PPCKO: Principal Predictive Components for Estimating an Autoregressive Operator
 
Loading...
Searching...
No Matches
PPC_KO_wrapper_cv_k< solver, k_imp, valid_err_ret, cv_strat, cv_err_eval > Class Template Reference

Derived-from-PPC_KO_wrapper class for wrapping class that performs PPCKO computations with cross-validation on the number of retained PPCs. More...

#include <PPC_KO_wrapper.hpp>

Inheritance diagram for PPC_KO_wrapper_cv_k< solver, k_imp, valid_err_ret, cv_strat, cv_err_eval >:
PPC_KO_wrapper< solver, k_imp, valid_err_ret, cv_strat, cv_err_eval >

Public Member Functions

template<typename STOR_OBJ>
 PPC_KO_wrapper_cv_k (STOR_OBJ &&data, double alpha, const std::vector< int > &k_s, double toll, int min_size_ts, int max_size_ts, int number_threads)
 Constructor.
 
void call_ko () override
 Override for calling the number of retained PPCs cross-validation PPCKO version at runtime.
 
- Public Member Functions inherited from PPC_KO_wrapper< solver, k_imp, valid_err_ret, cv_strat, cv_err_eval >
template<typename STOR_OBJ>
 PPC_KO_wrapper (STOR_OBJ &&data, int number_threads)
 Constructor.
 
virtual ~PPC_KO_wrapper ()=default
 Virtual destructor.
 
KO_Traits::StoringMatrix data () const
 Getter for the data matrix.
 
results_t< valid_err_ret > results () const
 Getter for the results.
 
int number_threads () const
 Getter for the number of threads for OMP.
 
results_t< valid_err_ret > & results ()
 Setter for the results.
 

Detailed Description

template<SOLVER solver, K_IMP k_imp, VALID_ERR_RET valid_err_ret, CV_STRAT cv_strat, CV_ERR_EVAL cv_err_eval>
class PPC_KO_wrapper_cv_k< solver, k_imp, valid_err_ret, cv_strat, cv_err_eval >

Derived-from-PPC_KO_wrapper class for wrapping class that performs PPCKO computations with cross-validation on the number of retained PPCs.

Template Parameters
solverif algorithm solved inverting the regularized covariance or avoiding it through gep (not possible if retaining the number of PPCs with explanatory power criterion)
k_impif k is imposed or has to be found through explanatory power criterion
valid_err_retif validation error are stored
cv_stratstrategy for splitting training/validation sets
err_evalhow to evaluate the loss between prediction on validation set and validation set

It is a derived class. Polymorphism is known at run-time through virtual polymorphism

Constructor & Destructor Documentation

◆ PPC_KO_wrapper_cv_k()

template<SOLVER solver, K_IMP k_imp, VALID_ERR_RET valid_err_ret, CV_STRAT cv_strat, CV_ERR_EVAL cv_err_eval>
template<typename STOR_OBJ>
PPC_KO_wrapper_cv_k< solver, k_imp, valid_err_ret, cv_strat, cv_err_eval >::PPC_KO_wrapper_cv_k ( STOR_OBJ && data,
double alpha,
const std::vector< int > & k_s,
double toll,
int min_size_ts,
int max_size_ts,
int number_threads )
inline

Constructor.

Parameters
datamatrix storing fts
alpharegularization parameter
k_snumber of retained PPCs input space
tollthe cv on the number of retained PPCs continues only if between two parameters, that are checked in increasing order, the absolute difference between two validation errors is bigger than tolerance*trace(covariance). If not, stops and look for k only between the tested ones
min_size_tsminimum size (number of time instants) of the training set
max_size_tsmaximum size (number of time instants) of the training set
number_threadsnumber of threads for OMP

Universal constructor: move semantic used to optimazing handling big size objects

Member Function Documentation

◆ call_ko()

template<SOLVER solver, K_IMP k_imp, VALID_ERR_RET valid_err_ret, CV_STRAT cv_strat, CV_ERR_EVAL cv_err_eval>
void PPC_KO_wrapper_cv_k< solver, k_imp, valid_err_ret, cv_strat, cv_err_eval >::call_ko ( )
overridevirtual

Override for calling the number of retained PPCs cross-validation PPCKO version at runtime.

Number of retained PPCs cross-validation overriding: wraps the class for computations (static polymorphism) accordingly.

Wraps the class for computations, performs them and then update the results in the wrapper class

Implements PPC_KO_wrapper< solver, k_imp, valid_err_ret, cv_strat, cv_err_eval >.


The documentation for this class was generated from the following files: