Dakota
Version 6.20
Explore and Predict with Confidence
|
Performs multilevel-multifidelity Monte Carlo sampling for uncertainty quantification. More...
Public Member Functions | |
NonDMultilevControlVarSampling (ProblemDescDB &problem_db, Model &model) | |
standard constructor More... | |
~NonDMultilevControlVarSampling () | |
destructor | |
Public Member Functions inherited from NonDMultilevelSampling | |
NonDMultilevelSampling (ProblemDescDB &problem_db, Model &model) | |
standard constructor More... | |
~NonDMultilevelSampling () | |
destructor | |
Public Member Functions inherited from NonDHierarchSampling | |
NonDHierarchSampling (ProblemDescDB &problem_db, Model &model) | |
standard constructor More... | |
virtual | ~NonDHierarchSampling () |
destructor (virtual declaration should be redundant with ~Iterator, but this is top of MLMF diamond so doesn't hurt to be explicit) | |
Public Member Functions inherited from NonDEnsembleSampling | |
NonDEnsembleSampling (ProblemDescDB &problem_db, Model &model) | |
standard constructor More... | |
~NonDEnsembleSampling () | |
destructor (virtual declaration should be redundant with ~Iterator, but this is top of MLMF diamond so doesn't hurt to be explicit) | |
bool | resize () |
reinitializes iterator based on new variable size | |
Public Member Functions inherited from NonDSampling | |
NonDSampling (Model &model, const RealMatrix &sample_matrix) | |
alternate constructor for evaluating and computing statistics for the provided set of samples More... | |
~NonDSampling () | |
destructor | |
void | compute_statistics (const RealMatrix &vars_samples, const IntResponseMap &resp_samples) |
For the input sample set, computes mean, standard deviation, and probability/reliability/response levels (aleatory uncertainties) or intervals (epsitemic or mixed uncertainties) | |
void | compute_intervals (RealRealPairArray &extreme_fns) |
called by compute_statistics() to calculate min/max intervals using allResponses | |
void | compute_intervals (const IntResponseMap &samples) |
called by compute_statistics() to calculate extremeValues from samples | |
void | compute_intervals (RealRealPairArray &extreme_fns, const IntResponseMap &samples) |
called by compute_statistics() to calculate min/max intervals using samples | |
void | compute_moments (const RealVectorArray &fn_samples) |
calculates sample moments from a matrix of observations for a set of QoI | |
void | compute_moments (const IntResponseMap &samples) |
calculate sample moments and confidence intervals from a map of response observations | |
void | compute_moments (const IntResponseMap &samples, RealMatrix &moment_stats, RealMatrix &moment_grads, RealMatrix &moment_conf_ints, short moments_type, const StringArray &labels) |
convert IntResponseMap to RealVectorArray and invoke helpers | |
void | compute_moment_gradients (const RealVectorArray &fn_samples, const RealMatrixArray &grad_samples, const RealMatrix &moment_stats, RealMatrix &moment_grads, short moments_type) |
compute moment_grads from function and gradient samples | |
void | compute_moment_confidence_intervals (const RealMatrix &moment_stats, RealMatrix &moment_conf_ints, const SizetArray &sample_counts, short moments_type) |
compute moment confidence intervals from moment values | |
void | archive_moments (size_t inc_id=0) |
archive moment statistics in results DB | |
void | archive_moment_confidence_intervals (size_t inc_id=0) |
archive moment confidence intervals in results DB | |
void | archive_std_regress_coeffs () |
archive standardized regression coefficients in results DB | |
void | archive_extreme_responses (size_t inc_id=0) |
archive extreme values (epistemic result) in results DB | |
void | compute_level_mappings (const IntResponseMap &samples) |
called by compute_statistics() to calculate CDF/CCDF mappings of z to p/beta and of p/beta to z as well as PDFs More... | |
void | print_statistics (std::ostream &s) const |
prints the statistics computed in compute_statistics() | |
void | print_intervals (std::ostream &s) const |
prints the intervals computed in compute_intervals() with default qoi_type and moment_labels | |
void | print_intervals (std::ostream &s, String qoi_type, const StringArray &interval_labels) const |
prints the intervals computed in compute_intervals() | |
void | print_moments (std::ostream &s) const |
prints the moments computed in compute_moments() with default qoi_type and moment_labels | |
void | print_moments (std::ostream &s, String qoi_type, const StringArray &moment_labels) const |
prints the moments computed in compute_moments() | |
void | print_wilks_stastics (std::ostream &s) const |
prints the Wilks stastics | |
void | print_tolerance_intervals_statistics (std::ostream &s) const |
prints the tolerance intervals stastics | |
void | archive_tolerance_intervals (size_t inc_id=0) |
archive the tolerance intervals statistics in results DB | |
void | update_final_statistics () |
update finalStatistics from minValues/maxValues, momentStats, and computedProbLevels/computedRelLevels/computedRespLevels | |
void | transform_samples (Model &src_model, Model &tgt_model, bool x_to_u=true) |
transform allSamples using configuration data from the source and target models | |
void | transform_samples (Pecos::ProbabilityTransformation &nataf, bool x_to_u=true) |
alternate version to transform allSamples. This is needed since random variable distribution parameters are not updated until run time and an imported sample_matrix is typically in x-space. More... | |
void | transform_samples (Pecos::ProbabilityTransformation &nataf, RealMatrix &sample_matrix, bool x_to_u=true) |
transform the specified samples matrix from x to u or u to x, assuming identical view and ids | |
void | transform_samples (Pecos::ProbabilityTransformation &nataf, RealMatrix &sample_matrix, SizetMultiArrayConstView src_cv_ids, SizetMultiArrayConstView tgt_cv_ids, bool x_to_u=true) |
transform the specified samples matrix from x to u or u to x | |
unsigned short | sampling_scheme () const |
return sampleType | |
const String & | random_number_generator () const |
return rngName | |
Public Member Functions inherited from NonD | |
void | requested_levels (const RealVectorArray &req_resp_levels, const RealVectorArray &req_prob_levels, const RealVectorArray &req_rel_levels, const RealVectorArray &req_gen_rel_levels, short resp_lev_tgt, short resp_lev_tgt_reduce, bool cdf_flag, bool pdf_output) |
set requestedRespLevels, requestedProbLevels, requestedRelLevels, requestedGenRelLevels, respLevelTarget, cdfFlag, and pdfOutput (used in combination with alternate ctors) | |
void | print_level_mappings (std::ostream &s) const |
prints the z/p/beta/beta* mappings reflected in {requested,computed}{Resp,Prob,Rel,GenRel}Levels for default qoi_type and qoi_labels | |
void | print_level_mappings (std::ostream &s, String qoi_type, const StringArray &qoi_labels) const |
prints the z/p/beta/beta* mappings reflected in {requested,computed}{Resp,Prob,Rel,GenRel}Levels More... | |
void | print_level_mappings (std::ostream &s, const RealVector &level_maps, bool moment_offset, const String &prepend="") |
print level mapping statistics using optional pre-pend More... | |
bool | pdf_output () const |
get pdfOutput | |
void | pdf_output (bool output) |
set pdfOutput | |
short | final_moments_type () const |
get finalMomentsType | |
void | final_moments_type (short type) |
set finalMomentsType | |
Public Member Functions inherited from Analyzer | |
const VariablesArray & | all_variables () |
return the complete set of evaluated variables | |
const RealMatrix & | all_samples () |
return the complete set of evaluated samples | |
const IntResponseMap & | all_responses () const |
return the complete set of computed responses | |
Public Member Functions inherited from Iterator | |
Iterator (std::shared_ptr< TraitsBase > traits=std::shared_ptr< TraitsBase >(new TraitsBase())) | |
default constructor More... | |
Iterator (ProblemDescDB &problem_db, std::shared_ptr< TraitsBase > traits=std::shared_ptr< TraitsBase >(new TraitsBase())) | |
standard envelope constructor, which constructs its own model(s) More... | |
Iterator (ProblemDescDB &problem_db, Model &model, std::shared_ptr< TraitsBase > traits=std::shared_ptr< TraitsBase >(new TraitsBase())) | |
alternate envelope constructor which uses the ProblemDescDB but accepts a model from a higher level (meta-iterator) context, instead of constructing its own More... | |
Iterator (const String &method_string, Model &model, std::shared_ptr< TraitsBase > traits=std::shared_ptr< TraitsBase >(new TraitsBase())) | |
alternate envelope constructor for instantiations by name without the ProblemDescDB More... | |
Iterator (const Iterator &iterator) | |
copy constructor More... | |
virtual | ~Iterator () |
destructor | |
Iterator | operator= (const Iterator &iterator) |
assignment operator | |
virtual void | derived_free_communicators (ParLevLIter pl_iter) |
derived class contributions to freeing the communicators associated with this Iterator instance | |
virtual void | post_input () |
read tabular data for post-run mode | |
virtual void | reset () |
restore initial state for repeated sub-iterator executions | |
virtual void | nested_variable_mappings (const SizetArray &c_index1, const SizetArray &di_index1, const SizetArray &ds_index1, const SizetArray &dr_index1, const ShortArray &c_target2, const ShortArray &di_target2, const ShortArray &ds_target2, const ShortArray &dr_target2) |
set primaryA{CV,DIV,DRV}MapIndices, secondaryA{CV,DIV,DRV}MapTargets within derived Iterators; supports computation of higher-level sensitivities in nested contexts (e.g., derivatives of statistics w.r.t. inserted design variables) | |
virtual void | initialize_iterator (int job_index) |
used by IteratorScheduler to set the starting data for a run | |
virtual void | pack_parameters_buffer (MPIPackBuffer &send_buffer, int job_index) |
used by IteratorScheduler to pack starting data for an iterator run | |
virtual void | unpack_parameters_buffer (MPIUnpackBuffer &recv_buffer, int job_index) |
used by IteratorScheduler to unpack starting data for an iterator run | |
virtual void | unpack_parameters_initialize (MPIUnpackBuffer &recv_buffer, int job_index) |
used by IteratorScheduler to unpack starting data and initialize an iterator run | |
virtual void | pack_results_buffer (MPIPackBuffer &send_buffer, int job_index) |
used by IteratorScheduler to pack results data from an iterator run | |
virtual void | unpack_results_buffer (MPIUnpackBuffer &recv_buffer, int job_index) |
used by IteratorScheduler to unpack results data from an iterator run | |
virtual void | update_local_results (int job_index) |
used by IteratorScheduler to update local results arrays | |
virtual bool | accepts_multiple_points () const |
indicates if this iterator accepts multiple initial points. Default return is false. Override to return true if appropriate. | |
virtual void | initial_point (const Variables &pt) |
sets the initial point for this iterator (user-functions mode for which Model updating is not used) | |
virtual void | initial_point (const RealVector &pt) |
sets the initial point (active continuous variables) for this iterator (user-functions mode for which Model updating is not used) | |
virtual void | initial_points (const VariablesArray &pts) |
sets the multiple initial points for this iterator. This should only be used if accepts_multiple_points() returns true. | |
