Dakota
Version 6.19
Explore and Predict with Confidence
|
Container class for response functions and their derivatives.
ExperimentResponse provides the body class.
More...
Public Member Functions | |
ExperimentResponse () | |
default constructor | |
ExperimentResponse (const Variables &vars, const ProblemDescDB &problem_db) | |
standard constructor built from problem description database | |
ExperimentResponse (const SharedResponseData &srd, const ActiveSet &set) | |
alternate constructor that shares a SharedResponseData instance | |
ExperimentResponse (const SharedResponseData &srd) | |
alternate constructor that shares a SharedResponseData instance | |
ExperimentResponse (const ActiveSet &set) | |
alternate constructor using limited data | |
~ExperimentResponse () | |
destructor | |
void | set_scalar_covariance (RealVector &scalars) override |
method to set the covariance matrix defined for ExperimentResponse | |
const ExperimentCovariance & | experiment_covariance () const override |
retrieve the ExperimentCovariance structure | |
void | set_full_covariance (std::vector< RealMatrix > &matrices, std::vector< RealVector > &diagonals, RealVector &scalars, IntVector matrix_map_indices, IntVector diagonal_map_indices, IntVector scalar_map_indices) override |
method to set the full covariance matrices for ExperimentResponse | |
Real | apply_covariance (const RealVector &residual) const override |
method to compute the triple product v'*inv(C)*v. | |
void | apply_covariance_inv_sqrt (const RealVector &residuals, RealVector &weighted_residuals) const override |
method to compute (v'*inv(C)^1/2), to compute weighted residual | |
void | apply_covariance_inv_sqrt (const RealMatrix &gradients, RealMatrix &weighted_gradients) const override |
void | apply_covariance_inv_sqrt (const RealSymMatrixArray &hessians, RealSymMatrixArray &weighted_hessians) const override |
void | get_covariance_diagonal (RealVector &diagonal) const override |
Real | covariance_determinant () const override |
covariance determinant for this experiment (default 1.0) | |
Real | log_covariance_determinant () const override |
log covariance determinant for this experiment (default 0.0) | |
Public Member Functions inherited from Response | |
Response () | |
default constructor | |
Response (short type, const Variables &vars, const ProblemDescDB &problem_db) | |
standard constructor built from problem description database | |
Response (const SharedResponseData &srd, const ActiveSet &set) | |
alternate constructor that shares response data | |
Response (short type, const ActiveSet &set) | |
alternate constructor using limited data without sharing | |
Response (const SharedResponseData &srd) | |
alternate constructor using limited data (explicit disallows implicit type conversion) | |
Response (const Response &response) | |
copy constructor | |
virtual | ~Response () |
destructor | |
Response | operator= (const Response &response) |
assignment operator | |
const SharedResponseData & | shared_data () const |
return sharedRespData | |
SharedResponseData & | shared_data () |
return sharedRespData | |
size_t | num_functions () const |
return the number of response functions | |
const ActiveSet & | active_set () const |
return the active set | |
void | active_set (const ActiveSet &set) |
set the active set | |
const ShortArray & | active_set_request_vector () const |
return the active set request vector | |
ShortArray & | active_set_request_vector () |
return the active set request vector | |
void | active_set_request_vector (const ShortArray &asrv) |
set the active set request vector and verify consistent number of response functions | |
const SizetArray & | active_set_derivative_vector () const |
return the active set derivative vector | |
SizetArray & | active_set_derivative_vector () |
return the active set derivative vector | |
void | active_set_derivative_vector (const SizetArray &asdv) |
set the active set derivative vector and reshape functionGradients/functionHessians if needed | |
void | active_set_derivative_vector (SizetMultiArrayConstView asdv) |
set the active set derivative vector and reshape functionGradients/functionHessians if needed | |
const Real & | function_value (size_t i) const |
return a function value | |
Real & | function_value_view (size_t i) |
return a "view" of a function value for updating in place | |
const RealVector & | function_values () const |
return all function values | |
RealVector | function_values_view () |
return all function values as a view for updating in place | |
RealVector | function_values_view () const |
return all function values as a view for accessing the function values vector from a const response | |
void | function_value (const Real &fn_val, size_t i) |
set a function value | |
void | function_values (const RealVector &fn_vals) |
set all function values | |
const Real * | function_gradient (int i) const |
return the i-th function gradient as a const Real* | |
RealVector | function_gradient_view (int i) |
return the i-th function gradient as a SerialDenseVector view (shallow copy) for updating in place | |
RealVector | function_gradient_view (int i) const |
return the i-th function gradient as a SerialDenseVector view (shallow copy) for accessing a column vector from a const matrix | |
RealVector | function_gradient_copy (int i) const |
return the i-th function gradient as a SerialDenseVector Teuchos::Copy (deep copy) | |
const RealMatrix & | function_gradients () const |
return all function gradients | |
