_chaos_max]
    type = PolynomialChaosTrainer
    execute_on = timestep_end
    order = 10
    regression_type = integration
    distributions = 'k_dist q_dist L_dist Tinf_dist'
    sampler = sample
    response = results/data:max:value
  []
[]

[Outputs]
  file_base = poly_chaos_training
  [out]
    type = SurrogateTrainerOutput
    trainers = 'poly_chaos_avg poly_chaos_max'
    execute_on = FINAL
  []
[]

(modules/stochastic_tools/examples/surrogates/poly_chaos_uniform_quad.i)

[StochasticTools]
[]

[Distributions]
  [k_dist]
    type = Uniform
    lower_bound = 1
    upper_bound = 10
  []
  [q_dist]
    type = Uniform
    lower_bound = 9000
    upper_bound = 11000
  []
  [L_dist]
    type = Uniform
    lower_bound = 0.01
    upper_bound = 0.05
  []
  [Tinf_dist]
    type = Uniform
    lower_bound = 290
    upper_bound = 310
  []
[]

[Samplers]
  [sample]
    type = Quadrature
    order = 10
    distributions = 'k_dist q_dist L_dist Tinf_dist'
    execute_on = PRE_MULTIAPP_SETUP
  []
[]

[MultiApps]
  [sub]
    type = SamplerFullSolveMultiApp
    input_files = sub.i
    sampler = sample
  []
[]

[Controls]
  [cmdline]
    type = MultiAppSamplerControl
    multi_app = sub
    sampler = sample
    param_names = 'Materials/conductivity/prop_values Kernels/source/value Mesh/xmax BCs/right/value'
  []
[]

[Transfers]
  [data]
    type = SamplerReporterTransfer
    from_multi_app = sub
    sampler = sample
    stochastic_reporter = results
    from_reporter = 'avg/value max/value'
  []
[]

[Reporters]
  [results]
    type = StochasticReporter
  []
[]

[Trainers]
  [poly_chaos_avg]
    type = PolynomialChaosTrainer
    execute_on = timestep_end
    order = 10
    distributions = 'k_dist q_dist L_dist Tinf_dist'
    sampler = sample
    response = results/data:avg:value
  []
  [poly_chaos_max]
    type = PolynomialChaosTrainer
    execute_on = timestep_end
    order = 10
    distributions = 'k_dist q_dist L_dist Tinf_dist'
    sampler = sample
    response = results/data:max:value
  []
[]

[Outputs]
  file_base = poly_chaos_training
  [out]
    type = SurrogateTrainerOutput
    trainers = 'poly_chaos_avg poly_chaos_max'
    execute_on = FINAL
  []
[]

(modules/stochastic_tools/test/tests/surrogates/nearest_point/train.i)

[StochasticTools]
[]

[Samplers]
  [sample]
    type = CartesianProduct
    linear_space_items = '0 1 10
                          0 1 10
                          0 1 10'
  []
[]

[VectorPostprocessors]
  [values]
    type = GFunction
    sampler = sample
    q_vector = '0 0 0'
    execute_on = INITIAL
    outputs = none
  []
[]

[Trainers]
  [train]
    type = NearestPointTrainer
    sampler = sample
    response = values/g_values
  []
[]

[Outputs]
  [out]
    type = SurrogateTrainerOutput
    trainers = 'train'
    execute_on = FINAL
  []
[]

(modules/stochastic_tools/examples/surrogates/polynomial_regression/normal_train.i)

[StochasticTools]
[]

[Distributions]
  [k_dist]
    type = Normal
    mean = 5
    standard_deviation = 2
  []
  [q_dist]
    type = Normal
    mean = 10000
    standard_deviation = 500
  []
  [L_dist]
    type = Normal
    mean = 0.03
    standard_deviation = 0.01
  []
  [Tinf_dist]
    type = Normal
    mean = 300
    standard_deviation = 10
  []
[]

[Samplers]
  [pc_sampler]
    type = Quadrature
    order = 8
    distributions = 'k_dist q_dist L_dist Tinf_dist'
    execute_on = PRE_MULTIAPP_SETUP
  []
  [pr_sampler]
    type = LatinHypercube
    distributions = 'k_dist q_dist L_dist Tinf_dist'
    num_rows = 6560
    execute_on = PRE_MULTIAPP_SETUP
  []
[]

