demo24t.cpp File Reference

Example 24t: Threaded Monte Carlo - random feature subset search. 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_5050.hpp"
#include "data_splitter_cv.hpp"
#include "data_scaler.hpp"
#include "data_scaler_void.hpp"
#include "data_accessor_splitting_memTRN.hpp"
#include "data_accessor_splitting_memARFF.hpp"
#include "criterion_wrapper.hpp"
#include "distance_euclid.hpp"
#include "classifier_knn.hpp"
#include "search_monte_carlo_threaded.hpp"

Include dependency graph for demo24t.cpp:

Functions
int	main ()

---

Detailed Description

Example 24t: Threaded Monte Carlo - random feature subset search.

---

Function Documentation

int main

(

)

Example 24t: Threaded Monte Carlo - random feature subset search.

Pure random search does not give any guarantees as of the optimality of results. It is unlikely to yield better results than any of the sub-optimal or optimal methods implemented in FST, except in isolated cases. The advantage of random search is neverhtheless its quick initial "convergence"; often it is capable of quickly revealing a solution that is only marginally worse than solutions that would take considerably longer to find using more advanced methods. In this example features are selected using Monte Carlo algorithm and 3-NN wrapper classification accuracy as FS criterion. The stopping condition is specified as a time limit in seconds. Classification accuracy (i.e, FS wrapper criterion value) is estimated on the first 50% of data samples by means of 3-fold cross-validation. The final classification performance on the selected subspace is eventually validated on the second 50% of data. SFFS is called in d-optimizing setting, invoked by parameter 0 in search(0,...), which is otherwise used to specify the required subset size.

Note:

Random search requires a user specified stopping condition.

In addition to threaded execution this example differs from Example 24: Monte Carlo - random feature subset search. also by using a different setting of random subset cardinality generator.

References FST::Search_Monte_Carlo_Threaded< RETURNTYPE, DIMTYPE, SUBSET, CRITERION, max_threads >::search(), FST::Search_Monte_Carlo_Threaded< RETURNTYPE, DIMTYPE, SUBSET, CRITERION, max_threads >::set_cardinality_randomization(), FST::Search< RETURNTYPE, DIMTYPE, SUBSET, CRITERION >::set_output_detail(), and FST::Search_Monte_Carlo_Threaded< RETURNTYPE, DIMTYPE, SUBSET, CRITERION, max_threads >::set_stopping_condition().