RealMatrix | function_gradients_view () |
return all function gradients as a view for updating in place | |
RealMatrix | function_gradients_view () const |
return all function gradients as a view for updating in place | |
void | function_gradient (const RealVector &assign_grad, int fn_index) |
set a function gradient | |
void | function_gradient (const RealVector &assign_grad, int fn_index, const SizetArray &assign_indices, const SizetArray &curr_indices) |
set a function gradient, managing dissimilar DVV | |
void | function_gradient (const RealVector &assign_grad, int fn_index, const SizetArray &assign_dvv) |
set a function gradient, managing dissimilar DVV | |
void | function_gradients (const RealMatrix &fn_grads) |
set all function gradients | |
const RealSymMatrix & | function_hessian (size_t i) const |
return the i-th function Hessian | |
RealSymMatrix | function_hessian_view (size_t i) |
return the i-th function Hessian as a Teuchos::View (shallow copy) for updating in place | |
RealSymMatrix | function_hessian_view (size_t i) const |
return the i-th function Hessian as a Teuchos::View (shallow copy) for accessing the i-th matrix within a const matrix array | |
const RealSymMatrixArray & | function_hessians () const |
return all function Hessians | |
RealSymMatrixArray | function_hessians_view () |
return all function Hessians as Teuchos::Views (shallow copies) for updating in place | |
RealSymMatrixArray | function_hessians_view () const |
return all function Hessians as Teuchos::Views (shallow copies) for updating in place | |
void | function_hessian (const RealSymMatrix &assign_hessian, size_t fn_index) |
set a function Hessian | |
void | function_hessian (const RealSymMatrix &assign_hessian, size_t fn_index, const SizetArray &assign_indices, const SizetArray &curr_indices) |
set a function Hessian, using DVV index mappings | |
void | function_hessian (const RealSymMatrix &assign_hessian, size_t fn_index, const SizetArray &assign_dvv) |
set a function Hessian, managing dissimilar DVV | |
void | function_hessians (const RealSymMatrixArray &fn_hessians) |
set all function Hessians | |
void | map_dvv_indices (const SizetArray &assign_dvv, SizetArray &assign_indices, SizetArray &curr_indices) |
define source and target indices for updating derivatives by matching assign_dvv against the current DVV | |
const IntVector & | field_lengths () const |
return the field lengths (from SharedResponseData) | |
void | field_lengths (const IntVector &field_lens) |
set the field lengths (within SharedResponseData) | |
RealVector | field_values_view (size_t i) const |
return const field values for the i-th field | |
RealVector | field_values_view (size_t i) |
return a "view" of the i-th field values for updating in place | |
void | field_values (const RealVector &field_val, size_t i) |
set the values for the i-th field | |
RealMatrix | field_gradients_view (size_t i) const |
return a view of the gradients of each element of the i-th field | |
RealSymMatrixArray | field_hessians_view (size_t i) const |
return a view of the hessians of each element of the i-th field | |
RealMatrix | field_coords_view (size_t i) |
return a "view" of the i-th field's coordinates | |
const RealMatrix | field_coords_view (size_t i) const |
return a const "view" of the i-th field's coordinates | |
void | field_coords (const RealMatrix &field_coords, size_t i) |
set the i-th field's coordinates | |
const IntVector & | num_coords_per_field () const |
return the number of coordinates each field has (from SharedResponseData) | |
const StringArray & | function_labels () const |
return the fine-grained (unrolled) response function identifier strings from sharedRespData | |
void | function_labels (const StringArray &labels) |
set the fine-grained (unrolled) response function identifier strings within sharedRespData | |
const StringArray & | field_group_labels () |
return the user-provided field group labels instead of the unrolled labels available through function_labels() | |
const std::vector< RespMetadataT > & | metadata () const |
get the (possibly empty) response metadata; (get labels through shared_data()) | |
void | metadata (const std::vector< RespMetadataT > &md) |
set the response metadata (set labels through shared_data()) | |
void | metadata (const std::vector< RespMetadataT > &md, size_t start) |
set a portion of the response metadata starting from given position | |
void | read (std::istream &s, const unsigned short format=FLEXIBLE_RESULTS) |
read a response object of specified format from a std::istream | |
void | write (std::ostream &s) const |
write a response object to a std::ostream | |
void | read_annotated (std::istream &s) |
read a response object in annotated format from a std::istream | |
void | write_annotated (std::ostream &s) const |
write a response object in annotated format to a std::ostream | |
void | read_tabular (std::istream &s) |
read responseRep::functionValues in tabular format from a std::istream | |
void | write_tabular (std::ostream &s, bool eol=true) const |
write responseRep::functionValues in tabular format to a std::ostream | |
void | write_tabular_partial (std::ostream &s, size_t start_index, size_t num_items) const |
write portion of responseRep::functionValues in tabular format to a std::ostream | |
void | write_tabular_labels (std::ostream &s, bool eol=true) const |
write the response labels in tabular format to a std::ostream | |
void | read (MPIUnpackBuffer &s) |
read a response object from a packed MPI buffer | |