[MultiApps]
  [pc_sub]
    type = SamplerFullSolveMultiApp
    input_files = sub.i
    sampler = pc_sampler
  []
  [pr_sub]
    type = SamplerFullSolveMultiApp
    input_files = sub.i
    sampler = pr_sampler
  []
[]

[Controls]
  [pc_cmdline]
    type = MultiAppSamplerControl
    multi_app = pc_sub
    sampler = pc_sampler
    param_names = 'Materials/conductivity/prop_values Kernels/source/value Mesh/xmax BCs/right/value'
  []
  [pr_cmdline]
    type = MultiAppSamplerControl
    multi_app = pr_sub
    sampler = pr_sampler
    param_names = 'Materials/conductivity/prop_values Kernels/source/value Mesh/xmax BCs/right/value'
  []
[]

[Transfers]
  [pc_data]
    type = SamplerReporterTransfer
    from_multi_app = pc_sub
    sampler = pc_sampler
    stochastic_reporter = results
    from_reporter = 'max/value'
  []
  [pr_data]
    type = SamplerReporterTransfer
    from_multi_app = pr_sub
    sampler = pr_sampler
    stochastic_reporter = results
    from_reporter = 'max/value'
  []
[]

[Reporters]
  [results]
    type = StochasticReporter
  []
[]

[Trainers]
  [pc_max]
    type = PolynomialChaosTrainer
    execute_on = timestep_end
    order = 8
    distributions = 'k_dist q_dist L_dist Tinf_dist'
    sampler = pc_sampler
    response = results/pc_data:max:value
  []
  [pr_max]
    type = PolynomialRegressionTrainer
    execute_on = timestep_end
    regression_type = "ols"
    max_degree = 4
    sampler = pr_sampler
    response = results/pr_data:max:value
    []
[]

[Outputs]
  [pc_out]
    type = SurrogateTrainerOutput
    trainers = 'pc_max'
    execute_on = FINAL
  []
  [pr_out]
    type = SurrogateTrainerOutput
    trainers = 'pr_max'
    execute_on = FINAL
  []
[]

(modules/stochastic_tools/test/tests/surrogates/polynomial_regression/train.i)

[StochasticTools]
[]

[Samplers]
  [sample]
    type = CartesianProduct
    linear_space_items = '0 1 10
                          0 1 10
                          0 1 10'
  []
[]

[VectorPostprocessors]
  [values]
    type = GFunction
    sampler = sample
    q_vector = '0 0 0'
    execute_on = INITIAL
    outputs = none
  []
[]

[Trainers]
  [train]
    type = PolynomialRegressionTrainer
    regression_type = "ols"
    sampler = sample
    response = values/g_values
    max_degree = 3
  []
[]

[Outputs]
  [out]
    type = SurrogateTrainerOutput
    trainers = 'train'
    execute_on = FINAL
  []
[]

(modules/stochastic_tools/test/tests/surrogates/libtorch_nn/train.i)

[StochasticTools]
[]

[Samplers]
  [sample]
    type = CartesianProduct
    linear_space_items = '0 0.0125 5
                          0 0.0125 5
                          0 0.0125 5'
  []
[]

[VectorPostprocessors]
  [values]
    type = GFunction
    sampler = sample
    q_vector = '0 0 0'
    execute_on = INITIAL
    outputs = none
  []
[]

[Trainers]
  [train]
    type = LibtorchANNTrainer
    sampler = sample
    response = values/g_values
    num_epochs = 40
    num_batches = 10
    num_neurons_per_layer = '64 32'
    learning_rate = 0.001
    nn_filename = mynet.pt
    read_from_file = false
    print_epoch_loss = 10
    activation_function = 'relu relu'
    max_processes = 1
    standardize_input = false
    standardize_output = false
  []
[]

[Outputs]
  [out]
    type = SurrogateTrainerOutput
    trainers = 'train'
    execute_on = FINAL
  []
[]

(modules/stochastic_tools/test/tests/userobjects/inverse_mapping/create_mapping_main.i)

[StochasticTools]
[]

[Distributions]
  [S_dist]
    type = Uniform
    lower_bound = 0
    upper_bound = 10
  []
  [D_dist]
    type = Uniform
    lower_bound = 0
    upper_bound = 10
  []
[]

[Samplers]
  [sample]
    type = MonteCarlo
    num_rows = 8
    distributions = 'S_dist D_dist'
    execute_on = PRE_MULTIAPP_SETUP
    min_procs_per_row = 2
  []
[]

