letter_recog.cpp 17.5 KB
Newer Older
wester committed
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562
#include "opencv2/core/core.hpp"
#include "opencv2/ml/ml.hpp"

#include <cstdio>
#include <vector>
#include <iostream>

using namespace std;
using namespace cv;
using namespace cv::ml;

static void help()
{
    printf("\nThe sample demonstrates how to train Random Trees classifier\n"
    "(or Boosting classifier, or MLP, or Knearest, or Nbayes, or Support Vector Machines - see main()) using the provided dataset.\n"
    "\n"
    "We use the sample database letter-recognition.data\n"
    "from UCI Repository, here is the link:\n"
    "\n"
    "Newman, D.J. & Hettich, S. & Blake, C.L. & Merz, C.J. (1998).\n"
    "UCI Repository of machine learning databases\n"
    "[http://www.ics.uci.edu/~mlearn/MLRepository.html].\n"
    "Irvine, CA: University of California, Department of Information and Computer Science.\n"
    "\n"
    "The dataset consists of 20000 feature vectors along with the\n"
    "responses - capital latin letters A..Z.\n"
    "The first 16000 (10000 for boosting)) samples are used for training\n"
    "and the remaining 4000 (10000 for boosting) - to test the classifier.\n"
    "======================================================\n");
    printf("\nThis is letter recognition sample.\n"
            "The usage: letter_recog [-data=<path to letter-recognition.data>] \\\n"
            "  [-save=<output XML file for the classifier>] \\\n"
            "  [-load=<XML file with the pre-trained classifier>] \\\n"
            "  [-boost|-mlp|-knearest|-nbayes|-svm] # to use boost/mlp/knearest/SVM classifier instead of default Random Trees\n" );
}

// This function reads data and responses from the file <filename>
static bool
read_num_class_data( const string& filename, int var_count,
                     Mat* _data, Mat* _responses )
{
    const int M = 1024;
    char buf[M+2];

    Mat el_ptr(1, var_count, CV_32F);
    int i;
    vector<int> responses;

    _data->release();
    _responses->release();

    FILE* f = fopen( filename.c_str(), "rt" );
    if( !f )
    {
        cout << "Could not read the database " << filename << endl;
        return false;
    }

    for(;;)
    {
        char* ptr;
        if( !fgets( buf, M, f ) || !strchr( buf, ',' ) )
            break;
        responses.push_back((int)buf[0]);
        ptr = buf+2;
        for( i = 0; i < var_count; i++ )
        {
            int n = 0;
            sscanf( ptr, "%f%n", &el_ptr.at<float>(i), &n );
            ptr += n + 1;
        }
        if( i < var_count )
            break;
        _data->push_back(el_ptr);
    }
    fclose(f);
    Mat(responses).copyTo(*_responses);

    cout << "The database " << filename << " is loaded.\n";

    return true;
}

template<typename T>
static Ptr<T> load_classifier(const string& filename_to_load)
{
    // load classifier from the specified file
    Ptr<T> model = StatModel::load<T>( filename_to_load );
    if( model.empty() )
        cout << "Could not read the classifier " << filename_to_load << endl;
    else
        cout << "The classifier " << filename_to_load << " is loaded.\n";

    return model;
}

static Ptr<TrainData>
prepare_train_data(const Mat& data, const Mat& responses, int ntrain_samples)
{
    Mat sample_idx = Mat::zeros( 1, data.rows, CV_8U );
    Mat train_samples = sample_idx.colRange(0, ntrain_samples);
    train_samples.setTo(Scalar::all(1));

    int nvars = data.cols;
    Mat var_type( nvars + 1, 1, CV_8U );
    var_type.setTo(Scalar::all(VAR_ORDERED));
    var_type.at<uchar>(nvars) = VAR_CATEGORICAL;

    return TrainData::create(data, ROW_SAMPLE, responses,
                             noArray(), sample_idx, noArray(), var_type);
}

inline TermCriteria TC(int iters, double eps)
{
    return TermCriteria(TermCriteria::MAX_ITER + (eps > 0 ? TermCriteria::EPS : 0), iters, eps);
}

static void test_and_save_classifier(const Ptr<StatModel>& model,
                                     const Mat& data, const Mat& responses,
                                     int ntrain_samples, int rdelta,
                                     const string& filename_to_save)
{
    int i, nsamples_all = data.rows;
    double train_hr = 0, test_hr = 0;

