classifier_tester.cpp File Reference

#include <stdio.h>
#include "base/commandlineflags.h"
#include "baseapi.h"
#include "commontraining.h"
#include "cubeclassifier.h"
#include "mastertrainer.h"
#include "params.h"
#include "strngs.h"
#include "tessclassifier.h"

Go to the source code of this file.

Enumerations
enum	ClassifierName {   CN_PRUNER, CN_FULL, CN_CUBE, CN_CUBETESS,   CN_COUNT }

Functions
	STRING_PARAM_FLAG (classifier, "", "Classifier to test")
	STRING_PARAM_FLAG (lang, "eng", "Language to test")
	STRING_PARAM_FLAG (tessdata_dir, "", "Directory of traineddata files")
	DECLARE_INT_PARAM_FLAG (debug_level)
	DECLARE_STRING_PARAM_FLAG (T)
int	main (int argc, char **argv)

Variables
const char *	names []

Enumeration Type Documentation

ClassifierName

enum ClassifierName

Enumerator
CN_PRUNER
CN_FULL
CN_CUBE
CN_CUBETESS
CN_COUNT

Definition at line 39 of file classifier_tester.cpp.

                    {
  CN_PRUNER,
  CN_FULL,
#ifndef NO_CUBE_BUILD
  CN_CUBE,
  CN_CUBETESS,
#endif  // NO_CUBE_BUILD
  CN_COUNT
};

Function Documentation

DECLARE_INT_PARAM_FLAG()

DECLARE_INT_PARAM_FLAG

(

debug_level

)

DECLARE_STRING_PARAM_FLAG()

DECLARE_STRING_PARAM_FLAG

(

T

)

main()

int main	(	int	argc,
		char **	argv
	)

This program reads in a text file consisting of feature samples from a training page in the following format:

   FontName UTF8-char-str xmin ymin xmax ymax page-number
    NumberOfFeatureTypes(N)
      FeatureTypeName1 NumberOfFeatures(M)
         Feature1
         ...
         FeatureM
      FeatureTypeName2 NumberOfFeatures(M)
         Feature1
         ...
         FeatureM
      ...
      FeatureTypeNameN NumberOfFeatures(M)
         Feature1
         ...
         FeatureM
   FontName CharName ...

The result of this program is a binary inttemp file used by the OCR engine.

Parameters

argc	number of command line arguments
argv	array of command line arguments

Returns

none

Note

Exceptions: none

History: Fri Aug 18 08:56:17 1989, DSJ, Created.

History: Mon May 18 1998, Christy Russson, Revistion started.

Definition at line 149 of file classifier_tester.cpp.

                                {
  ParseArguments(&argc, &argv);
  STRING file_prefix;
  tesseract::MasterTrainer* trainer = tesseract::LoadTrainingData(
      argc, argv, false, NULL, &file_prefix);
  tesseract::TessBaseAPI* api;
  // Decode the classifier string.
  tesseract::ShapeClassifier* shape_classifier = InitializeClassifier(
      FLAGS_classifier.c_str(), trainer->unicharset(), argc, argv, &api);
  if (shape_classifier == NULL) {
    fprintf(stderr, "Classifier init failed!:%s\n", FLAGS_classifier.c_str());
    return 1;
  }

  // We want to test junk as well if it is available.
  // trainer->IncludeJunk();
  // We want to test with replicated samples too.
  trainer->ReplicateAndRandomizeSamplesIfRequired();

  trainer->TestClassifierOnSamples(tesseract:: CT_UNICHAR_TOP1_ERR,
                                   MAX(3, FLAGS_debug_level), false,
                                   shape_classifier, NULL);
  delete shape_classifier;
  delete api;
  delete trainer;

  return 0;
} /* main */

STRING_PARAM_FLAG() [1/3]

STRING_PARAM_FLAG	(	classifier	,
		""	,
		"Classifier to test"
	)

STRING_PARAM_FLAG() [2/3]

STRING_PARAM_FLAG	(	lang	,
		"eng"	,
		"Language to test"
	)

STRING_PARAM_FLAG() [3/3]

STRING_PARAM_FLAG	(	tessdata_dir	,
		""	,
		"Directory of traineddata files"
	)

Variable Documentation

names

const char* names[]

Initial value:

= {"pruner", "full",

                       "cube", "cubetess",

                       NULL}

Definition at line 49 of file classifier_tester.cpp.