[MultiApps]
  [worker]
    type = SamplerFullSolveMultiApp
    input_files = sub.i
    sampler = sample
    mode = batch-reset
    min_procs_per_app = 2
  []
[]

[Trainers]
  [polyreg_v]
    type = PolynomialRegressionTrainer
    sampler = sample
    regression_type = ols
    max_degree = 1
    response = reduced_solutions/v_pod_mapping
    response_type = vector_real
    execute_on = FINAL
  []
  [polyreg_v_aux]
    type = PolynomialRegressionTrainer
    sampler = sample
    regression_type = ols
    max_degree = 1
    response = reduced_solutions/v_aux_pod_mapping
    response_type = vector_real
    execute_on = FINAL
  []
[]

[VariableMappings]
  [pod_mapping]
    type = PODMapping
    solution_storage = parallel_storage
    variables = "v v_aux"
    num_modes_to_compute = '8 8'
    extra_slepc_options = "-svd_monitor_all"
  []
[]

[Transfers]
  [param_transfer]
    type = SamplerParameterTransfer
    to_multi_app = worker
    sampler = sample
    parameters = 'Kernels/source_v/value BCs/right_v/value'
  []
  [solution_transfer]
    type = SerializedSolutionTransfer
    parallel_storage = parallel_storage
    from_multi_app = worker
    sampler = sample
    solution_container = solution_storage
    variables = 'v'
    serialize_on_root = false
  []
  [solution_transfer_aux]
    type = SerializedSolutionTransfer
    parallel_storage = parallel_storage
    from_multi_app = worker
    sampler = sample
    solution_container = solution_storage_aux
    variables = 'v_aux'
    serialize_on_root = false
  []
[]

[Controls]
  [cmd_line]
    type = MultiAppSamplerControl
    multi_app = worker
    sampler = sample
    param_names = 'S D'
  []
[]

[Reporters]
  [parallel_storage]
    type = ParallelSolutionStorage
    variables = 'v v_aux'
    outputs = none
  []
  [reduced_solutions]
    type = MappingReporter
    sampler = sample
    parallel_storage = parallel_storage
    mapping = pod_mapping
    variables = "v v_aux"
    execute_on = timestep_end # To make sure the trainer sees the results on FINAL
  []
[]

[Outputs]
  [out]
    type = JSON
    execute_on = FINAL
  []
  [mapping]
    type = MappingOutput
    mappings = pod_mapping
    execute_on = FINAL
  []
  [rom]
    type = SurrogateTrainerOutput
    trainers = 'polyreg_v polyreg_v_aux'
    execute_on = FINAL
  []
[]

(modules/stochastic_tools/examples/surrogates/polynomial_regression/uniform_train.i)

[StochasticTools]
[]

[Distributions]
  [k_dist]
    type = Uniform
    lower_bound = 1
    upper_bound = 10
  []
  [q_dist]
    type = Uniform
    lower_bound = 9000
    upper_bound = 11000
  []
  [L_dist]
    type = Uniform
    lower_bound = 0.01
    upper_bound = 0.05
  []
  [Tinf_dist]
    type = Uniform
    lower_bound = 290
    upper_bound = 310
  []
[]

[Samplers]
  [pc_sampler]
    type = Quadrature
    order = 8
    distributions = 'k_dist q_dist L_dist Tinf_dist'
    execute_on = PRE_MULTIAPP_SETUP
  []
  [pr_sampler]
    type = LatinHypercube
    distributions = 'k_dist q_dist L_dist Tinf_dist'
    num_rows = 6560
    execute_on = PRE_MULTIAPP_SETUP
  []
[]

[MultiApps]
  [pc_sub]
    type = SamplerFullSolveMultiApp
    input_files = sub.i
    sampler = pc_sampler
  []
  [pr_sub]
    type = SamplerFullSolveMultiApp
    input_files = sub.i
    sampler = pr_sampler
  []
[]

[Controls]
  [pc_cmdline]
    type = MultiAppSamplerControl
    multi_app = pc_sub
    sampler = pc_sampler
    param_names = 'Materials/conductivity/prop_values Kernels/source/value Mesh/xmax BCs/right/value'
  []
  [pr_cmdline]
    type = MultiAppSamplerControl
    multi_app = pr_sub
    sampler = pr_sampler
    param_names = 'Materials/conductivity/prop_values Kernels/source/value Mesh/xmax BCs/right/value'
  []
[]