    // compute prediction error on train and test data
    for( i = 0; i < nsamples_all; i++ )
    {
        Mat sample = data.row(i);

        float r = model->predict( sample );
        r = std::abs(r + rdelta - responses.at<int>(i)) <= FLT_EPSILON ? 1.f : 0.f;

        if( i < ntrain_samples )
            train_hr += r;
        else
            test_hr += r;
    }

    test_hr /= nsamples_all - ntrain_samples;
    train_hr = ntrain_samples > 0 ? train_hr/ntrain_samples : 1.;

    printf( "Recognition rate: train = %.1f%%, test = %.1f%%\n",
            train_hr*100., test_hr*100. );

    if( !filename_to_save.empty() )
    {
        model->save( filename_to_save );
    }
}


static bool
build_rtrees_classifier( const string& data_filename,
                         const string& filename_to_save,
                         const string& filename_to_load )
{
    Mat data;
    Mat responses;
    bool ok = read_num_class_data( data_filename, 16, &data, &responses );
    if( !ok )
        return ok;

    Ptr<RTrees> model;

    int nsamples_all = data.rows;
    int ntrain_samples = (int)(nsamples_all*0.8);

    // Create or load Random Trees classifier
    if( !filename_to_load.empty() )
    {
        model = load_classifier<RTrees>(filename_to_load);
        if( model.empty() )
            return false;
        ntrain_samples = 0;
    }
    else
    {
        // create classifier by using <data> and <responses>
        cout << "Training the classifier ...\n";
//        Params( int maxDepth, int minSampleCount,
//                   double regressionAccuracy, bool useSurrogates,
//                   int maxCategories, const Mat& priors,
//                   bool calcVarImportance, int nactiveVars,
//                   TermCriteria termCrit );
        Ptr<TrainData> tdata = prepare_train_data(data, responses, ntrain_samples);
        model = RTrees::create();
        model->setMaxDepth(10);
        model->setMinSampleCount(10);
        model->setRegressionAccuracy(0);
        model->setUseSurrogates(false);
        model->setMaxCategories(15);
        model->setPriors(Mat());
        model->setCalculateVarImportance(true);
        model->setActiveVarCount(4);
        model->setTermCriteria(TC(100,0.01f));
        model->train(tdata);
        cout << endl;
    }

    test_and_save_classifier(model, data, responses, ntrain_samples, 0, filename_to_save);
    cout << "Number of trees: " << model->getRoots().size() << endl;

    // Print variable importance
    Mat var_importance = model->getVarImportance();
    if( !var_importance.empty() )
    {
        double rt_imp_sum = sum( var_importance )[0];
        printf("var#\timportance (in %%):\n");
        int i, n = (int)var_importance.total();
        for( i = 0; i < n; i++ )
            printf( "%-2d\t%-4.1f\n", i, 100.f*var_importance.at<float>(i)/rt_imp_sum);
    }

    return true;
}


static bool
build_boost_classifier( const string& data_filename,
                        const string& filename_to_save,
                        const string& filename_to_load )
{
    const int class_count = 26;
    Mat data;
    Mat responses;
    Mat weak_responses;

    bool ok = read_num_class_data( data_filename, 16, &data, &responses );
    if( !ok )
        return ok;

    int i, j, k;
    Ptr<Boost> model;

    int nsamples_all = data.rows;
    int ntrain_samples = (int)(nsamples_all*0.5);
    int var_count = data.cols;

    // Create or load Boosted Tree classifier
    if( !filename_to_load.empty() )
    {
        model = load_classifier<Boost>(filename_to_load);
        if( model.empty() )
            return false;
        ntrain_samples = 0;
    }
    else
    {
        // !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
        //
        // As currently boosted tree classifier in MLL can only be trained
        // for 2-class problems, we transform the training database by
        // "unrolling" each training sample as many times as the number of
        // classes (26) that we have.
        //
        // !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

        Mat new_data( ntrain_samples*class_count, var_count + 1, CV_32F );
        Mat new_responses( ntrain_samples*class_count, 1, CV_32S );

        // 1. unroll the database type mask
        printf( "Unrolling the database...\n");
        for( i = 0; i < ntrain_samples; i++ )
        {
            const float* data_row = data.ptr<float>(i);
            for( j = 0; j < class_count; j++ )
            {
                float* new_data_row = (float*)new_data.ptr<float>(i*class_count+j);
                memcpy(new_data_row, data_row, var_count*sizeof(data_row[0]));
                new_data_row[var_count] = (float)j;
                new_responses.at<int>(i*class_count + j) = responses.at<int>(i) == j+'A';
            }
        }

