Common subdirectories: modules/ml/src//.svn and modules/ml/src.update//.svn
diff -c modules/ml/src//em.cpp modules/ml/src.update//em.cpp
*** modules/ml/src//em.cpp	Thu Sep  1 14:10:21 2011
--- modules/ml/src.update//em.cpp	Thu Sep  1 12:42:16 2011
***************
*** 797,803 ****
      else
      {
          int max_count = 0;
! 
          CV_CALL( class_ranges = cvCreateMat( 1, nclusters+1, CV_32SC1 ));
          if( nclusters > 1 )
          {
--- 797,803 ----
      else
      {
          int max_count = 0;
! 	int nclusters_found;
          CV_CALL( class_ranges = cvCreateMat( 1, nclusters+1, CV_32SC1 ));
          if( nclusters > 1 )
          {
***************
*** 808,814 ****
                      params.means ? 1 : 10, 0.5 ), params.means );
  
              CV_CALL( cvSortSamplesByClasses( (const float**)train_data.data.fl,
!                                             labels, class_ranges->data.i ));
          }
          else
          {
--- 808,814 ----
                      params.means ? 1 : 10, 0.5 ), params.means );
  
              CV_CALL( cvSortSamplesByClasses( (const float**)train_data.data.fl,
! 					      labels, class_ranges->data.i, &nclusters_found ));
          }
          else
          {
***************
*** 816,822 ****
              class_ranges->data.i[1] = nsamples;
          }
  
!         for( i = 0; i < nclusters; i++ )
          {
              int left = class_ranges->data.i[i], right = class_ranges->data.i[i+1];
              max_count = MAX( max_count, right - left );
--- 816,822 ----
              class_ranges->data.i[1] = nsamples;
          }
  
!         for( i = 0; i < nclusters_found; i++ )
          {
              int left = class_ranges->data.i[i], right = class_ranges->data.i[i+1];
              max_count = MAX( max_count, right - left );
***************
*** 830,843 ****
              hdr[i] = hdr[0];
          }
  
!         for( i = 0; i < nclusters; i++ )
          {
              int left = class_ranges->data.i[i], right = class_ranges->data.i[i+1];
              int cluster_size = right - left;
              CvMat avg;
  
!             if( cluster_size <= 0 )
!                 continue;
  
              for( j = left; j < right; j++ )
                  hdr[j - left].data.fl = train_data.data.fl[j];
--- 830,842 ----
              hdr[i] = hdr[0];
          }
  
!         for( i = 0; i < nclusters_found; i++ )
          {
              int left = class_ranges->data.i[i], right = class_ranges->data.i[i+1];
              int cluster_size = right - left;
              CvMat avg;
  
!             // if( cluster_size <= 0 ) continue;
  
              for( j = left; j < right; j++ )
                  hdr[j - left].data.fl = train_data.data.fl[j];
***************
*** 847,852 ****
--- 846,856 ----
                  &avg, CV_COVAR_NORMAL | CV_COVAR_SCALE ));
              weights->data.db[i] = (double)cluster_size/(double)nsamples;
          }
+         for( i = nclusters_found; i < nclusters; i++ )
+         {
+ 	  cvZero(covs[i]);
+ 	  weights->data.db[i] = 0;
+ 	}
      }
  
      __END__;
***************
*** 859,873 ****
  
  
  void CvEM::kmeans( const CvVectors& train_data, int nclusters, CvMat* labels,
!                    CvTermCriteria termcrit, const CvMat* /*centers0*/ )
  {
!     int i, nsamples = train_data.count, dims = train_data.dims;
!     cv::Ptr<CvMat> temp_mat = cvCreateMat(nsamples, dims, CV_32F);
!     
      for( i = 0; i < nsamples; i++ )
!         memcpy( temp_mat->data.ptr + temp_mat->step*i, train_data.data.fl[i], dims*sizeof(float));
!     
!     cvKMeans2(temp_mat, nclusters, labels, termcrit, 10);
  }
  