[Transfers]
  [pc_data]
    type = SamplerReporterTransfer
    from_multi_app = pc_sub
    sampler = pc_sampler
    stochastic_reporter = results
    from_reporter = 'max/value'
  []
  [pr_data]
    type = SamplerReporterTransfer
    from_multi_app = pr_sub
    sampler = pr_sampler
    stochastic_reporter = results
    from_reporter = 'max/value'
  []
[]

[Reporters]
  [results]
    type = StochasticReporter
  []
[]

[Trainers]
  [pc_max]
    type = PolynomialChaosTrainer
    execute_on = timestep_end
    order = 8
    distributions = 'k_dist q_dist L_dist Tinf_dist'
    sampler = pc_sampler
    response = results/pc_data:max:value
  []
  [pr_max]
    type = PolynomialRegressionTrainer
    regression_type = "ols"
    execute_on = timestep_end
    max_degree = 4
    sampler = pr_sampler
    response = results/pr_data:max:value
  []
[]

[Outputs]
  [pc_out]
    type = SurrogateTrainerOutput
    trainers = 'pc_max'
    execute_on = FINAL
  []
  [pr_out]
    type = SurrogateTrainerOutput
    trainers = 'pr_max'
    execute_on = FINAL
  []
[]

(modules/stochastic_tools/examples/surrogates/gaussian_process/GP_normal_mc.i)

[StochasticTools]
[]

[Distributions]
  [k_dist]
    type = Uniform
    lower_bound = 0
    upper_bound = 20
  []
  [q_dist]
    type = Uniform
    lower_bound = 7000
    upper_bound = 13000
  []
  [L_dist]
    type = Uniform
    lower_bound = 0.0
    upper_bound = 0.1
  []
  [Tinf_dist]
    type = Uniform
    lower_bound = 270
    upper_bound = 330
  []
[]

[Samplers]
  [sample]
    type = MonteCarlo
    num_rows = 500
    distributions = 'k_dist q_dist L_dist Tinf_dist'
    execute_on = PRE_MULTIAPP_SETUP
  []
[]

[MultiApps]
  [sub]
    type = SamplerFullSolveMultiApp
    input_files = sub.i
    sampler = sample
  []
[]

[Controls]
  [cmdline]
    type = MultiAppSamplerControl
    multi_app = sub
    sampler = sample
    param_names = 'Materials/conductivity/prop_values Kernels/source/value Mesh/xmax BCs/right/value'
  []
[]

[Transfers]
  [data]
    type = SamplerReporterTransfer
    from_multi_app = sub
    sampler = sample
    stochastic_reporter = results
    from_reporter = 'avg/value'
  []
[]

[Reporters]
  [results]
    type = StochasticReporter
  []
[]

[Trainers]
  [GP_avg]
    type = GaussianProcessTrainer
    execute_on = timestep_end
    covariance_function = 'rbf'
    standardize_params = 'true'               #Center and scale the training params
    standardize_data = 'true'                 #Center and scale the training data
    sampler = sample
    response = results/data:avg:value
    tune_parameters = ' signal_variance length_factor'
    tuning_algorithm = 'adam'
    tuning_min = ' 1e-9 1e-3'
    tuning_max = ' 100  100'
    iter_adam = 200
    learning_rate_adam = 0.005
  []
[]

[Covariance]
  [rbf]
    type=SquaredExponentialCovariance
    noise_variance = 1e-3                     #A small amount of noise can help with numerical stability
    signal_variance = 1
    length_factor = '0.038971 0.038971 0.038971 0.038971' #Select a length factor for each parameter
  []
[]

[Outputs]
  file_base = GP_training_normal
  [out]
    type = SurrogateTrainerOutput
    trainers = 'GP_avg'
    execute_on = FINAL
  []
[]

(modules/stochastic_tools/test/tests/surrogates/pod_rb/boundary/trainer.i)

[StochasticTools]
[]

[Distributions]
  [k_dist]
    type = Uniform
    lower_bound = 2.5
    upper_bound = 7.5
  []
  [alpha_dist]
    type = Uniform
    lower_bound = 2.5
    upper_bound = 7.5
  []
  [S_dist]
    type = Uniform
    lower_bound = 2.5
    upper_bound = 7.5
  []
  [Dir_dist]
    type = Uniform
    lower_bound = 0
    upper_bound = 1
  []
[]