void | write (MPIPackBuffer &s) const |
write a response object to a packed MPI buffer | |
Response | copy (bool deep_srd=false) const |
return a deep response copy of the contained responseRep for use in history mechanisms (SharedResponseData uses a shallow copy by default) | |
int | data_size () |
return the number of doubles active in response. Used for sizing double* response_data arrays passed into read_data and write_data. | |
void | read_data (double *response_data) |
read from an incoming double* array | |
void | write_data (double *response_data) |
write to an incoming double* array | |
void | overlay (const Response &response) |
add incoming response to functionValues/Gradients/Hessians | |
void | update (const Response &response, bool pull_metadata=false) |
Used in place of operator= when only results data updates are desired (functionValues/functionGradients/functionHessians are updated, ASV/labels/id's/etc. are not). Care is taken to allow different derivative array sizing between the two response objects. | |
void | update (const RealVector &source_fn_vals, const RealMatrix &source_fn_grads, const RealSymMatrixArray &source_fn_hessians, const ActiveSet &source_set) |
Overloaded form which allows update from components of a response object. Care is taken to allow different derivative array sizing. | |
void | update_partial (size_t start_index_target, size_t num_items, const Response &response, size_t start_index_source) |
partial update of this response object from another response object. The response objects may have different numbers of response functions. | |
void | update_partial (size_t start_index_target, size_t num_items, const RealVector &source_fn_vals, const RealMatrix &source_fn_grads, const RealSymMatrixArray &source_fn_hessians, const ActiveSet &source_set, size_t start_index_source) |
Overloaded form which allows partial update from components of a response object. The response objects may have different numbers of response functions. | |
void | reshape (size_t num_fns, size_t num_params, bool grad_flag, bool hess_flag) |
rehapes response data arrays | |
void | reshape_metadata (size_t num_meta) |
rehapes response metadata arrays | |
void | reset () |
resets all response data to zero | |
void | reset_inactive () |
resets all inactive response data to zero | |
bool | is_null () const |
function to check responseRep (does this handle contain a body) | |
Protected Member Functions | |
void | copy_rep (std::shared_ptr< Response > source_resp_rep) override |
Specialization of copy_rep; pulls base class data as well as derived specific data from the source rep into the this object. | |
Protected Member Functions inherited from Response | |
Response (BaseConstructor, const Variables &vars, const ProblemDescDB &problem_db) | |
constructor initializes the base class part of letter classes (BaseConstructor overloading avoids infinite recursion in the derived class constructors - Coplien, p. 139) | |
Response (BaseConstructor, const SharedResponseData &srd, const ActiveSet &set) | |
constructor initializes the base class part of letter classes (BaseConstructor overloading avoids infinite recursion in the derived class constructors - Coplien, p. 139) | |
Response (BaseConstructor, const ActiveSet &set) | |
constructor initializes the base class part of letter classes (BaseConstructor overloading avoids infinite recursion in the derived class constructors - Coplien, p. 139) | |
Response (BaseConstructor, const SharedResponseData &srd) | |
constructor initializes the base class part of letter classes (BaseConstructor overloading avoids infinite recursion in the derived class constructors - Coplien, p. 139) | |
Private Attributes | |
ExperimentCovariance | expDataCovariance |
sigma terms... | |
Additional Inherited Members | |
Protected Attributes inherited from Response | |
SharedResponseData | sharedRespData |
reference-counted instance of shared response data: id's, labels | |
RealVector | functionValues |
Abstract set of response functions. Ordered: [primary_scalar, primary_field, nonlinear_inequality, nonlinear_equality]. | |
RealMatrix | functionGradients |
first derivatives of the response functions More... | |
RealSymMatrixArray | functionHessians |
second derivatives of the response functions | |
IntRealMatrixMap | fieldCoords |
coordinates (independent vars like x,t) on which field values depend | |
ActiveSet | responseActiveSet |
copy of the ActiveSet used by the Model to generate a Response instance | |
std::vector< RespMetadataT > | metaData |
metadata storage | |
Container class for response functions and their derivatives.
ExperimentResponse provides the body class.
The ExperimentResponse class is the "representation" of the response container class. It is the "body" portion of the "handle-body idiom" (see Coplien "Advanced C++", p. 58). The handle class (Response) provides for memory efficiency in management of multiple response objects through reference counting and representation sharing. The body class (ExperimentResponse) actually contains the response data (functionValues, functionGradients, functionHessians, etc.). The representation is hidden in that an instance of ExperimentResponse may only be created by Response. Therefore, programmers create instances of the Response handle class, and only need to be aware of the handle/body mechanisms when it comes to managing shallow copies (shared representation) versus deep copies (separate representation used for history mechanisms).