virtual void | update_callback_data (const RealVector &cv_initial, const RealVector &cv_lower_bnds, const RealVector &cv_upper_bnds, const RealMatrix &lin_ineq_coeffs, const RealVector &lin_ineq_lb, const RealVector &lin_ineq_ub, const RealMatrix &lin_eq_coeffs, const RealVector &lin_eq_tgt, const RealVector &nln_ineq_lb, const RealVector &nln_ineq_ub, const RealVector &nln_eq_tgt) |
assign variable values and bounds and constraint coefficients and bounds for this iterator (user-functions mode for which iteratedModel is null) | |
virtual const RealMatrix & | callback_linear_ineq_coefficients () const |
return linear constraint coefficients for this iterator (user-functions mode for which iteratedModel is null) | |
virtual const RealVector & | callback_linear_ineq_lower_bounds () const |
return linear constraint lower bounds for this iterator (user-functions mode for which iteratedModel is null) | |
virtual const RealVector & | callback_linear_ineq_upper_bounds () const |
return linear constraint upper bounds for this iterator (user-functions mode for which iteratedModel is null) | |
virtual void | initialize_graphics (int iterator_server_id=1) |
initialize the 2D graphics window and the tabular graphics data More... | |
virtual void | check_sub_iterator_conflict () |
detect any conflicts due to recursive use of the same Fortran solver More... | |
virtual unsigned short | uses_method () const |
return name of any enabling iterator used by this iterator | |
virtual void | method_recourse (unsigned short method_name) |
perform a method switch, if possible, due to a detected conflict with the simultaneous use of method_name at an higher-level | |
virtual void | sampling_increment () |
increment to next in sequence of refinement samples | |
virtual IntIntPair | estimate_partition_bounds () |
estimate the minimum and maximum partition sizes that can be utilized by this Iterator | |
virtual void | declare_sources () |
Declare sources to the evaluations database. | |
void | init_communicators (ParLevLIter pl_iter) |
initialize the communicators associated with this Iterator instance | |
void | set_communicators (ParLevLIter pl_iter) |
set the communicators associated with this Iterator instance | |
void | free_communicators (ParLevLIter pl_iter) |
free the communicators associated with this Iterator instance | |
void | resize_communicators (ParLevLIter pl_iter, bool reinit_comms) |
Resize the communicators. This is called from the letter's resize() | |
void | parallel_configuration_iterator (ParConfigLIter pc_iter) |
set methodPCIter | |
ParConfigLIter | parallel_configuration_iterator () const |
return methodPCIter | |
void | parallel_configuration_iterator_map (std::map< size_t, ParConfigLIter > pci_map) |
set methodPCIterMap | |
std::map< size_t, ParConfigLIter > | parallel_configuration_iterator_map () const |
return methodPCIterMap | |
void | run (ParLevLIter pl_iter) |
invoke set_communicators(pl_iter) prior to run() | |
void | run () |
orchestrate initialize/pre/core/post/finalize phases More... | |
void | assign_rep (std::shared_ptr< Iterator > iterator_rep) |
replaces existing letter with a new one More... | |
void | iterated_model (const Model &model) |
set the iteratedModel (iterators and meta-iterators using a single model instance) | |
Model & | iterated_model () |
return the iteratedModel (iterators & meta-iterators using a single model instance) | |
ProblemDescDB & | problem_description_db () const |
return the problem description database (probDescDB) | |
ParallelLibrary & | parallel_library () const |
return the parallel library (parallelLib) | |
void | method_name (unsigned short m_name) |
set the method name to an enumeration value | |
unsigned short | method_name () const |
return the method name via its native enumeration value | |
void | method_string (const String &m_str) |
set the method name by string | |
String | method_string () const |
return the method name by string | |
String | method_enum_to_string (unsigned short method_enum) const |
convert a method name enumeration value to a string | |
unsigned short | method_string_to_enum (const String &method_str) const |
convert a method name string to an enumeration value | |
String | submethod_enum_to_string (unsigned short submethod_enum) const |
convert a sub-method name enumeration value to a string | |
const String & | method_id () const |
return the method identifier (methodId) | |
int | maximum_evaluation_concurrency () const |
return the maximum evaluation concurrency supported by the iterator | |
void | maximum_evaluation_concurrency (int max_conc) |
set the maximum evaluation concurrency supported by the iterator | |
size_t | maximum_iterations () const |
return the maximum iterations for this iterator | |
void | maximum_iterations (size_t max_iter) |
set the maximum iterations for this iterator | |
void | convergence_tolerance (Real conv_tol) |
set the method convergence tolerance (convergenceTol) | |
Real | convergence_tolerance () const |
return the method convergence tolerance (convergenceTol) | |
void | output_level (short out_lev) |
set the method output level (outputLevel) | |
short | output_level () const |
return the method output level (outputLevel) | |
void | summary_output (bool summary_output_flag) |
Set summary output control; true enables evaluation/results summary. | |
size_t | num_final_solutions () const |
return the number of solutions to retain in best variables/response arrays | |
void | num_final_solutions (size_t num_final) |
set the number of solutions to retain in best variables/response arrays | |
void | active_set (const ActiveSet &set) |
set the default active set (for use with iterators that employ evaluate_parameter_sets()) | |
const ActiveSet & | active_set () const |
return the default active set (used by iterators that employ evaluate_parameter_sets()) | |
void | active_set_request_vector (const ShortArray &asv) |
return the default active set request vector (used by iterators that employ evaluate_parameter_sets()) | |
const ShortArray & | active_set_request_vector () const |
return the default active set request vector (used by iterators that employ evaluate_parameter_sets()) | |
void | active_set_request_values (short asv_val) |
return the default active set request vector (used by iterators that employ evaluate_parameter_sets()) | |
void | sub_iterator_flag (bool si_flag) |
set subIteratorFlag (and update summaryOutputFlag if needed) | |
bool | is_null () const |
function to check iteratorRep (does this envelope contain a letter?) | |
std::shared_ptr< Iterator > | iterator_rep () const |
returns iteratorRep for access to derived class member functions that are not mapped to the top Iterator level | |
virtual void | eval_tag_prefix (const String &eval_id_str) |
set the hierarchical eval ID tag prefix More... | |
std::shared_ptr< TraitsBase > | traits () const |
returns methodTraits for access to derived class member functions that are not mapped to the top TraitsBase level | |
bool | top_level () |
Return whether the iterator is the top level iterator. | |
void | top_level (bool tflag) |
Set the iterator's top level flag. | |
Protected Member Functions | |
void | pre_run () |
pre-run portion of run (optional); re-implemented by Iterators which can generate all Variables (parameter sets) a priori More... | |
void | core_run () |
void | print_variance_reduction (std::ostream &s) |
bool | discrepancy_sample_counts () const |
flag identifying whether sample counts correspond to level discrepancies | |
Protected Member Functions inherited from NonDMultilevelSampling | |
void | core_run () |
core portion of run; implemented by all derived classes and may include pre/post steps in lieu of separate pre/post More... | |
void | print_variance_reduction (std::ostream &s) |
void | nested_response_mappings (const RealMatrix &primary_coeffs, const RealMatrix &secondary_coeffs) |
set primaryResponseCoefficients, secondaryResponseCoefficients within derived Iterators; Necessary for scalarization case in MLMC NonDMultilevelSampling to map scalarization in nested context | |
bool | discrepancy_sample_counts () const |
flag identifying whether sample counts correspond to level discrepancies | |
void | evaluate_ml_sample_increment (String prepend, unsigned short step) |
helper that consolidates sequence advancement, sample generation, sample export, and sample evaluation | |
void | increment_ml_equivalent_cost (size_t new_N_l, Real lev_cost, Real ref_cost, Real &equiv_hf) |
increment the equivalent number of HF evaluations based on new model evaluations | |
void | compute_ml_estimator_variance (const RealMatrix &var_Y, const Sizet2DArray &num_Y, RealVector &ml_est_var) |
compute MLMC estimator variance from level QoI variances | |
void | recover_variance (const RealMatrix &moment_stats, RealVector &var_H) |
recover variance from raw moments | |
void | initialize_ml_Ysums (IntRealMatrixMap &sum_Y, size_t num_lev, size_t num_mom=4) |
initialize the ML accumulators for computing means, variances, and covariances across resolution levels | |
void | initialize_ml_Qsums (IntRealMatrixMap &sum_Ql, IntRealMatrixMap &sum_Qlm1, IntIntPairRealMatrixMap &sum_QlQlm1, size_t num_lev) |
initialize the ML accumulators for computing means, variances, and covariances across resolution levels | |
void | accumulate_ml_Ysums (IntRealMatrixMap &sum_Y, RealMatrix &sum_YY, size_t lev, SizetArray &num_Y) |
update accumulators for multilevel telescoping running sums using set of model evaluations within allResponses | |
void | accumulate_ml_Ysums (RealMatrix &sum_Y, RealMatrix &sum_YY, size_t lev, SizetArray &num_Y) |
update accumulators for multilevel telescoping running sums using set of model evaluations within allResponses | |
void | accumulate_ml_Qsums (IntRealMatrixMap &sum_Q, size_t lev, SizetArray &num_Q) |
update running QoI sums for one model (sum_Q) using set of model evaluations within allResponses; used for level 0 from other accumulators | |
Real | variance_Ysum (Real sum_Y, Real sum_YY, size_t Nlq) |
compute variance scalar from sum accumulators | |
void | variance_Ysum (const Real *sum_Y, const Real *sum_YY, const SizetArray &N_l, Real *var_Y) |
compute variance column vec (all QoI for one level) from sum accumulators | |
Real | variance_Qsum (Real sum_Ql, Real sum_Qlm1, Real sum_QlQl, Real sum_QlQlm1, Real sum_Qlm1Qlm1, size_t Nlq) |
compute variance from sum accumulators | |
Real | aggregate_variance_Ysum (const Real *sum_Y, const Real *sum_YY, const SizetArray &N_l) |
sum up variances across QoI (using sum_YY with means from sum_Y) | |
Real | aggregate_mse_Yvar (const Real *var_Y, const SizetArray &N_l) |
sum up Monte Carlo estimates for mean squared error (MSE) across QoI using discrepancy variances | |
Real | aggregate_mse_Ysum (const Real *sum_Y, const Real *sum_YY, const SizetArray &N_l) |
sum up Monte Carlo estimates for mean squared error (MSE) across QoI using discrepancy sums | |
void | ml_Q_raw_moments (const RealMatrix &sum_Hl_1, const RealMatrix &sum_Hlm1_1, const RealMatrix &sum_Hl_2, const RealMatrix &sum_Hlm1_2, const Sizet2DArray &N_actual, size_t start, size_t end, RealMatrix &ml_raw_mom) |
accumulate ML-only contributions (levels with no CV) to 2 raw moments | |
void | ml_Y_raw_moments (const RealMatrix &sum_H1, const RealMatrix &sum_H2, const Sizet2DArray &N_l, size_t start, size_t end, RealMatrix &ml_raw_mom) |
accumulate ML-only contributions (levels with no CV) to 2 raw moments | |
void | ml_Y_raw_moments (const RealMatrix &sum_H1, const RealMatrix &sum_H2, const RealMatrix &sum_H3, const RealMatrix &sum_H4, const Sizet2DArray &N_l, size_t start, size_t end, RealMatrix &ml_raw_mom) |
accumulate ML-only contributions (levels with no CV) to 4 raw moments | |
void | configure_indices (unsigned short group, unsigned short form, size_t lev, short seq_type) |