        Mat var_type( 1, var_count + 2, CV_8U );
        var_type.setTo(Scalar::all(VAR_ORDERED));
        var_type.at<uchar>(var_count) = var_type.at<uchar>(var_count+1) = VAR_CATEGORICAL;

        Ptr<TrainData> tdata = TrainData::create(new_data, ROW_SAMPLE, new_responses,
                                                 noArray(), noArray(), noArray(), var_type);
        vector<double> priors(2);
        priors[0] = 1;
        priors[1] = 26;

        cout << "Training the classifier (may take a few minutes)...\n";
        model = Boost::create();
        model->setBoostType(Boost::GENTLE);
        model->setWeakCount(100);
        model->setWeightTrimRate(0.95);
        model->setMaxDepth(5);
        model->setUseSurrogates(false);
        model->setPriors(Mat(priors));
        model->train(tdata);
        cout << endl;
    }

    Mat temp_sample( 1, var_count + 1, CV_32F );
    float* tptr = temp_sample.ptr<float>();

    // compute prediction error on train and test data
    double train_hr = 0, test_hr = 0;
    for( i = 0; i < nsamples_all; i++ )
    {
        int best_class = 0;
        double max_sum = -DBL_MAX;
        const float* ptr = data.ptr<float>(i);
        for( k = 0; k < var_count; k++ )
            tptr[k] = ptr[k];

        for( j = 0; j < class_count; j++ )
        {
            tptr[var_count] = (float)j;
            float s = model->predict( temp_sample, noArray(), StatModel::RAW_OUTPUT );
            if( max_sum < s )
            {
                max_sum = s;
                best_class = j + 'A';
            }
        }

        double r = std::abs(best_class - responses.at<int>(i)) < FLT_EPSILON ? 1 : 0;
        if( i < ntrain_samples )
            train_hr += r;
        else
            test_hr += r;
    }

    test_hr /= nsamples_all-ntrain_samples;
    train_hr = ntrain_samples > 0 ? train_hr/ntrain_samples : 1.;
    printf( "Recognition rate: train = %.1f%%, test = %.1f%%\n",
            train_hr*100., test_hr*100. );

    cout << "Number of trees: " << model->getRoots().size() << endl;

    // Save classifier to file if needed
    if( !filename_to_save.empty() )
        model->save( filename_to_save );

    return true;
}


static bool
build_mlp_classifier( const string& data_filename,
                      const string& filename_to_save,
                      const string& filename_to_load )
{
    const int class_count = 26;
    Mat data;
    Mat responses;

    bool ok = read_num_class_data( data_filename, 16, &data, &responses );
    if( !ok )
        return ok;

    Ptr<ANN_MLP> model;

    int nsamples_all = data.rows;
    int ntrain_samples = (int)(nsamples_all*0.8);

    // Create or load MLP classifier
    if( !filename_to_load.empty() )
    {
        model = load_classifier<ANN_MLP>(filename_to_load);
        if( model.empty() )
            return false;
        ntrain_samples = 0;
    }
    else
    {
        // !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
        //
        // MLP does not support categorical variables by explicitly.
        // So, instead of the output class label, we will use
        // a binary vector of <class_count> components for training and,
        // therefore, MLP will give us a vector of "probabilities" at the
        // prediction stage
        //
        // !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

        Mat train_data = data.rowRange(0, ntrain_samples);
        Mat train_responses = Mat::zeros( ntrain_samples, class_count, CV_32F );

        // 1. unroll the responses
        cout << "Unrolling the responses...\n";
        for( int i = 0; i < ntrain_samples; i++ )
        {
            int cls_label = responses.at<int>(i) - 'A';
            train_responses.at<float>(i, cls_label) = 1.f;
        }

        // 2. train classifier
        int layer_sz[] = { data.cols, 100, 100, class_count };
        int nlayers = (int)(sizeof(layer_sz)/sizeof(layer_sz[0]));
        Mat layer_sizes( 1, nlayers, CV_32S, layer_sz );

#if 1
        int method = ANN_MLP::BACKPROP;
        double method_param = 0.001;
        int max_iter = 300;
#else
        int method = ANN_MLP::RPROP;
        double method_param = 0.1;
        int max_iter = 1000;
#endif

