demo33t.cpp File Reference

Example 33t: Threaded Oscillating Search in very high-dimensional feature selection. More...

#include <boost/smart_ptr.hpp>
#include <exception>
#include <iostream>
#include <cstdlib>
#include <string>
#include <vector>
#include "error.hpp"
#include "global.hpp"
#include "subset.hpp"
#include "data_intervaller.hpp"
#include "data_splitter.hpp"
#include "data_splitter_randrand.hpp"
#include "data_scaler.hpp"
#include "data_scaler_void.hpp"
#include "data_accessor_splitting_memTRN.hpp"
#include "data_accessor_splitting_memARFF.hpp"
#include "criterion_multinom_bhattacharyya.hpp"
#include "criterion_wrapper.hpp"
#include "classifier_multinom_naivebayes.hpp"
#include "search_bif.hpp"
#include "seq_step_straight_threaded.hpp"
#include "search_seq_os.hpp"

Include dependency graph for demo33t.cpp:

Functions
int	main ()

---

Detailed Description

Example 33t: Threaded Oscillating Search in very high-dimensional feature selection.

---

Function Documentation

int main

(

)

Example 33t: Threaded Oscillating Search in very high-dimensional feature selection.

As alternative to standard sequential evaluation of feature subset candidates FST3 enables threaded candidate subset evaluation. All FST3 sequential search methods can be easily parallelized by using Sequential_Step_Straight_Threaded instead of Sequential_Step_Straight evaluator object. The actual search speed gain depends on particular problem setting. In small-sample, low-dimensional settings, or when criterion evaluation is very fast, the actual gain may remain negligible or even negative due to thread management overhead (see Example 11: Wrapper-based feature selection with Floating Search.). However, in computationally more complex cases as illustrated here on the threaded version of very-high-dimensional problem (see Example 33: Oscillating Search in very high-dimensional feature selection.) the gain is becoming substantial.

Note:

With many higher-dimensional problems the FST3 threading capability can become key in making the feature selection task tractable. Note that maximum permitted number of threads to run at once is to be user-specified depending on hardware capabilities.

References FST::Search_OS< RETURNTYPE, DIMTYPE, SUBSET, CRITERION, EVALUATOR >::search(), FST::Search_BIF< RETURNTYPE, DIMTYPE, SUBSET, CRITERION >::search(), and FST::Search_OS< RETURNTYPE, DIMTYPE, SUBSET, CRITERION, EVALUATOR >::set_delta().