manage response mode and active model key from {group,form,lev} triplet. seq_type defines the active dimension for a 1D model sequence. | |
void | configure_indices (size_t group, size_t form, size_t lev, short seq_type) |
convert group and form and call overload | |
Real | level_cost (const RealVector &cost, size_t step) |
return (aggregate) level cost | |
Protected Member Functions inherited from NonDHierarchSampling | |
Real | estimator_accuracy_metric () |
void | average_online_cost (const RealVector &accum_cost, const SizetArray &num_cost, RealVector &seq_cost) |
average costs once accumulations are complete | |
void | accumulate_paired_online_cost (RealVector &accum_cost, SizetArray &num_cost, size_t step) |
recover partial estimates of simulation cost using aggregated (paired) response metadata | |
void | recover_paired_online_cost (RealVector &seq_cost, size_t step) |
accumulate cost and counts and then perform averaging | |
Protected Member Functions inherited from NonDEnsembleSampling | |
virtual void | print_multimodel_summary (std::ostream &s, const String &summary_type, bool projections) |
virtual void | print_multigroup_summary (std::ostream &s, const String &summary_type, bool projections) |
void | post_run (std::ostream &s) |
post-run portion of run (optional); verbose to print results; re-implemented by Iterators that can read all Variables/Responses and perform final analysis phase in a standalone way More... | |
void | print_results (std::ostream &s, short results_state=FINAL_RESULTS) |
print the final iterator results More... | |
void | initialize_final_statistics () |
initializes finalStatistics for storing NonD final results More... | |
void | update_final_statistics () |
update finalStatistics::functionValues | |
bool | seed_updated () |
void | active_set_mapping () |
in the case of sub-iteration, map from finalStatistics.active_set() requests to activeSet used in evaluate_parameter_sets() More... | |
Real | estimator_cost_metric () |
return cost metric for entry into finalStatistics | |
void | assign_specification_sequence (size_t index) |
advance any sequence specifications | |
int | seed_sequence (size_t index) |
extract current random seed from randomSeedSeqSpec More... | |
void | resize_active_set () |
synchronize activeSet with iteratedModel's response size | |
void | increment_samples (SizetArray &N_l, size_t incr) |
increment samples array with a shared scalar | |
void | increment_samples (SizetArray &N_l, const SizetArray &incr) |
increment samples array with a shared scalar | |
void | increment_samples (Sizet2DArray &N_l, const SizetArray &incr) |
increment 2D samples array with a shared 1D array (additional dim is QoI) | |
void | increment_sums (Real *sum_l, const Real *incr, size_t len) |
increment samples array with a shared scalar | |
void | compute_mc_estimator_variance (const RealVector &var_l, const SizetArray &N_l, RealVector &mc_est_var) |
compute the variance of the mean estimator (Monte Carlo sample average) | |
void | project_mc_estimator_variance (const RealVector &var_l, const SizetArray &N_l, size_t new_samp, RealVector &mc_est_var) |
compute the variance of the mean estimator (Monte Carlo sample average) after projection with additional samples (var_l remains fixed) | |
Real | estvar_ratios_to_avg_estvar (const RealVector &estvar_ratios, const RealVector &var_H, const SizetArray &N_H) |
convert estimator variance ratios to average estimator variance | |
void | reset_relaxation () |
initialize relaxFactor prior to iteration | |
void | advance_relaxation () |
update relaxFactor based on iteration number | |
void | compute_mf_control (Real sum_L, Real sum_H, Real sum_LL, Real sum_LH, size_t N_shared, Real &beta) |
compute scalar control variate parameters | |
void | compute_mf_control (const RealMatrix &sum_L, const RealMatrix &sum_H, const RealMatrix &sum_LL, const RealMatrix &sum_LH, const SizetArray &N_shared, size_t lev, RealVector &beta) |
compute matrix control variate parameters | |
void | compute_mf_control (const RealVector &sum_L, const RealVector &sum_H, const RealVector &sum_LL, const RealVector &sum_LH, const SizetArray &N_shared, RealVector &beta) |
compute vector control variate parameters | |
void | export_all_samples (String root_prepend, const Model &model, size_t iter, size_t step) |
export allSamples to tagged tabular file | |
void | convert_moments (const RealMatrix &raw_mom, RealMatrix &final_mom) |
convert uncentered raw moments (multilevel expectations) to standardized moments | |
void | uncentered_to_centered (const RealMatrix &raw_mom, RealMatrix ¢_mom) |
convert uncentered (raw) moments to centered moments; biased estimators More... | |
void | centered_to_standard (const RealMatrix ¢_mom, RealMatrix &std_mom) |
convert centered moments to standardized moments More... | |
Protected Member Functions inherited from NonDSampling | |
NonDSampling (ProblemDescDB &problem_db, Model &model) | |
constructor More... | |
NonDSampling (unsigned short method_name, Model &model, unsigned short sample_type, size_t samples, int seed, const String &rng, bool vary_pattern, short sampling_vars_mode) | |
alternate constructor for sample generation and evaluation "on the fly" More... | |
NonDSampling (unsigned short sample_type, size_t samples, int seed, const String &rng, const RealVector &lower_bnds, const RealVector &upper_bnds) | |
alternate constructor for sample generation "on the fly" More... | |
NonDSampling (unsigned short sample_type, size_t samples, int seed, const String &rng, const RealVector &means, const RealVector &std_devs, const RealVector &lower_bnds, const RealVector &upper_bnds, RealSymMatrix &correl) | |
alternate constructor for sample generation of correlated normals "on the fly" More... | |
void | pre_run () |
pre-run portion of run (optional); re-implemented by Iterators which can generate all Variables (parameter sets) a priori More... | |
void | core_run () |
size_t | num_samples () const |
void | sampling_reset (size_t min_samples, bool all_data_flag, bool stats_flag) |
resets number of samples and sampling flags More... | |
void | sampling_reference (size_t samples_ref) |
set reference number of samples, which is a lower bound during reset | |
void | random_seed (int seed) |
assign randomSeed | |
void | vary_pattern (bool pattern_flag) |
set varyPattern | |
void | get_parameter_sets (Model &model) |
Uses lhsDriver to generate a set of samples from the distributions/bounds defined in the incoming model. More... | |
void | get_parameter_sets (Model &model, const size_t num_samples, RealMatrix &design_matrix) |
Uses lhsDriver to generate a set of samples from the distributions/bounds defined in the incoming model and populates the specified design matrix. More... | |
void | get_parameter_sets (Model &model, const size_t num_samples, RealMatrix &design_matrix, bool write_msg) |
core of get_parameter_sets that accepts message print control | |
void | get_parameter_sets (const RealVector &lower_bnds, const RealVector &upper_bnds) |
Uses lhsDriver to generate a set of uniform samples over lower_bnds/upper_bnds. More... | |
void | get_parameter_sets (const RealVector &means, const RealVector &std_devs, const RealVector &lower_bnds, const RealVector &upper_bnds, RealSymMatrix &correl) |
Uses lhsDriver to generate a set of normal samples. More... | |
void | update_model_from_sample (Model &model, const Real *sample_vars) |
Override default update of continuous vars only. | |
void | sample_to_variables (const Real *sample_vars, Variables &vars) |
override default mapping of continuous variables only | |
void | variables_to_sample (const Variables &vars, Real *sample_vars) |
override default mapping of continuous variables only | |
const RealSymMatrix & | response_error_estimates () const |
return error estimates associated with each of the finalStatistics | |
void | initialize_sample_driver (bool write_message, size_t num_samples) |
increments numLHSRuns, sets random seed, and initializes lhsDriver | |
void | mode_counts (const Variables &vars, size_t &cv_start, size_t &num_cv, size_t &div_start, size_t &num_div, size_t &dsv_start, size_t &num_dsv, size_t &drv_start, size_t &num_drv) const |
compute sampled subsets (all, active, uncertain) within all variables (acv/adiv/adrv) from samplingVarsMode and model More... | |
void | mode_bits (const Variables &vars, BitArray &active_vars, BitArray &active_corr) const |
define subset views for sampling modes | |
Protected Member Functions inherited from NonD | |
NonD (ProblemDescDB &problem_db, Model &model) | |
constructor | |
NonD (unsigned short method_name, Model &model) | |
alternate constructor for sample generation and evaluation "on the fly" | |
NonD (unsigned short method_name, Model &model, const ShortShortPair &approx_view) | |
alternate constructor for sample generation and evaluation "on the fly" | |
NonD (unsigned short method_name, const RealVector &lower_bnds, const RealVector &upper_bnds) | |
alternate constructor for sample generation "on the fly" | |
~NonD () | |
destructor | |
void | derived_set_communicators (ParLevLIter pl_iter) |
derived class contributions to setting the communicators associated with this Iterator instance | |
void | initialize_run () |
utility function to perform common operations prior to pre_run(); typically memory initialization; setting of instance pointers More... | |
void | finalize_run () |
utility function to perform common operations following post_run(); deallocation and resetting of instance pointers More... | |
const Response & | response_results () const |
return the final statistics from the nondeterministic iteration | |
void | response_results_active_set (const ActiveSet &set) |
set the active set within finalStatistics | |
virtual void | initialize_response_covariance () |
initializes respCovariance | |
void | pull_level_mappings (RealVector &level_maps, size_t offset) |
concatenate computed{Resp,Prob,Rel,GenRel}Levels into level_maps | |
void | push_level_mappings (const RealVector &level_maps, size_t offset) |
update computed{Resp,Prob,Rel,GenRel}Levels from level_maps | |
void | configure_sequence (size_t &num_steps, size_t &secondary_index, short &seq_type) |
configure fidelity/level counts from model hierarchy More... | |
void | configure_enumeration (size_t &num_combinations) |
configure the total number of fidelity/level options More... | |
void | configure_cost (unsigned short num_steps, short seq_type, RealVector &cost) |
extract cost estimates from model hierarchy (forms or resolutions) | |
bool | query_cost (unsigned short num_steps, short seq_type, RealVector &cost) |
extract cost estimates from model hierarchy, if available | |
bool | query_cost (unsigned short num_steps, Model &model, RealVector &cost) |
extract cost estimates from model hierarchy, if available | |
bool | valid_cost_values (const RealVector &cost) |
test cost for valid values > 0 | |
void | load_pilot_sample (const SizetArray &pilot_spec, size_t num_steps, SizetArray &delta_N_l) |
distribute pilot sample specification across model forms or levels | |
void | load_pilot_sample (const SizetArray &pilot_spec, short seq_type, const Sizet3DArray &N_l, Sizet2DArray &delta_N_l) |
distribute pilot sample specification across model forms and levels | |
template<typename ArrayType > | |
void | inflate_approx_samples (const ArrayType &N_l, bool multilev, size_t secondary_index, std::vector< ArrayType > &N_l_vec) |
update the relevant slice of N_l_3D from the final 2D multilevel or 2D multifidelity sample profile | |
template<typename ArrayType > | |