  
--- 863,1052 ----
  
  
  void CvEM::kmeans( const CvVectors& train_data, int nclusters, CvMat* labels,
!                    CvTermCriteria termcrit, const CvMat* centers0 )
  {
!     CvMat* centers = 0;
!     CvMat* old_centers = 0;
!     CvMat* counters = 0;
! 
!     CV_FUNCNAME( "CvEM::kmeans" );
! 
!     __BEGIN__;
! 
! 
!     cv::RNG* rng = &cv::theRNG();
!     int i, j, k, nsamples, dims;
!     int iter = 0;
!     double max_dist = DBL_MAX;
! 
!     termcrit = cvCheckTermCriteria( termcrit, 1e-6, 100 );
!     termcrit.epsilon *= termcrit.epsilon;
!     nsamples = train_data.count;
!     dims = train_data.dims;
!     nclusters = MIN( nclusters, nsamples );
! 
!     CV_CALL( centers = cvCreateMat( nclusters, dims, CV_64FC1 ));
!     CV_CALL( old_centers = cvCreateMat( nclusters, dims, CV_64FC1 ));
!     CV_CALL( counters = cvCreateMat( 1, nclusters, CV_32SC1 ));
!     cvZero( old_centers );
! 
!     if( centers0 )
!     {
!         CV_CALL( cvConvert( centers0, centers ));
!     }
!     else
!     {
!         for( i = 0; i < nsamples; i++ )
!             labels->data.i[i] = i*nclusters/nsamples;
!         cvRandShuffle( labels, &rng->state );
!     }
! 
!     for( ;; )
!     {
!         CvMat* temp;
! 
!         if( iter > 0 || centers0 )
!         {
!             for( i = 0; i < nsamples; i++ )
!             {
!                 const float* s = train_data.data.fl[i];
!                 int k_best = 0;
!                 double min_dist = DBL_MAX;
! 
!                 for( k = 0; k < nclusters; k++ )
!                 {
!                     const double* c = (double*)(centers->data.ptr + k*centers->step);
!                     double dist = 0;
! 
!                     for( j = 0; j <= dims - 4; j += 4 )
!                     {
!                         double t0 = c[j] - s[j];
!                         double t1 = c[j+1] - s[j+1];
!                         dist += t0*t0 + t1*t1;
!                         t0 = c[j+2] - s[j+2];
!                         t1 = c[j+3] - s[j+3];
!                         dist += t0*t0 + t1*t1;
!                     }
! 
!                     for( ; j < dims; j++ )
!                     {
!                         double t = c[j] - s[j];
!                         dist += t*t;
!                     }
! 
!                     if( min_dist > dist )
!                     {
!                         min_dist = dist;
!                         k_best = k;
!                     }
!                 }
! 
!                 labels->data.i[i] = k_best;
!             }
!         }
! 
!         if( ++iter > termcrit.max_iter )
!             break;
! 
!         CV_SWAP( centers, old_centers, temp );
!         cvZero( centers );
!         cvZero( counters );
! 
!         // update centers
!         for( i = 0; i < nsamples; i++ )
!         {
!             const float* s = train_data.data.fl[i];
!             k = labels->data.i[i];
!             double* c = (double*)(centers->data.ptr + k*centers->step);
! 
!             for( j = 0; j <= dims - 4; j += 4 )
!             {
!                 double t0 = c[j] + s[j];
!                 double t1 = c[j+1] + s[j+1];
! 
!                 c[j] = t0;
!                 c[j+1] = t1;
! 
!                 t0 = c[j+2] + s[j+2];
!                 t1 = c[j+3] + s[j+3];
! 
!                 c[j+2] = t0;
!                 c[j+3] = t1;
!             }
!             for( ; j < dims; j++ )
!                 c[j] += s[j];
!             counters->data.i[k]++;
!         }
! 
!         if( iter > 1 )
!             max_dist = 0;
! 
!         for( k = 0; k < nclusters; k++ )
!         {
!             double* c = (double*)(centers->data.ptr + k*centers->step);
!             if( counters->data.i[k] != 0 )
!             {
!                 double scale = 1./counters->data.i[k];
!                 for( j = 0; j < dims; j++ )
!                     c[j] *= scale;
!             }
!             else
!             {
!                 const float* s;
!                 for( j = 0; j < 10; j++ )
!                 {
!                     i = (*rng)(nsamples);
!                     if( counters->data.i[labels->data.i[i]] > 1 )
!                         break;
!                 }
!                 s = train_data.data.fl[i];
!                 for( j = 0; j < dims; j++ )
!                     c[j] = s[j];
!             }
! 
!             if( iter > 1 )
!             {
!                 double dist = 0;
!                 const double* c_o = (double*)(old_centers->data.ptr + k*old_centers->step);
!                 for( j = 0; j < dims; j++ )
!                 {
!                     double t = c[j] - c_o[j];
!                     dist += t*t;
!                 }
!                 if( max_dist < dist )
!                     max_dist = dist;
!             }
!         }
! 
!         if( max_dist < termcrit.epsilon )
!             break;
!     }
! 
!     cvZero( counters );
      for( i = 0; i < nsamples; i++ )
!         counters->data.i[labels->data.i[i]]++;
! 
!     // ensure that we do not have empty clusters
!     for( k = 0; k < nclusters; k++ )
!         if( counters->data.i[k] == 0 )
!             for(;;)
!             {
!                 i = (*rng)(nsamples);
!                 j = labels->data.i[i];
!                 if( counters->data.i[j] > 1 )
!                 {
!                     labels->data.i[i] = k;
!                     counters->data.i[j]--;
!                     counters->data.i[k]++;
!                     break;
!                 }
!             }
! 
!     __END__;
! 
!     cvReleaseMat( &centers );
!     cvReleaseMat( &old_centers );
!     cvReleaseMat( &counters );
  }
  