        Ptr<TrainData> tdata = TrainData::create(train_data, ROW_SAMPLE, train_responses);

        cout << "Training the classifier (may take a few minutes)...\n";
        model = ANN_MLP::create();
        model->setLayerSizes(layer_sizes);
        model->setActivationFunction(ANN_MLP::SIGMOID_SYM, 0, 0);
        model->setTermCriteria(TC(max_iter,0));
        model->setTrainMethod(method, method_param);
        model->train(tdata);
        cout << endl;
    }

    test_and_save_classifier(model, data, responses, ntrain_samples, 'A', filename_to_save);
    return true;
}

static bool
build_knearest_classifier( const string& data_filename, int K )
{
    Mat data;
    Mat responses;
    bool ok = read_num_class_data( data_filename, 16, &data, &responses );
    if( !ok )
        return ok;


    int nsamples_all = data.rows;
    int ntrain_samples = (int)(nsamples_all*0.8);

    // create classifier by using <data> and <responses>
    cout << "Training the classifier ...\n";
    Ptr<TrainData> tdata = prepare_train_data(data, responses, ntrain_samples);
    Ptr<KNearest> model = KNearest::create();
    model->setDefaultK(K);
    model->setIsClassifier(true);
    model->train(tdata);
    cout << endl;

    test_and_save_classifier(model, data, responses, ntrain_samples, 0, string());
    return true;
}

static bool
build_nbayes_classifier( const string& data_filename )
{
    Mat data;
    Mat responses;
    bool ok = read_num_class_data( data_filename, 16, &data, &responses );
    if( !ok )
        return ok;

    Ptr<NormalBayesClassifier> model;

    int nsamples_all = data.rows;
    int ntrain_samples = (int)(nsamples_all*0.8);

    // create classifier by using <data> and <responses>
    cout << "Training the classifier ...\n";
    Ptr<TrainData> tdata = prepare_train_data(data, responses, ntrain_samples);
    model = NormalBayesClassifier::create();
    model->train(tdata);
    cout << endl;

    test_and_save_classifier(model, data, responses, ntrain_samples, 0, string());
    return true;
}

static bool
build_svm_classifier( const string& data_filename,
                      const string& filename_to_save,
                      const string& filename_to_load )
{
    Mat data;
    Mat responses;
    bool ok = read_num_class_data( data_filename, 16, &data, &responses );
    if( !ok )
        return ok;

    Ptr<SVM> model;

    int nsamples_all = data.rows;
    int ntrain_samples = (int)(nsamples_all*0.8);

    // Create or load Random Trees classifier
    if( !filename_to_load.empty() )
    {
        model = load_classifier<SVM>(filename_to_load);
        if( model.empty() )
            return false;
        ntrain_samples = 0;
    }
    else
    {
        // create classifier by using <data> and <responses>
        cout << "Training the classifier ...\n";
        Ptr<TrainData> tdata = prepare_train_data(data, responses, ntrain_samples);
        model = SVM::create();
        model->setType(SVM::C_SVC);
        model->setKernel(SVM::LINEAR);
        model->setC(1);
        model->train(tdata);
        cout << endl;
    }

    test_and_save_classifier(model, data, responses, ntrain_samples, 0, filename_to_save);
    return true;
}

int main( int argc, char *argv[] )
{
    string filename_to_save = "";
    string filename_to_load = "";
    string data_filename;
    int method = 0;

    cv::CommandLineParser parser(argc, argv, "{data|../data/letter-recognition.data|}{save||}{load||}{boost||}"
            "{mlp||}{knn knearest||}{nbayes||}{svm||}{help h||}");
    data_filename = parser.get<string>("data");
    if (parser.has("save"))
        filename_to_save = parser.get<string>("save");
    if (parser.has("load"))
        filename_to_load = parser.get<string>("load");
    if (parser.has("boost"))
        method = 1;
    else if (parser.has("mlp"))
        method = 2;
    else if (parser.has("knearest"))
        method = 3;
    else if (parser.has("nbayes"))
        method = 4;
    else if (parser.has("svm"))
        method = 5;
    if (parser.has("help"))
    {
        help();
        return 0;
    }
    if( (method == 0 ?
        build_rtrees_classifier( data_filename, filename_to_save, filename_to_load ) :
        method == 1 ?
        build_boost_classifier( data_filename, filename_to_save, filename_to_load ) :
        method == 2 ?
        build_mlp_classifier( data_filename, filename_to_save, filename_to_load ) :
        method == 3 ?
        build_knearest_classifier( data_filename, 10 ) :
        method == 4 ?
        build_nbayes_classifier( data_filename) :
        method == 5 ?
        build_svm_classifier( data_filename, filename_to_save, filename_to_load ):
        -1) < 0)
    {
        help();
    }
    return 0;
}