void | inflate_sequence_samples (const ArrayType &N_l, bool multilev, size_t secondary_index, std::vector< ArrayType > &N_l_vec) |
update the relevant slice of N_l_3D from the final 2D multilevel or 2D multifidelity sample profile | |
void | resize_final_statistics_gradients () |
resizes finalStatistics::functionGradients based on finalStatistics ASV | |
void | update_aleatory_final_statistics () |
update finalStatistics::functionValues from momentStats and computed{Prob,Rel,GenRel,Resp}Levels | |
void | update_system_final_statistics () |
update system metrics from component metrics within finalStatistics | |
void | update_system_final_statistics_gradients () |
update finalStatistics::functionGradients | |
void | initialize_level_mappings () |
size computed{Resp,Prob,Rel,GenRel}Levels | |
void | compute_densities (const RealRealPairArray &min_max_fns, bool prob_refinement=false, bool all_levels_computed=false) |
compute the PDF bins from the CDF/CCDF values and store in computedPDF{Abscissas,Ordinates} More... | |
void | print_densities (std::ostream &s) const |
output the PDFs reflected in computedPDF{Abscissas,Ordinates} using default qoi_type and pdf_labels | |
void | print_densities (std::ostream &s, String qoi_type, const StringArray &pdf_labels) const |
output the PDFs reflected in computedPDF{Abscissas,Ordinates} | |
void | print_system_mappings (std::ostream &s) const |
print system series/parallel mappings for response levels | |
void | print_multilevel_evaluation_summary (std::ostream &s, const SizetArray &N_m) |
print evaluation summary for multilevel sampling across 1D level profile | |
void | print_multilevel_evaluation_summary (std::ostream &s, const Sizet2DArray &N_m) |
print evaluation summary for multilevel sampling across 2D level+QoI profile | |
void | print_multilevel_discrepancy_summary (std::ostream &s, const SizetArray &N_m) |
print evaluation summary for multilevel sampling across 1D level profile for discrepancy across levels | |
void | print_multilevel_discrepancy_summary (std::ostream &s, const SizetArray &N_m, const SizetArray &N_mp1) |
print evaluation summary for multilevel sampling across 1D level profile for discrepancy across model forms | |
void | print_multilevel_discrepancy_summary (std::ostream &s, const Sizet2DArray &N_m) |
print evaluation summary for multilevel sampling across 2D level+QoI profile for discrepancy across levels | |
void | print_multilevel_discrepancy_summary (std::ostream &s, const Sizet2DArray &N_m, const Sizet2DArray &N_mp1) |
print evaluation summary for multilevel sampling across 2D level+QoI profile for discrepancy across model forms | |
template<typename ArrayType > | |
void | print_multilevel_model_summary (std::ostream &s, const std::vector< ArrayType > &N_samp, String type, short seq_type, bool discrep_flag) |
print evaluation summary for multilevel sampling across 2D model+level profile (allocations) or 3D model+level+QoI profile (actual) | |
void | construct_lhs (Iterator &u_space_sampler, Model &u_model, unsigned short sample_type, int num_samples, int seed, const String &rng, bool vary_pattern, short sampling_vars_mode=ACTIVE) |
assign a NonDLHSSampling instance within u_space_sampler | |
unsigned short | sub_optimizer_select (unsigned short requested_sub_method, unsigned short default_sub_method=SUBMETHOD_NPSOL) |
utility for vetting sub-method request against optimizers within the package configuration | |
size_t | one_sided_relax_round (Real diff, Real relax_factor=1.) |
compute a one-sided sample increment for multilevel methods to move current sampling level to a new target | |
size_t | one_sided_delta (Real current, Real target, Real relax_factor=1.) |
compute a one-sided sample increment for multilevel methods to move current sampling level to a new target | |
size_t | one_sided_delta (const SizetArray ¤t, const RealVector &targets, Real relax_factor=1., size_t power=1) |
compute a one-sided sample increment for multilevel methods to move current sampling level to a new target | |
size_t | one_sided_delta (const SizetArray ¤t, Real target, Real relax_factor=1., size_t power=1) |
compute a one-sided sample increment for multilevel methods to move current sampling level to a new target | |
void | one_sided_delta (const SizetArray ¤t, const RealVector &targets, SizetArray &delta_N, Real relax_factor=1.) |
compute a one-sided sample increment vector to move current sampling levels to new targets | |
void | one_sided_delta (const Sizet2DArray ¤t, const RealVector &targets, SizetArray &delta_N, Real relax_factor=1.) |
compute a one-sided sample increment vector to move current sampling levels to new targets | |
bool | differ (size_t N_alloc_ij, const SizetArray &N_actual_ij) const |
return true if fine-grained reporting differs from coarse-grained | |
bool | differ (const SizetArray &N_alloc_i, const Sizet2DArray &N_actual_i) const |
return true if fine-grained reporting differs from coarse-grained | |
bool | differ (const Sizet2DArray &N_alloc, const Sizet3DArray &N_actual) const |
return true if fine-grained reporting differs from coarse-grained | |
void | archive_allocate_mappings () |
allocate results array storage for distribution mappings | |
void | archive_from_resp (size_t fn_index, size_t inc_id=0) |
archive the mappings from specified response levels for specified fn | |
void | archive_to_resp (size_t fn_index, size_t inc_id=0) |
archive the mappings to computed response levels for specified fn and (optional) increment id. | |
void | archive_allocate_pdf () |
allocate results array storage for pdf histograms | |
void | archive_pdf (size_t fn_index, size_t inc_id=0) |
archive a single pdf histogram for specified function | |
void | archive_equiv_hf_evals (const Real equiv_hf_evals) |
archive the equivalent number of HF evals (used by ML/MF methods) | |
bool | zeros (const SizetArray &N_m) const |
return true if N_m is empty or only populated with zeros | |
bool | zeros (const Sizet2DArray &N_m) const |
return true if N_m is empty or only populated with zeros | |
bool | zeros (const SizetVector &N_m) const |
return true if N_m is empty or only populated with zeros | |
bool | homogeneous (const SizetArray &N_l) const |
return true if N_l has consistent values | |
Protected Member Functions inherited from Analyzer | |
Analyzer () | |
default constructor | |
Analyzer (ProblemDescDB &problem_db, Model &model) | |
standard constructor | |
Analyzer (unsigned short method_name, Model &model) | |
alternate constructor for instantiations "on the fly" with a Model | |
Analyzer (unsigned short method_name, Model &model, const ShortShortPair &view_override) | |
alternate constructor for instantiations "on the fly" with a Model | |
Analyzer (unsigned short method_name) | |
alternate constructor for instantiations "on the fly" without a Model | |
~Analyzer () | |
destructor | |
virtual void | update_model_from_variables (Model &model, const Variables &vars) |
update model's current variables with data from vars | |
void | update_from_model (const Model &model) |
set inherited data attributes based on extractions from incoming model | |
void | pre_output () |
const Model & | algorithm_space_model () const |
const Variables & | variables_results () const |
return a single final iterator solution (variables) | |
const VariablesArray & | variables_array_results () |
return multiple final iterator solutions (variables). This should only be used if returns_multiple_points() returns true. | |
const ResponseArray & | response_array_results () |
return multiple final iterator solutions (response). This should only be used if returns_multiple_points() returns true. | |
bool | compact_mode () const |
returns Analyzer::compactMode | |
bool | returns_multiple_points () const |
indicates if this iterator returns multiple final points. Default return is false. Override to return true if appropriate. | |
void | evaluate_parameter_sets (Model &model, bool log_resp_flag=true, bool log_best_flag=false) |
perform function evaluations to map parameter sets (allVariables) into response sets (allResponses) More... | |
void | evaluate_batch (Model &model, int batch_id, bool log_best_flag=false) |
perform function evaluations to map a keyed batch of parameter sets (allVariablesMap[key]) into a corresponding batch of response sets (allResponsesMap[key]) | |
const IntResponse2DMap & | synchronize_batches (Model &model, bool log_best_flag=false) |
perform function evaluations to map a keyed batch of parameter sets (allVariablesMap[key]) into a corresponding batch of response sets (allResponsesMap[key]) | |
void | clear_batches () |
since synchronize returns the aggregation of all evaluated batches, we use a separate call to indicate when processing of this data is complete and bookkeeping can be cleared | |
void | get_vbd_parameter_sets (Model &model, size_t num_samples) |
generate replicate parameter sets for use in variance-based decomposition More... | |
virtual void | archive_model_variables (const Model &, size_t idx) const |
archive model evaluation points | |
virtual void | archive_model_response (const Response &, size_t idx) const |
archive model evaluation responses | |
void | read_variables_responses (int num_evals, size_t num_vars) |
convenience function for reading variables/responses (used in derived classes post_input) More... | |
void | samples_to_variables_array (const RealMatrix &sample_matrix, VariablesArray &vars_array) |
convert samples array to variables array; e.g., allSamples to allVariables | |
void | variables_array_to_samples (const VariablesArray &vars_array, RealMatrix &sample_matrix) |
convert variables array to samples array; e.g., allVariables to allSamples | |
Protected Member Functions inherited from Iterator | |
Iterator (BaseConstructor, ProblemDescDB &problem_db, std::shared_ptr< TraitsBase > traits=std::shared_ptr< TraitsBase >(new TraitsBase())) | |
constructor initializes the base class part of letter classes (BaseConstructor overloading avoids infinite recursion in the derived class constructors - Coplien, p. 139) More... | |
Iterator (NoDBBaseConstructor, unsigned short method_name, Model &model, std::shared_ptr< TraitsBase > traits=std::shared_ptr< TraitsBase >(new TraitsBase())) | |
alternate constructor for base iterator classes constructed on the fly More... | |
Iterator (NoDBBaseConstructor, unsigned short method_name, std::shared_ptr< TraitsBase > traits=std::shared_ptr< TraitsBase >(new TraitsBase())) | |
alternate constructor for base iterator classes constructed on the fly More... | |
Iterator (NoDBBaseConstructor, Model &model, size_t max_iter, size_t max_eval, Real conv_tol, std::shared_ptr< TraitsBase > traits=std::shared_ptr< TraitsBase >(new TraitsBase())) | |
alternate envelope constructor for instantiations without ProblemDescDB More... | |
virtual void | derived_init_communicators (ParLevLIter pl_iter) |
derived class contributions to initializing the communicators associated with this Iterator instance | |
virtual const VariablesArray & | initial_points () const |
gets the multiple initial points for this iterator. This will only be meaningful after a call to initial_points mutator. | |
StrStrSizet | run_identifier () const |
get the unique run identifier based on method name, id, and number of executions | |
void | initialize_model_graphics (Model &model, int iterator_server_id) |
helper function that encapsulates initialization operations, modular on incoming Model instance More... | |
void | export_final_surrogates (Model &data_fit_surr_model) |
export final surrogates generated, e.g., GP in EGO and friends More... | |
Private Member Functions | |