  
Only in modules/ml/src.update/: em.cpp.original
diff -c modules/ml/src//inner_functions.cpp modules/ml/src.update//inner_functions.cpp
*** modules/ml/src//inner_functions.cpp	Thu Sep  1 14:10:28 2011
--- modules/ml/src.update//inner_functions.cpp	Wed Aug 31 14:03:20 2011
***************
*** 1026,1032 ****
  
  void
  cvSortSamplesByClasses( const float** samples, const CvMat* classes,
!                         int* class_ranges, const uchar** mask )
  {
      CvSampleResponsePair* pairs = 0;
      CV_FUNCNAME( "cvSortSamplesByClasses" );
--- 1026,1032 ----
  
  void
  cvSortSamplesByClasses( const float** samples, const CvMat* classes,
!                         int* const class_ranges, int* const num_classes_found, const uchar** mask )
  {
      CvSampleResponsePair* pairs = 0;
      CV_FUNCNAME( "cvSortSamplesByClasses" );
***************
*** 1035,1041 ****
  
      int i, k = 0, sample_count;
  
!     if( !samples || !classes || !class_ranges )
          CV_ERROR( CV_StsNullPtr, "INTERNAL ERROR: some of the args are NULL pointers" );
  
      if( classes->rows != 1 || CV_MAT_TYPE(classes->type) != CV_32SC1 )
--- 1035,1041 ----
  
      int i, k = 0, sample_count;
  
!     if( !samples || !classes || !class_ranges || !num_classes_found)
          CV_ERROR( CV_StsNullPtr, "INTERNAL ERROR: some of the args are NULL pointers" );
  
      if( classes->rows != 1 || CV_MAT_TYPE(classes->type) != CV_32SC1 )
***************
*** 1068,1073 ****
--- 1068,1075 ----
              class_ranges[++k] = i+1;
      }
  
+     *num_classes_found = k;
+ 
      __END__;
  
      cvFree( &pairs );
Only in modules/ml/src.update/: inner_functions.cpp.new
Only in modules/ml/src.update/: inner_functions.cpp.original
diff -c modules/ml/src//precomp.hpp modules/ml/src.update//precomp.hpp
*** modules/ml/src//precomp.hpp	Thu Sep  1 14:10:33 2011
--- modules/ml/src.update//precomp.hpp	Wed Aug 31 14:02:54 2011
***************
*** 299,305 ****
  
  void
  cvSortSamplesByClasses( const float** samples, const CvMat* classes, 
!                         int* class_ranges, const uchar** mask CV_DEFAULT(0) );
  
  void 
  cvCombineResponseMaps (CvMat*  _responses,
--- 299,305 ----
  
  void
  cvSortSamplesByClasses( const float** samples, const CvMat* classes, 
!                         int* const class_ranges, int* const num_classes_found, const uchar** mask CV_DEFAULT(0) );
  
  void 
  cvCombineResponseMaps (CvMat*  _responses,
Only in modules/ml/src.update/: precomp.hpp.original
diff -c modules/ml/src//svm.cpp modules/ml/src.update//svm.cpp
*** modules/ml/src//svm.cpp	Thu Sep  1 14:10:24 2011
--- modules/ml/src.update//svm.cpp	Thu Sep  1 13:38:56 2011
***************
*** 1396,1405 ****
                              sample_count*sizeof(temp_samples[0])));
          CV_CALL( temp_y = (schar*)cvMemStorageAlloc( temp_storage, sample_count));
  
!         class_ranges[class_count] = 0;
!         cvSortSamplesByClasses( samples, responses, class_ranges, 0 );
          //check that while cross-validation there were the samples from all the classes
!         if( class_ranges[class_count] <= 0 )
              CV_ERROR( CV_StsBadArg, "While cross-validation one or more of the classes have "
              "been fell out of the sample. Try to enlarge <CvSVMParams::k_fold>" );
  
--- 1396,1405 ----
                              sample_count*sizeof(temp_samples[0])));
          CV_CALL( temp_y = (schar*)cvMemStorageAlloc( temp_storage, sample_count));
  
! 	int class_count_found; 
!         cvSortSamplesByClasses( samples, responses, class_ranges, &class_count_found, 0 );
          //check that while cross-validation there were the samples from all the classes
! 	if (class_count_found < class_count)
              CV_ERROR( CV_StsBadArg, "While cross-validation one or more of the classes have "
              "been fell out of the sample. Try to enlarge <CvSVMParams::k_fold>" );
  
Only in modules/ml/src.update/: svm.cpp.original