[Samplers]
  [sample]
    type = LatinHypercube
    distributions = 'k_dist alpha_dist S_dist Dir_dist'
    num_rows = 5
    execute_on = PRE_MULTIAPP_SETUP
    max_procs_per_row = 1
  []
[]

[MultiApps]
  [sub]
    type = PODFullSolveMultiApp
    input_files = sub.i
    sampler = sample
    trainer_name = 'pod_rb'
    execute_on = 'timestep_begin final'
    max_procs_per_app = 1
  []
[]

[Transfers]
  [param]
    type = SamplerParameterTransfer
    to_multi_app = sub
    sampler = sample
    parameters = 'Materials/k/prop_values Materials/alpha/prop_values Kernels/source/value BCs/left/value'
    execute_on = 'timestep_begin'
    check_multiapp_execute_on = false
  []
  [data]
    type = PODSamplerSolutionTransfer
    from_multi_app = sub
    sampler = sample
    trainer_name = 'pod_rb'
    execute_on = 'timestep_begin'
    check_multiapp_execute_on = false
  []
  [mode]
    type = PODSamplerSolutionTransfer
    to_multi_app = sub
    sampler = sample
    trainer_name = 'pod_rb'
    execute_on = 'final'
    check_multiapp_execute_on = false
  []
  [res]
    type = PODResidualTransfer
    from_multi_app = sub
    sampler = sample
    trainer_name = "pod_rb"
    execute_on = 'final'
    check_multiapp_execute_on = false
  []
[]

[Trainers]
  [pod_rb]
    type = PODReducedBasisTrainer
    var_names = 'u'
    error_res = '1e-9'
    tag_names = 'diff react bodyf dir_src dir_imp'
    tag_types = 'op op src src_dir op_dir'
    execute_on = 'timestep_begin final'
  []
[]

[Outputs]
  [out]
    type = SurrogateTrainerOutput
    trainers = 'pod_rb'
    execute_on = FINAL
  []
[]

(modules/stochastic_tools/test/tests/surrogates/pod_rb/internal/trainer.i)

[StochasticTools]
[]

[Distributions]
  [k_dist]
    type = Uniform
    lower_bound = 2.5
    upper_bound = 7.5
  []
  [alpha_dist]
    type = Uniform
    lower_bound = 2.5
    upper_bound = 7.5
  []
  [S_dist]
    type = Uniform
    lower_bound = 2.5
    upper_bound = 7.5
  []
[]

[Samplers]
  [sample]
    type = LatinHypercube
    distributions = 'k_dist alpha_dist S_dist'
    num_rows = 3
    execute_on = PRE_MULTIAPP_SETUP
    max_procs_per_row = 1
  []
[]

[MultiApps]
  [sub]
    type = PODFullSolveMultiApp
    input_files = sub.i
    sampler = sample
    trainer_name = 'pod_rb'
    execute_on = 'timestep_begin final'
    max_procs_per_app = 1
  []
[]

[Transfers]
  [param]
    type = SamplerParameterTransfer
    to_multi_app = sub
    sampler = sample
    parameters = 'Materials/k/prop_values Materials/alpha/prop_values Kernels/source/value'
    execute_on = 'timestep_begin'
    check_multiapp_execute_on = false
  []
  [snapshots]
    type = PODSamplerSolutionTransfer
    from_multi_app = sub
    sampler = sample
    trainer_name = 'pod_rb'
    execute_on = 'timestep_begin'
    check_multiapp_execute_on = false
  []
  [pod_modes]
    type = PODSamplerSolutionTransfer
    to_multi_app = sub
    sampler = sample
    trainer_name = 'pod_rb'
    execute_on = 'final'
    check_multiapp_execute_on = false
  []
  [res]
    type = PODResidualTransfer
    from_multi_app = sub
    sampler = sample
    trainer_name = "pod_rb"
    execute_on = 'final'
    check_multiapp_execute_on = false
  []
[]

[Trainers]
  [pod_rb]
    type = PODReducedBasisTrainer
    var_names = 'u'
    error_res = '1e-9'
    tag_names = 'diff react bodyf'
    tag_types = 'op op src'
    execute_on = 'timestep_begin final'
  []
[]

[Outputs]
  [out]
    type = SurrogateTrainerOutput
    trainers = 'pod_rb'
    execute_on = FINAL
  []
[]

Overview
Example Input File Syntax
Input Parameters
Input Files