void | multilevel_control_variate_mc_online_pilot () |
Perform multilevel Monte Carlo across levels in combination with control variate Monte Carlo across model forms at each level; CV computes correlations for Q (LH correlations for QoI) More... | |
void | multilevel_control_variate_mc_offline_pilot () |
Qcorr approach using a pilot sample treated as separate offline cost. | |
void | multilevel_control_variate_mc_pilot_projection () |
Qcorr approach projecting estimator performance from a pilot sample. | |
void | evaluate_pilot (RealVector &hf_cost, RealVector &lf_cost, RealVectorArray &eval_ratios, RealMatrix &Lambda, RealMatrix &var_YH, SizetArray &N_alloc, Sizet2DArray &N_actual, RealVector &hf_targets, RealMatrix &pilot_mom, bool accumulate_cost, bool pilot_estvar) |
helper for shared code among offline-pilot and pilot-projection modes | |
bool | lf_increment (const RealVector &eval_ratios, const SizetArray &N_lf, Real hf_target, RealVector &lf_targets, size_t iter, size_t lev) |
perform LF sample increment as indicated by evaluation ratios | |
bool | lf_increment (const RealVector &eval_ratios, size_t N_lf, Real hf_target, Real &lf_target, size_t iter, size_t lev) |
perform LF sample increment as indicated by evaluation ratios | |
bool | lf_perform_samples (size_t iter, size_t lev) |
parameter set definition and evaluation for LF sample increment | |
Real | compute_mlmf_equivalent_cost (const SizetArray &raw_N_hf, const RealVector &hf_cost, const SizetArray &raw_N_lf, const RealVector &lf_cost) |
compute the equivalent number of HF evaluations (includes any sim faults) | |
void | increment_mlmf_equivalent_cost (size_t new_N_hf, Real hf_lev_cost, size_t new_N_lf, Real lf_lev_cost, Real hf_ref_cost, Real &equiv_hf) |
increment the equivalent number of HF evaluations | |
void | compute_mlmf_estimator_variance (const RealMatrix &var_Y, const Sizet2DArray &num_Y, const RealMatrix &Lambda, RealVector &mlmf_est_var) |
compute the variance of the MLMF estimator | |
void | compute_eval_ratios (RealMatrix &sum_L_shared, RealMatrix &sum_H, RealMatrix &sum_LL, RealMatrix &sum_LH, RealMatrix &sum_HH, Real cost_ratio, size_t lev, const SizetArray &N_shared, RealMatrix &var_H, RealMatrix &rho2_LH, RealVector &eval_ratios) |
compute the LF/HF evaluation ratio, averaged over the QoI | |
void | compute_eval_ratios (RealMatrix &sum_Ll, RealMatrix &sum_Llm1, RealMatrix &sum_Hl, RealMatrix &sum_Hlm1, RealMatrix &sum_Ll_Ll, RealMatrix &sum_Ll_Llm1, RealMatrix &sum_Llm1_Llm1, RealMatrix &sum_Hl_Ll, RealMatrix &sum_Hl_Llm1, RealMatrix &sum_Hlm1_Ll, RealMatrix &sum_Hlm1_Llm1, RealMatrix &sum_Hl_Hl, RealMatrix &sum_Hl_Hlm1, RealMatrix &sum_Hlm1_Hlm1, Real cost_ratio, size_t lev, const SizetArray &N_shared, RealMatrix &var_YHl, RealMatrix &rho_dot2_LH, RealVector &eval_ratios) |
compute the LF/HF evaluation ratio, averaged over the QoI | |
void | compute_mf_correlation (Real sum_L, Real sum_H, Real sum_LL, Real sum_LH, Real sum_HH, size_t N_shared, Real &var_H, Real &rho2_LH) |
compute scalar variance and correlation parameters for control variates | |
void | cv_raw_moments (IntRealMatrixMap &sum_L_shared, IntRealMatrixMap &sum_H, IntRealMatrixMap &sum_LL, IntRealMatrixMap &sum_LH, const SizetArray &N_shared, IntRealMatrixMap &sum_L_refined, const SizetArray &N_refined, size_t lev, RealMatrix &H_raw_mom) |
apply control variate parameters for MLMF MC to estimate raw moment contributions | |
void | cv_raw_moments (IntRealMatrixMap &sum_Ll, IntRealMatrixMap &sum_Llm1, IntRealMatrixMap &sum_Hl, IntRealMatrixMap &sum_Hlm1, IntRealMatrixMap &sum_Ll_Ll, IntRealMatrixMap &sum_Ll_Llm1, IntRealMatrixMap &sum_Llm1_Llm1, IntRealMatrixMap &sum_Hl_Ll, IntRealMatrixMap &sum_Hl_Llm1, IntRealMatrixMap &sum_Hlm1_Ll, IntRealMatrixMap &sum_Hlm1_Llm1, IntRealMatrixMap &sum_Hl_Hl, IntRealMatrixMap &sum_Hl_Hlm1, IntRealMatrixMap &sum_Hlm1_Hlm1, const SizetArray &N_shared, IntRealMatrixMap &sum_Ll_refined, IntRealMatrixMap &sum_Llm1_refined, const SizetArray &N_refined, size_t lev, RealMatrix &H_raw_mom) |
apply control variate parameters for MLMF MC to estimate raw moment contributions | |
void | compute_mlmf_control (Real sum_Ll, Real sum_Llm1, Real sum_Hl, Real sum_Hlm1, Real sum_Ll_Ll, Real sum_Ll_Llm1, Real sum_Llm1_Llm1, Real sum_Hl_Ll, Real sum_Hl_Llm1, Real sum_Hlm1_Ll, Real sum_Hlm1_Llm1, Real sum_Hl_Hl, Real sum_Hl_Hlm1, Real sum_Hlm1_Hlm1, size_t N_shared, Real &var_YHl, Real &rho_dot2_LH, Real &beta_dot, Real &gamma) |
compute scalar control variate parameters | |
void | compute_mlmf_control (const RealMatrix &sum_Ll, const RealMatrix &sum_Llm1, const RealMatrix &sum_Hl, const RealMatrix &sum_Hlm1, const RealMatrix &sum_Ll_Ll, const RealMatrix &sum_Ll_Llm1, const RealMatrix &sum_Llm1_Llm1, const RealMatrix &sum_Hl_Ll, const RealMatrix &sum_Hl_Llm1, const RealMatrix &sum_Hlm1_Ll, const RealMatrix &sum_Hlm1_Llm1, const RealMatrix &sum_Hl_Hl, const RealMatrix &sum_Hl_Hlm1, const RealMatrix &sum_Hlm1_Hlm1, const SizetArray &N_shared, size_t lev, RealVector &beta_dot, RealVector &gamma) |
compute matrix control variate parameters | |
void | apply_mf_control (Real sum_H, Real sum_L_shared, size_t N_shared, Real sum_L_refined, size_t N_refined, Real beta, Real &H_raw_mom) |
apply scalar control variate parameter (beta) to approximate HF moment | |
void | apply_mf_control (const RealMatrix &sum_H, const RealMatrix &sum_L_shared, const SizetArray &N_shared, const RealMatrix &sum_L_refined, const SizetArray &N_refined, size_t lev, const RealVector &beta, RealMatrix &H_raw_mom, int mom_index) |
apply matrix control variate parameter (beta) to approximate HF moment | |
void | apply_mlmf_control (Real sum_Hl, Real sum_Hlm1, Real sum_Ll, Real sum_Llm1, size_t N_shared, Real sum_Ll_refined, Real sum_Llm1_refined, size_t N_refined, Real beta_dot, Real gamma, Real &H_raw_mom) |
apply scalar control variate parameter (beta) to approximate HF moment | |
void | apply_mlmf_control (const RealMatrix &sum_Hl, const RealMatrix &sum_Hlm1, const RealMatrix &sum_Ll, const RealMatrix &sum_Llm1, const SizetArray &N_shared, const RealMatrix &sum_Ll_refined, const RealMatrix &sum_Llm1_refined, const SizetArray &N_refined, size_t lev, const RealVector &beta_dot, const RealVector &gamma, RealMatrix &H_raw_mom, int mom_index) |
apply matrix control variate parameter (beta) to approximate HF moment | |
void | update_projected_samples (const RealVector &hf_targets, const RealVectorArray &eval_ratios, const Sizet2DArray &N_actual_hf, SizetArray &N_alloc_hf, const RealVector &hf_cost, const Sizet2DArray &N_actual_lf, SizetArray &N_alloc_lf, const RealVector &lf_cost, SizetArray &delta_N_actual_hf, Real &delta_equiv_hf) |
for pilot projection mode, advance sample counts and accumulated cost | |
void | update_projected_lf_samples (const RealVector &hf_targets, const RealVectorArray &eval_ratios, const RealVector &hf_cost, const Sizet2DArray &N_actual_lf, SizetArray &N_alloc_lf, const RealVector &lf_cost, Real &delta_equiv_hf) |
for pilot projection mode, advance sample counts and accumulated cost | |
void | initialize_mlmf_sums (IntRealMatrixMap &sum_L_shared, IntRealMatrixMap &sum_L_refined, IntRealMatrixMap &sum_H, IntRealMatrixMap &sum_LL, IntRealMatrixMap &sum_LH, IntRealMatrixMap &sum_HH, size_t num_ml_lev, size_t num_cv_lev, size_t num_mom=4) |
initialize the MLMF accumulators for computing means, variances, and covariances across fidelity levels | |
void | initialize_mlmf_sums (IntRealMatrixMap &sum_Ll, IntRealMatrixMap &sum_Llm1, IntRealMatrixMap &sum_Ll_refined, IntRealMatrixMap &sum_Llm1_refined, IntRealMatrixMap &sum_Hl, IntRealMatrixMap &sum_Hlm1, IntRealMatrixMap &sum_Ll_Ll, IntRealMatrixMap &sum_Ll_Llm1, IntRealMatrixMap &sum_Llm1_Llm1, IntRealMatrixMap &sum_Hl_Ll, IntRealMatrixMap &sum_Hl_Llm1, IntRealMatrixMap &sum_Hlm1_Ll, IntRealMatrixMap &sum_Hlm1_Llm1, IntRealMatrixMap &sum_Hl_Hl, IntRealMatrixMap &sum_Hl_Hlm1, IntRealMatrixMap &sum_Hlm1_Hlm1, size_t num_ml_lev, size_t num_cv_lev, size_t num_mom=4) |
initialize the MLMF accumulators for computing means, variances, and covariances across fidelity levels | |
void | accumulate_mlmf_Qsums (IntRealMatrixMap &sum_Ql, IntRealMatrixMap &sum_Qlm1, size_t lev, SizetArray &num_Q) |
update running QoI sums for one model at two levels (sum_Ql, sum_Qlm1) using set of model evaluations within allResponses | |
void | accumulate_mlmf_Ysums (IntRealMatrixMap &sum_Y, size_t lev, SizetArray &num_Y) |
update running discrepancy sums for one model (sum_Y) using set of model evaluations within allResponses | |
void | accumulate_mlmf_Qsums (const IntResponseMap &lf_resp_map, const IntResponseMap &hf_resp_map, IntRealMatrixMap &sum_L_shared, IntRealMatrixMap &sum_L_refined, IntRealMatrixMap &sum_H, IntRealMatrixMap &sum_LL, IntRealMatrixMap &sum_LH, IntRealMatrixMap &sum_HH, size_t lev, SizetArray &num_L, SizetArray &num_H) |
update running QoI sums for two models (sum_L, sum_H, sum_LL, sum_LH, and sum_HH) from set of low/high fidelity model evaluations within {lf,hf}_resp_map; used for level 0 from other accumulators | |
void | accumulate_mlmf_Qsums (const IntResponseMap &lf_resp_map, const IntResponseMap &hf_resp_map, RealMatrix &sum_L_shared, RealMatrix &sum_L_refined, IntRealMatrixMap &sum_H, RealMatrix &sum_LL, RealMatrix &sum_LH, RealMatrix &sum_HH, size_t lev, SizetArray &N_shared) |
update running QoI sums for two models (sum_L, sum_H, sum_LL, sum_LH, and sum_HH) from set of low/high fidelity model evaluations within {lf,hf}_resp_map; used for level 0 from other accumulators | |
void | accumulate_mlmf_Ysums (const IntResponseMap &lf_resp_map, const IntResponseMap &hf_resp_map, IntRealMatrixMap &sum_L_shared, IntRealMatrixMap &sum_L_refined, IntRealMatrixMap &sum_H, IntRealMatrixMap &sum_LL, IntRealMatrixMap &sum_LH, IntRealMatrixMap &sum_HH, size_t lev, SizetArray &num_L, SizetArray &num_H) |
update running two-level discrepancy sums for two models (sum_L, sum_H, sum_LL, sum_LH, and sum_HH) from set of low/high fidelity model evaluations within {lf,hf}resp_map | |
void | accumulate_mlmf_Qsums (const IntResponseMap &lf_resp_map, const IntResponseMap &hf_resp_map, IntRealMatrixMap &sum_Ll, IntRealMatrixMap &sum_Llm1, IntRealMatrixMap &sum_Ll_refined, IntRealMatrixMap &sum_Llm1_refined, IntRealMatrixMap &sum_Hl, IntRealMatrixMap &sum_Hlm1, IntRealMatrixMap &sum_Ll_Ll, IntRealMatrixMap &sum_Ll_Llm1, IntRealMatrixMap &sum_Llm1_Llm1, IntRealMatrixMap &sum_Hl_Ll, IntRealMatrixMap &sum_Hl_Llm1, IntRealMatrixMap &sum_Hlm1_Ll, IntRealMatrixMap &sum_Hlm1_Llm1, IntRealMatrixMap &sum_Hl_Hl, IntRealMatrixMap &sum_Hl_Hlm1, IntRealMatrixMap &sum_Hlm1_Hlm1, size_t lev, SizetArray &num_L, SizetArray &num_H) |
update running QoI sums for two models and two levels from set of low/high fidelity model evaluations within {lf,hf}_resp_map | |
void | accumulate_mlmf_Qsums (const IntResponseMap &lf_resp_map, const IntResponseMap &hf_resp_map, RealMatrix &sum_Ll, RealMatrix &sum_Llm1, RealMatrix &sum_Ll_refined, RealMatrix &sum_Llm1_refined, IntRealMatrixMap &sum_Hl, IntRealMatrixMap &sum_Hlm1, RealMatrix &sum_Ll_Ll, RealMatrix &sum_Ll_Llm1, RealMatrix &sum_Llm1_Llm1, RealMatrix &sum_Hl_Ll, RealMatrix &sum_Hl_Llm1, RealMatrix &sum_Hlm1_Ll, RealMatrix &sum_Hlm1_Llm1, RealMatrix &sum_Hl_Hl, RealMatrix &sum_Hl_Hlm1, RealMatrix &sum_Hlm1_Hlm1, size_t lev, SizetArray &N_shared) |
update running QoI sums for two models and two levels from set of low/high fidelity model evaluations within {lf,hf}_resp_map | |
Private Attributes | |
short | delegateMethod |
core_run() can delegate execution to either ML or CV if hierarchy does not support MLCV; in this case output must also be delegated | |
Additional Inherited Members | |
