source: trunk/test/crossvalidation_test.cc @ 514

// $Id: crossvalidation_test.cc 514 2006-02-20 09:45:34Z peter $

#include <c++_tools/classifier/CrossSplitter.h>
#include <c++_tools/classifier/MatrixLookup.h>
#include <c++_tools/classifier/Target.h>
#include <c++_tools/gslapi/matrix.h>

#include <cstdlib>
#include <fstream>
#include <iostream>
#include <string>
#include <vector>

int main(const int argc,const char* argv[])
{  
  using namespace theplu;

  std::ostream* error;
  if (argc>1 && argv[1]==std::string("-v"))
    error = &std::cerr;
  else {
    error = new std::ofstream("/dev/null");
    if (argc>1)
      std::cout << "crossvalidation_test -v : for printing extra information\n";
  }
  *error << "testing crosssplitter" << std::endl;
  bool ok = true;

  std::vector<std::string> label(10,"default");
  label[2]=label[7]="white";
  label[4]=label[5]="black";
  label[6]=label[3]="green";
  label[8]=label[9]="red";
                 
  classifier::Target target(label);
  gslapi::matrix raw_data(10,10);
  classifier::MatrixLookup data(raw_data);
  classifier::CrossSplitter cv(target,data,3,3);
  
  std::vector<size_t> sample_count(10,0);
  for (cv.reset(); cv.more(); cv.next()){
    std::vector<size_t> class_count(5,0);
    if (cv.training_index().size()+cv.validation_index().size()!=target.size()){
      ok = false;
      *error << "ERROR: size of training samples plus " 
             << "size of validation samples is invalid." << std::endl;
    }
    if (cv.validation_index().size()!=3 && cv.validation_index().size()!=4){
      ok = false;
      *error << "ERROR: size of validation samples is invalid." 
             << "expected size to be 3 or 4" << std::endl;
    }
    for (size_t i=0; i<cv.validation_index().size(); i++) {
      assert(cv.validation_index()[i]<sample_count.size());
      sample_count[cv.validation_index()[i]]++;
    }
    for (size_t i=0; i<cv.training_index().size(); i++) {
      class_count[target(cv.training_index()[i])]++;
    }
    for (size_t i=0; i<class_count.size(); i++)
      if (class_count[i]==0){
        ok = false;
        *error << "ERROR: class " << i << " was not in training set." 
                  << " Expected at least one sample from each class." 
                  << std::endl;
      }
  }
  for (size_t i=0; i<sample_count.size(); i++){
    if (sample_count[i]!=1){
      ok = false;
      *error << "ERROR: sample " << i << " was validated " << sample_count[i] 
                << " times." << " Expected to be 1 time" << std::endl;
    }
  }
  
  if (error!=&std::cerr)
    delete error;

  if (ok)
    return 0;
  return -1;
}