A class used to perform Monte Carlo Sampling with active learning.
virtual Real computeSample(dof_id_type row_index, dof_id_type col_index) override
Return the sample for the given row and column.
const std::vector< bool > & _flag_sample
Flag samples if the surrogate prediction was inadequate.
int _retraining_steps
Number of retraining performed.
const int & _num_samples
Number of samples requested.
std::vector< std::vector< Real > > _inputs_gp_fails
Store the input params for which the GP fails.
std::vector< Distribution const * > _distributions
Storage for distribution objects to be utilized.
const int & _step
Track the current step of the main App.
std::vector< std::vector< Real > > _inputs_sto
Storage for previously accepted samples by the decision reporter system.
bool _is_sampling_completed
True if the sampling is completed.
DIE A HORRIBLE DEATH HERE typedef LIBMESH_DEFAULT_SCALAR_TYPE Real
virtual bool isAdaptiveSamplingCompleted() const override
Returns true if the adaptive sampling is completed.
const InputParameters & parameters() const
static InputParameters validParams()
int _check_step
Ensure that the sampler proceeds in a sequential fashion.
const unsigned int _num_batch
The maximum number of GP fails.
virtual void sampleSetUp(const Sampler::SampleMode mode) override
Gather all the samples.
ActiveLearningMonteCarloSampler(const InputParameters ¶meters)