Static Public Member Functions inherited from NonDSampling | |
static void | compute_moments (const RealVectorArray &fn_samples, SizetArray &sample_counts, RealMatrix &moment_stats, short moments_type, const StringArray &labels) |
core compute_moments() implementation with all data as inputs | |
static void | compute_moments (const RealVectorArray &fn_samples, RealMatrix &moment_stats, short moments_type) |
core compute_moments() implementation with all data as inputs | |
static void | compute_moments (const RealMatrix &fn_samples, RealMatrix &moment_stats, short moments_type) |
alternate RealMatrix samples API for use by external clients | |
static void | print_moments (std::ostream &s, const RealMatrix &moment_stats, const RealMatrix moment_cis, String qoi_type, short moments_type, const StringArray &moment_labels, bool print_cis) |
core print moments that can be called without object | |
static int | compute_wilks_sample_size (unsigned short order, Real alpha, Real beta, bool twosided=false) |
calculates the number of samples using the Wilks formula Static so I can test without instantiating a NonDSampling object - RWH | |
static Real | compute_wilks_residual (unsigned short order, int nsamples, Real alpha, Real beta, bool twosided) |
Helper function - calculates the Wilks residual. | |
static Real | compute_wilks_alpha (unsigned short order, int nsamples, Real beta, bool twosided=false) |
calculates the alpha paramter given number of samples using the Wilks formula Static so I can test without instantiating a NonDSampling object - RWH | |
static Real | compute_wilks_beta (unsigned short order, int nsamples, Real alpha, bool twosided=false) |
calculates the beta parameter given number of samples using the Wilks formula Static so I can test without instantiating a NonDSampling object - RWH | |
static Real | get_wilks_alpha_min () |
Get the lower and upper bounds supported by Wilks bisection solves. | |
static Real | get_wilks_alpha_max () |
static Real | get_wilks_beta_min () |
static Real | get_wilks_beta_max () |
Static Protected Member Functions inherited from NonDEnsembleSampling | |
static void | uncentered_to_centered (Real rm1, Real rm2, Real rm3, Real rm4, Real &cm1, Real &cm2, Real &cm3, Real &cm4) |
convert uncentered (raw) moments to centered moments; unbiased estimators More... | |
static void | uncentered_to_centered (Real rm1, Real rm2, Real rm3, Real rm4, Real &cm1, Real &cm2, Real &cm3, Real &cm4, size_t Nlq) |
convert uncentered (raw) moments to centered moments; biased estimators More... | |
static void | uncentered_to_centered (const Real *rm, Real *cm, size_t num_mom) |
convert uncentered (raw) moments to centered moments; biased estimators More... | |
static void | centered_to_standard (Real cm1, Real cm2, Real cm3, Real cm4, Real &sm1, Real &sm2, Real &sm3, Real &sm4) |
convert centered moments to standardized moments | |
static void | centered_to_standard (const Real *cm, Real *sm, size_t num_mom) |
convert centered moments to standardized moments | |
static void | check_negative (Real &cm) |
detect, warn, and repair a negative central moment (for even orders) | |
Static Protected Member Functions inherited from Iterator | |
static void | gnewton_set_recast (const Variables &recast_vars, const ActiveSet &recast_set, ActiveSet &sub_model_set) |
conversion of request vector values for the Gauss-Newton Hessian approximation More... | |
Protected Attributes inherited from NonDMultilevelSampling | |
RealVector | estVar |
final estimator variance for output in print_variance_reduction() | |
Protected Attributes inherited from NonDHierarchSampling | |
size_t | numSteps |
number of model forms/resolution in the (outer) sequence | |
size_t | secondaryIndex |
setting for inactive model dimension not traversed by loop over numSteps | |
Real | avgEstVar |
final estimator variance for targeted moment (usually mean), averaged across QoI | |
Protected Attributes inherited from NonDEnsembleSampling | |
short | sequenceType |
type of model sequence enumerated with primary MF/ACV loop over steps | |
RealVector | sequenceCost |
relative costs of model forms/resolution levels within a 1D sequence | |
Sizet3DArray | NLevActual |
total number of successful sample evaluations (excluding faults) for each model form, discretization level, and QoI | |
Sizet2DArray | NLevAlloc |
total number of allocated sample evaluations (prior to any faults) for each model form and discretization level (same for all QoI) | |
SizetArray | pilotSamples |
store the pilot_samples input specification, prior to run-time invocation of load_pilot_sample() | |
short | pilotMgmtMode |
enumeration for pilot management modes: ONLINE_PILOT (default), OFFLINE_PILOT, ONLINE_PILOT_PROJECTION, or OFFLINE_PILOT_PROJECTION | |
bool | onlineCost |
indicates use of online cost recovery rather than offline user-specified cost ratios | |
SizetSizetPairArray | costMetadataIndices |
indices of cost data within response metadata, one per model form | |
SizetArray | randomSeedSeqSpec |
user specification for seed_sequence | |
size_t | mlmfIter |
major iteration counter | |
bool | backfillFailures |
(inactive) option to backfill simulation failures by comparing targets against successful sample completions rather than sample allocations | |
Real | equivHFEvals |
equivalent number of high fidelity evaluations accumulated using samples across multiple model forms and/or discretization levels | |
Real | deltaEquivHF |
for sample projections, the calculated increment in equivHFEvals that would be incurred if full iteration/statistics were needed | |
RealVector | varH |
variances for HF truth (length numFunctions) | |
RealVector | estVarIter0 |
initial estimator variance from shared pilot (no CV reduction) | |
short | finalStatsType |
QOI_STATISTICS (moments, level mappings) or ESTIMATOR_PERFORMANCE (for model tuning of estVar,equivHFEvals by an outer loop) | |
bool | exportSampleSets |
if defined, export each of the sample increments in ML, CV, MLCV using tagged tabular files | |
unsigned short | exportSamplesFormat |
format for exporting sample increments using tagged tabular files | |
Real | relaxFactor |
the current relaxation factor applied to the predicted sample increment; in typical use, this is an under-relaxation factor to mitigate over-estimation of the sample allocation based on an initial approximation to response covariance data | |
size_t | relaxIndex |
index into relaxFactorSequence | |
RealVector | relaxFactorSequence |
a sequence of relaxation factors to use across ML/MF iterations (see DataMethod.hpp for usage notes) | |
Real | relaxRecursiveFactor |
a recursive relaxation factor (see DataMethod.hpp for usage notes) | |
Protected Attributes inherited from NonDSampling | |
int | seedSpec |
the user seed specification (default is 0) | |
int | randomSeed |
the current seed | |
const int | samplesSpec |
initial specification of number of samples | |
size_t | samplesRef |
reference number of samples updated for refinement | |
size_t | numSamples |
the current number of samples to evaluate | |
String | rngName |
name of the random number generator | |
unsigned short | sampleType |
the sample type: default, random, lhs, incremental random, or incremental lhs | |
bool | wilksFlag |
flags use of Wilks formula to calculate num samples | |
unsigned short | wilksOrder |
Real | wilksAlpha |
Real | wilksBeta |
short | wilksSidedness |
RealMatrix | momentGrads |
gradients of standardized or central moments of response functions, as determined by finalMomentsType. Calculated in compute_moments() and indexed as (var,moment) when moment id runs from 1:2*numFunctions. | |
RealSymMatrix | finalStatErrors |
standard errors (estimator std deviation) for each of the finalStatistics | |
int | samplesIncrement |
current increment in a sequence of samples | |
Pecos::LHSDriver | lhsDriver |
the C++ wrapper for the F90 LHS library | |
size_t | numLHSRuns |
counter for number of sample set generations | |
bool | stdRegressionCoeffs |
flags computation/output of standardized regression coefficients | |
bool | toleranceIntervalsFlag |
flags of double sided tolerance interval equivalent normal | |
Real | tiCoverage |
coverage to be used in the calculation of the double sided tolerance interval equivaluent normal | |
Real | tiConfidenceLevel |
confidence interval to be used in the calculation of the double sided tolerance interval equivalent normal | |
size_t | tiNumValidSamples |
RealVector | tiDstienMus |
Real | tiDeltaMultiplicativeFactor |
RealVector | tiSampleSigmas |
RealVector | tiDstienSigmas |
bool | statsFlag |
flags computation/output of statistics | |
bool | allDataFlag |
flags update of allResponses (allVariables or allSamples already defined) | |
short | samplingVarsMode |
the sampling mode: ALEATORY_UNCERTAIN{,_UNIFORM}, EPISTEMIC_UNCERTAIN{,_UNIFORM}, UNCERTAIN{,_UNIFORM}, ACTIVE{,_UNIFORM}, or ALL{,_UNIFORM}. This is a secondary control on top of the variables view that allows sampling over subsets of variables that may differ from the view. | |
short | sampleRanksMode |
mode for input/output of LHS sample ranks: IGNORE_RANKS, GET_RANKS, SET_RANKS, or SET_GET_RANKS | |
bool | varyPattern |
flag for generating a sequence of seed values within multiple get_parameter_sets() calls so that these executions (e.g., for SBO/SBNLS) are not repeated, but are still repeatable | |
RealMatrix | sampleRanks |
data structure to hold the sample ranks | |
SensAnalysisGlobal | nonDSampCorr |
initialize statistical post processing | |
bool | backfillDuplicates |
flags whether to use backfill to enforce uniqueness of discrete LHS samples | |
RealRealPairArray | extremeValues |
Minimum and maximum values of response functions for epistemic calculations (calculated in compute_intervals()),. | |
bool | functionMomentsComputed |
Function moments have been computed; used to determine whether to archive the moments. | |
Protected Attributes inherited from NonD | |
NonD * | prevNondInstance |
pointer containing previous value of nondInstance | |
size_t | startCAUV |
starting index of continuous aleatory uncertain variables within active continuous variables (convenience for managing offsets) | |
size_t | numCAUV |
number of active continuous aleatory uncertain variables | |
bool | epistemicStats |
flag for computing interval-type metrics instead of integrated metrics If any epistemic vars are active in a metric evaluation, then flag is set. | |
RealMatrix | momentStats |
standardized or central resp moments, as determined by finalMomentsType. Calculated in compute_moments()) and indexed as (moment,fn). | |
RealVectorArray | requestedRespLevels |
requested response levels for all response functions | |
RealVectorArray | computedProbLevels |
output probability levels for all response functions resulting from requestedRespLevels | |
RealVectorArray | computedRelLevels |
output reliability levels for all response functions resulting from requestedRespLevels | |
RealVectorArray | computedGenRelLevels |
output generalized reliability levels for all response functions resulting from requestedRespLevels | |
short | respLevelTarget |
indicates mapping of z->p (PROBABILITIES), z->beta (RELIABILITIES), or z->beta* (GEN_RELIABILITIES) | |
short | respLevelTargetReduce |
indicates component or system series/parallel failure metrics | |
RealVectorArray | requestedProbLevels |
requested probability levels for all response functions | |
RealVectorArray | requestedRelLevels |
requested reliability levels for all response functions | |
RealVectorArray | requestedGenRelLevels |
requested generalized reliability levels for all response functions | |
RealVectorArray | computedRespLevels |
output response levels for all response functions resulting from requestedProbLevels, requestedRelLevels, or requestedGenRelLevels | |
size_t | totalLevelRequests |
total number of levels specified within requestedRespLevels, requestedProbLevels, and requestedRelLevels | |
bool | cdfFlag |
flag for type of probabilities/reliabilities used in mappings: cumulative/CDF (true) or complementary/CCDF (false) | |
bool | pdfOutput |
flag for managing output of response probability density functions (PDFs) | |
RealVectorArray | computedPDFAbscissas |
sorted response PDF intervals bounds extracted from min/max sample and requested/computedRespLevels (vector lengths = num bins + 1) | |
RealVectorArray | computedPDFOrdinates |
response PDF densities computed from bin counts divided by (unequal) bin widths (vector lengths = num bins) | |
Response | finalStatistics |
final statistics from the uncertainty propagation used in strategies: response means, standard deviations, and probabilities of failure | |
short | finalMomentsType |
type of moments logged within finalStatistics: none, central, standard | |
size_t | miPLIndex |
index for the active ParallelLevel within ParallelConfiguration::miPLIters | |
BitArray | pdfComputed |
Whether PDF was computed for function i; used to determine whether a pdf should be archived. | |
Protected Attributes inherited from Analyzer | |
size_t | numFunctions |
number of response functions | |
size_t | numContinuousVars |
number of active continuous vars | |
size_t | numDiscreteIntVars |
number of active discrete integer vars | |
size_t | numDiscreteStringVars |
number of active discrete string vars | |
size_t | numDiscreteRealVars |
number of active discrete real vars | |
bool | compactMode |
switch for allSamples (compact mode) instead of allVariables (normal mode) | |
VariablesArray | allVariables |
array of all variables to be evaluated in evaluate_parameter_sets() | |
RealMatrix | allSamples |
compact alternative to allVariables | |
IntResponseMap | allResponses |
array of all responses to be computed in evaluate_parameter_sets() | |
IntVariables2DMap | batchVariablesMap |
alternate container for Variables samples supporting evaluate_batch() and synchronize_batches(), a 2D map with outer batch_id and inner eval_id | |
IntRealVector2DMap | batchSamplesMap |
alternate container for RealVector samples supporting evaluate_batch() and synchronize_batches(), a 2D map with outer batch_id and inner eval_id | |
IntResponse2DMap | batchResponsesMap |
alternate container for Response samples supporting evaluate_batch() and synchronize_batches(), a 2D map with outer batch_id and inner eval_id | |
StringArray | allHeaders |
array of headers to insert into output while evaluating allVariables | |
size_t | numObjFns |
number of objective functions | |
size_t | numLSqTerms |
number of least squares terms | |
RealPairPRPMultiMap | bestVarsRespMap |
map which stores best set of solutions | |
bool | vbdFlag |
flag indicating the activation of variance-bsaed decomposition for computing Sobol' indices, via either PCE or sampling | |
Real | vbdDropTol |
tolerance for omitting output of small VBD indices computed via either PCE or sampling | |
Protected Attributes inherited from Iterator | |
ProblemDescDB & | probDescDB |
class member reference to the problem description database More... | |
ParallelLibrary & | parallelLib |
class member reference to the parallel library | |
ParConfigLIter | methodPCIter |
the active ParallelConfiguration used by this Iterator instance | |
Model | iteratedModel |
the model to be iterated (for iterators and meta-iterators employing a single model instance) | |
size_t | myModelLayers |
number of Models locally (in Iterator or derived classes) wrapped around the initially passed in Model | |
unsigned short | methodName |
name of the iterator (the user's method spec) | |
Real | convergenceTol |
iteration convergence tolerance | |
size_t | maxIterations |
maximum number of iterations for the method | |
size_t | maxFunctionEvals |
maximum number of fn evaluations for the method | |
int | maxEvalConcurrency |
maximum number of concurrent model evaluations More... | |
ActiveSet | activeSet |
the response data requirements on each function evaluation | |
size_t | numFinalSolutions |
number of solutions to retain in best variables/response arrays | |
VariablesArray | bestVariablesArray |
collection of N best solution variables found during the study; always in context of Model originally passed to the Iterator (any in-flight Recasts must be undone) | |
ResponseArray | bestResponseArray |
collection of N best solution responses found during the study; always in context of Model originally passed to the Iterator (any in-flight Recasts must be undone) | |
bool | subIteratorFlag |
flag indicating if this Iterator is a sub-iterator (NestedModel::subIterator or DataFitSurrModel::daceIterator) | |
short | outputLevel |
output verbosity level: {SILENT,QUIET,NORMAL,VERBOSE,DEBUG}_OUTPUT | |
bool | summaryOutputFlag |
flag for summary output (evaluation stats, final results); default true, but false for on-the-fly (helper) iterators and sub-iterator use cases | |
ResultsManager & | resultsDB |
reference to the global iterator results database | |
EvaluationStore & | evaluationsDB |
reference to the global evaluation database | |
EvaluationsDBState | evaluationsDBState |
State of evaluations DB for this iterator. | |
ResultsNames | resultsNames |
valid names for iterator results | |
std::shared_ptr< TraitsBase > | methodTraits |
pointer that retains shared ownership of a TraitsBase object, or child thereof | |
bool | topLevel |
Whether this is the top level iterator. | |
bool | exportSurrogate = false |
whether to export final surrogates | |
String | surrExportPrefix |
base filename for exported surrogates | |
unsigned short | surrExportFormat = NO_MODEL_FORMAT |
(bitwise) format(s) to export | |
Static Protected Attributes inherited from NonD | |
static NonD * | nondInstance |
pointer to the active object instance used within static evaluator functions in order to avoid the need for static data | |
Performs multilevel-multifidelity Monte Carlo sampling for uncertainty quantification.
Multilevel-multifidelity Monte Carlo (MLMFMC) combines variance decay across model resolutions with variance reduction from a control variate across model fidelities.
NonDMultilevControlVarSampling | ( | ProblemDescDB & | problem_db, |
Model & | model | ||
) |
standard constructor
This constructor is called for a standard letter-envelope iterator instantiation. In this case, set_db_list_nodes has been called and probDescDB can be queried for settings from the method specification.
References Model::ensemble_precedence(), Iterator::iteratedModel, and Model::multilevel_multifidelity().
|
protectedvirtual |
pre-run portion of run (optional); re-implemented by Iterators which can generate all Variables (parameter sets) a priori
pre-run phase, which a derived iterator may optionally reimplement; when not present, pre-run is likely integrated into the derived run function. This is a virtual function; when re-implementing, a derived class must call its nearest parent's pre_run(), if implemented, typically before performing its own implementation steps.
Reimplemented from NonDEnsembleSampling.
References Iterator::iteratedModel, NonDEnsembleSampling::NLevActual, NonDEnsembleSampling::NLevAlloc, Analyzer::numFunctions, NonDEnsembleSampling::pre_run(), Model::solution_levels(), and Model::truth_model().
|
protectedvirtual |
The primary run function manages the general case: a hierarchy of model forms (from the ordered model fidelities within an EnsembleSurrModel), each of which may contain multiple discretization levels.
Reimplemented from Iterator.
References Model::active_model_key(), NonDMultilevelSampling::core_run(), NonDMultilevControlVarSampling::delegateMethod, NonDEnsembleSampling::finalStatsType, Iterator::iteratedModel, NonDMultilevControlVarSampling::multilevel_control_variate_mc_offline_pilot(), NonDMultilevControlVarSampling::multilevel_control_variate_mc_online_pilot(), NonDMultilevControlVarSampling::multilevel_control_variate_mc_pilot_projection(), NonDEnsembleSampling::NLevActual, NonDEnsembleSampling::pilotMgmtMode, NonDEnsembleSampling::sequenceType, Model::subordinate_models(), and Dakota::SZ_MAX.
|
private |
Perform multilevel Monte Carlo across levels in combination with control variate Monte Carlo across model forms at each level; CV computes correlations for Q (LH correlations for QoI)
This function performs "geometrical" MLMC across discretization levels for the high fidelity model form where CVMC si employed across two model forms to exploit correlation in the discrepancies at each level (Y_l). void NonDMultilevControlVarSampling::multilevel_control_variate_mc_Ycorr() { Model& truth_model = iteratedModel.truth_model(); Model& surr_model = iteratedModel.surrogate_model(); size_t qoi, lev, num_mf = NLevActual.size(), num_hf_lev = truth_model.solution_levels(), num_cv_lev = std::min(num_hf_lev, surr_model.solution_levels()); bool budget_constrained = (maxFunctionEvals != SZ_MAX);
retrieve cost estimates across solution levels for HF model RealVector hf_targets(num_hf_lev), agg_var_hf(num_hf_lev), hf_cost = truth_model.solution_level_costs(), lf_cost = surr_model.solution_level_costs(); Real eps_sq_div_2, sum_sqrt_var_cost, agg_estvar_iter0 = 0., budget, r_lq, lf_lev_cost, hf_lev_cost, hf_ref_cost = hf_cost[num_hf_lev-1]; if (budget_constrained) budget = (Real)maxFunctionEvals * hf_ref_cost; RealVectorArray eval_ratios(num_cv_lev); For moment estimation, we accumulate telescoping sums for Q^i using discrepancies Yi = Q^i_{lev} - Q^i_{lev-1} (Y_diff_Qpow[i] for i=1:4). For computing N_l from estimator variance, we accumulate square of Y1 estimator (YY[1] = (Y^i)^2 for i=1). IntRealMatrixMap sum_L_refined, sum_L_shared, sum_H, sum_LL, sum_LH, sum_HH; initialize_mlmf_sums(sum_L_shared, sum_L_refined, sum_H, sum_LL, sum_LH, sum_HH, num_hf_lev, num_cv_lev); RealMatrix var_YH(numFunctions, num_hf_lev, false), rho2_LH(numFunctions, num_cv_lev, false), Lambda(numFunctions, num_cv_lev, false); RealVector avg_rho2_LH(num_cv_lev, false), avg_lambda(num_cv_lev, false);
Initialize for pilot sample unsigned short group, lf_form = 0, hf_form = num_mf - 1;// 2 models @ extremes Sizet2DArray& N_lf = NLevActual[lf_form]; Sizet2DArray& N_hf = NLevActual[hf_form]; Sizet2DArray delta_N_l; load_pilot_sample(pilotSamples, sequenceType, NLevActual, delta_N_l); SizetArray& delta_N_lf = delta_N_l[lf_form]; SizetArray& delta_N_hf = delta_N_l[hf_form];
now converge on sample counts per level (N_hf) while (!zeros(delta_N_hf) && mlmfIter <= maxIterations) {
sum_sqrt_var_cost = 0.; for (lev=0, group=0; lev<num_hf_lev; ++lev, ++group) {
configure_indices(group, hf_form, lev, sequenceType); hf_lev_cost = level_cost(hf_cost, lev);
set the number of current samples from the defined increment numSamples = delta_N_hf[lev];
aggregate variances across QoI for estimating N_hf (justification: for independent QoI, sum of QoI variances = variance of QoI sum) Real& agg_var_hf_l = agg_var_hf[lev];//carried over from prev iter if!samp if (numSamples) {
assign sequence, get samples, export, evaluate evaluate_ml_sample_increment("mlcv_", lev);
control variate betwen LF and HF for this discretization level: if unequal number of levels, loop over all HF levels for MLMC and apply CVMC when LF levels are available. LF levels are assigned as control variates to the leading set of HF levels, since these will tend to have larger variance.
if (lev < num_cv_lev) {
store allResponses used for sum_H (and sum_HH) IntResponseMap hf_resp = allResponses; // shallow copy activate LF response (lev 0) or LF response discrepancy (lev > 0) within the hierarchical surrogate model. Level indices & surrogate response mode are same as HF above, only the model form changes. However, we must pass the unchanged level index to update the corresponding variable values for the new model form. configure_indices(group, lf_form, lev, sequenceType); lf_lev_cost = level_cost(lf_cost, lev); compute allResp w/ LF model form reusing allVars from MLMC step
Note: this CV sample set is not separately exported (see below).
evaluate_parameter_sets(iteratedModel); process previous and new set of allResponses for CV sums accumulate_mlmf_Ysums(allResponses, hf_resp, sum_L_shared, sum_L_refined, sum_H, sum_LL, sum_LH, sum_HH, lev, N_lf[lev], N_hf[lev]); if (outputLevel == DEBUG_OUTPUT) Cout << "Accumulated sums (L_shared[1,2], L_refined[1,2], LH[1,2])" << ":\n" << sum_L_shared[1] << sum_L_shared[2] << sum_L_refined[1]<< sum_L_refined[2]<< sum_LH[1]<< sum_LH[2]; increment_mlmf_equivalent_cost(numSamples, hf_lev_cost, numSamples, lf_lev_cost, hf_ref_cost, equivHFEvals);
compute the average evaluation ratio and Lambda factor RealVector& eval_ratios_l = eval_ratios[lev]; compute_eval_ratios(sum_L_shared[1], sum_H[1], sum_LL[1], sum_LH[1], sum_HH[1], hf_lev_cost/lf_lev_cost, lev, N_hf[lev], var_YH, rho2_LH, eval_ratios_l);
retain Lambda per QoI and level, but apply QoI-average where needed for (qoi=0; qoi<numFunctions; ++qoi) { r_lq = eval_ratios_l[qoi]; Lambda(qoi,lev) = 1. - rho2_LH(qoi,lev) * (r_lq - 1.) / r_lq; } avg_lambda[lev] = average(Lambda[lev], numFunctions); avg_rho2_LH[lev] = average(rho2_LH[lev], numFunctions); } else { // no LF model for this level; accumulate only multilevel sums RealMatrix& sum_HH1 = sum_HH[1]; accumulate H sums for lev = 0, Y sums for lev > 0 accumulate_ml_Ysums(sum_H, sum_HH1, lev, N_hf[lev]); if (outputLevel == DEBUG_OUTPUT) Cout << "Accumulated sums (H[1], H[2], HH):\n" << sum_H[1] << sum_H[2] << sum_HH1; increment_ml_equivalent_cost(numSamples, hf_lev_cost, hf_ref_cost, equivHFEvals); compute Y variances for this level and aggregate across QoI: variance_Ysum(sum_H[1][lev], sum_HH1[lev], N_hf[lev], var_YH[lev]); } agg_var_hf_l = sum(var_YH[lev], numFunctions); }
accumulate sum of sqrt's of estimator var * cost used in N_target if (lev < num_cv_lev) { Real om_rho2 = 1. - avg_rho2_LH[lev]; sum_sqrt_var_cost += (budget_constrained) ? std::sqrt(agg_var_hf_l / hf_lev_cost * om_rho2) * (hf_lev_cost + (1. + average(eval_ratios[lev])) * lf_lev_cost) : std::sqrt(agg_var_hf_l * hf_lev_cost / om_rho2) * avg_lambda[lev]; } else sum_sqrt_var_cost += std::sqrt(agg_var_hf_l * hf_lev_cost);
MSE reference is MLMF MC applied to {HF,LF} pilot sample aggregated across qoi. Note: if the pilot sample for LF is not shaped, then r=1 will result in no additional variance reduction beyond MLMC. if (mlmfIter == 0 && !budget_constrained) agg_estvar_iter0 += aggregate_mse_Yvar(var_YH[lev], N_hf[lev]); } compute epsilon target based on relative tolerance: total MSE = eps^2 which is equally apportioned (eps^2 / 2) among discretization MSE and estimator variance (\Sum var_Y_l / N_l). Since we do not know the discretization error, we compute an initial estimator variance and then seek to reduce it by a relative_factor <= 1. if (mlmfIter == 0) { MLMC estimator variance for final estvar reporting is not aggregated (reduction from control variate is applied subsequently) compute_ml_estimator_variance(var_YH, N_hf, estVarIter0); compute eps^2 / 2 = aggregated estvar0 * rel tol if (!budget_constrained) { eps_sq_div_2 = agg_estvar_iter0 * convergenceTol; if (outputLevel == DEBUG_OUTPUT) Cout << "Epsilon squared target = " << eps_sq_div_2 << std::endl; } }
update sample targets based on variance estimates Note: sum_sqrt_var_cost is defined differently for the two cases Real fact = (budget_constrained) ? budget / sum_sqrt_var_cost : // budget constraint sum_sqrt_var_cost / eps_sq_div_2; // error balance constraint for (lev=0; lev<num_hf_lev; ++lev) { hf_lev_cost = level_cost(hf_cost, lev); hf_targets[lev] = (lev < num_cv_lev) ? fact * std::sqrt(agg_var_hf[lev] / hf_lev_cost * (1. - avg_rho2_LH[lev])) : fact * std::sqrt(agg_var_hf[lev] / hf_lev_cost); delta_N_hf[lev] = one_sided_delta(N_alloc_hf[lev], hf_targets[lev], relaxFactor); }
++mlmfIter; Cout << "\nMLMF MC iteration " << mlmfIter << " sample increments:\n" << delta_N_hf << std::endl; }
All CV lf_increment() calls now follow convergence of ML iteration: for (lev=0, group=0; lev<num_cv_lev; ++lev, ++group) { configure_indices(group, lf_form, lev, sequenceType); execute additional LF sample increment if (lf_increment(eval_ratios[lev], N_lf[lev], hf_targets[lev], mlmfIter, lev)) { accumulate_mlmf_Ysums(sum_L_refined, lev, N_lf[lev]); increment_ml_equivalent_cost(numSamples, level_cost(lf_cost, lev), hf_ref_cost, equivHFEvals); if (outputLevel == DEBUG_OUTPUT) Cout << "Accumulated sums (L_refined[1,2]):\n" << sum_L_refined[1] << sum_L_refined[2]; } }
Roll up raw moments from MLCVMC and MLMC levels RealMatrix Y_mom(4, numFunctions), Y_cvmc_mom(4, numFunctions, false); for (lev=0; lev<num_cv_lev; ++lev) { cv_raw_moments(sum_L_shared, sum_H, sum_LL, sum_LH, N_hf[lev], sum_L_refined, N_lf[lev], //rho2_LH, lev, Y_cvmc_mom); Y_mom += Y_cvmc_mom; } if (num_hf_lev > num_cv_lev) ml_Y_raw_moments(sum_H[1], sum_H[2], sum_H[3], sum_H[4], N_hf, num_cv_lev, num_hf_lev, Y_mom); convert_moments(Y_mom, momentStats); // raw to final (central or std)
RealMatrix Y_mlmc_mom(4, numFunctions), mlmc_stats; ml_Q_raw_moments(..., N_actual_hf, 0, num_cv_lev, Y_mlmc_mom); ml_Y_raw_moments(sum_Hl[1], sum_Hl[2], sum_Hl[3], sum_Hl[4], N_actual_hf, num_cv_lev, num_hf_lev, Y_mlmc_mom); convert_moments(Y_mlmc_mom, mlmc_stats); // raw to final (central or std) recover_variance(mlmc_stats, varH);
compute_mlmf_estimator_variance(var_YH, N_hf, Lambda, estVar); avgEstVar = average(estVar); } This function performs "geometrical" MLMC across discretization levels for the high fidelity model form where CVMC is employed across two model forms. It generalizes the Y_l correlation case to separately target correlations for each QoI level embedded within the level discrepancies.
References NonDMultilevelSampling::accumulate_ml_Ysums(), NonDMultilevControlVarSampling::accumulate_mlmf_Qsums(), NonDHierarchSampling::accumulate_paired_online_cost(), NonDEnsembleSampling::advance_relaxation(), NonDMultilevelSampling::aggregate_mse_Yvar(), Analyzer::allResponses, Dakota::average(), NonDHierarchSampling::average_online_cost(), NonDHierarchSampling::avgEstVar, NonDEnsembleSampling::backfillFailures, NonDMultilevControlVarSampling::compute_eval_ratios(), NonDMultilevelSampling::compute_ml_estimator_variance(), NonDMultilevControlVarSampling::compute_mlmf_estimator_variance(), NonDMultilevelSampling::configure_indices(), Iterator::convergenceTol, NonDEnsembleSampling::convert_moments(), NonDMultilevControlVarSampling::cv_raw_moments(), NonDEnsembleSampling::deltaEquivHF, NonDEnsembleSampling::equivHFEvals, NonDMultilevelSampling::estVar, NonDEnsembleSampling::estVarIter0, NonDMultilevelSampling::evaluate_ml_sample_increment(), Analyzer::evaluate_parameter_sets(), NonDEnsembleSampling::finalStatsType, NonDMultilevelSampling::increment_ml_equivalent_cost(), NonDMultilevControlVarSampling::increment_mlmf_equivalent_cost(), NonDMultilevControlVarSampling::initialize_mlmf_sums(), Iterator::iteratedModel, NonDMultilevelSampling::level_cost(), NonDMultilevControlVarSampling::lf_increment(), NonD::load_pilot_sample(), Iterator::maxFunctionEvals, Iterator::maxIterations, NonDMultilevelSampling::ml_Q_raw_moments(), NonDMultilevelSampling::ml_Y_raw_moments(), NonDEnsembleSampling::mlmfIter, NonD::momentStats, NonDEnsembleSampling::NLevActual, NonDEnsembleSampling::NLevAlloc, Analyzer::numFunctions, NonDSampling::numSamples, NonD::one_sided_delta(), Iterator::outputLevel, NonDEnsembleSampling::pilotSamples, NonD::query_cost(), NonDMultilevelSampling::recover_variance(), NonDEnsembleSampling::relaxFactor, NonDEnsembleSampling::sequenceType, Model::solution_levels(), Dakota::sum(), Model::surrogate_model(), Dakota::SZ_MAX, Model::truth_model(), NonDMultilevControlVarSampling::update_projected_lf_samples(), NonDEnsembleSampling::varH, NonDMultilevelSampling::variance_Ysum(), and NonD::zeros().
Referenced by NonDMultilevControlVarSampling::core_run().