good parallization

semseg/SemSegContextTree.cpp

@@ -196,7 +196,7 @@ double SemSegContextTree::getBestSplit ( std::vector<NICE::MultiChannelImageT<do
 
   try
   {
-    imgCount = (int)feats.size();
+    imgCount = ( int ) feats.size();
   }
   catch ( Exception )
   {
@@ -225,15 +225,15 @@ double SemSegContextTree::getBestSplit ( std::vector<NICE::MultiChannelImageT<do
     if ( forbidden_classes.find ( labelmapback[i] ) != forbidden_classes.end() )
       fraction[i] = 0;
     else
-      fraction[i] = ( (double)maxSamples ) / ( (double)featcounter * a[i] * a.size() );
+      fraction[i] = ( ( double ) maxSamples ) / ( ( double ) featcounter * a[i] * a.size() );
   }
 
   featcounter = 0;
 
   for ( int iCounter = 0; iCounter < imgCount; iCounter++ )
   {
-    int xsize = (int)currentfeats[iCounter].width();
-    int ysize = (int)currentfeats[iCounter].height();
+    int xsize = ( int ) currentfeats[iCounter].width();
+    int ysize = ( int ) currentfeats[iCounter].height();
 
     for ( int x = 0; x < xsize; x++ )
     {
@@ -242,7 +242,7 @@ double SemSegContextTree::getBestSplit ( std::vector<NICE::MultiChannelImageT<do
         if ( currentfeats[iCounter].get ( x, y, tree ) == node )
         {
           int cn = labels[iCounter] ( x, y );
-          double randD = (double)rand() / (double)RAND_MAX;
+          double randD = ( double ) rand() / ( double ) RAND_MAX;
 
           if ( labelmap.find ( cn ) == labelmap.end() )
             continue;
@@ -268,7 +268,7 @@ double SemSegContextTree::getBestSplit ( std::vector<NICE::MultiChannelImageT<do
 
   for ( mapit = e.begin() ; mapit != e.end(); mapit++ )
   {
-    double p = (double)( *mapit ).second / (double)featcounter;
+    double p = ( double ) ( *mapit ).second / ( double ) featcounter;
     globent += p * log2 ( p );
   }
 
@@ -278,14 +278,14 @@ double SemSegContextTree::getBestSplit ( std::vector<NICE::MultiChannelImageT<do
   {
     return 0.0;
   }
-  
+
   /** vector of all possible features */
   std::vector<Operation*> featsel;
 
   for ( int i = 0; i < featsPerSplit; i++ )
   {
     int x1, x2, y1, y2;
-    int ft = (int)( (double)rand() / (double)RAND_MAX * (double)ftypes );
+    int ft = ( int ) ( ( double ) rand() / ( double ) RAND_MAX * ( double ) ftypes );
 
     int tmpws = windowSize;
 
@@ -303,14 +303,14 @@ double SemSegContextTree::getBestSplit ( std::vector<NICE::MultiChannelImageT<do
       tmpws *= 4;
     }
 
-    x1 = (int)( (double)rand() / (double)RAND_MAX * (double)tmpws ) - tmpws / 2;
-    x2 = (int)( (double)rand() / (double)RAND_MAX * (double)tmpws ) - tmpws / 2;
-    y1 = (int)( (double)rand() / (double)RAND_MAX * (double)tmpws ) - tmpws / 2;
-    y2 = (int)( (double)rand() / (double)RAND_MAX * (double)tmpws ) - tmpws / 2;
+    x1 = ( int ) ( ( double ) rand() / ( double ) RAND_MAX * ( double ) tmpws ) - tmpws / 2;
+    x2 = ( int ) ( ( double ) rand() / ( double ) RAND_MAX * ( double ) tmpws ) - tmpws / 2;
+    y1 = ( int ) ( ( double ) rand() / ( double ) RAND_MAX * ( double ) tmpws ) - tmpws / 2;
+    y2 = ( int ) ( ( double ) rand() / ( double ) RAND_MAX * ( double ) tmpws ) - tmpws / 2;
 
-    int f1 = ( int ) ( (double) rand() / (double) RAND_MAX * (double) channelsPerType[ft].size() );
-    int f2 = ( int ) ( (double) rand() / (double) RAND_MAX * (double) channelsPerType[ft].size() );
-    int o = ( int ) ( (double) rand() / (double) RAND_MAX * (double) ops[ft].size() );
+    int f1 = ( int ) ( ( double ) rand() / ( double ) RAND_MAX * ( double ) channelsPerType[ft].size() );
+    int f2 = ( int ) ( ( double ) rand() / ( double ) RAND_MAX * ( double ) channelsPerType[ft].size() );
+    int o = ( int ) ( ( double ) rand() / ( double ) RAND_MAX * ( double ) ops[ft].size() );
 
     Operation *op = ops[ft][o]->clone();
 
@@ -361,7 +361,7 @@ double SemSegContextTree::getBestSplit ( std::vector<NICE::MultiChannelImageT<do
 
     for ( int r = 0; r < randomTests; r++ )
     {
-      splits.push_back ( ( (double)rand() / (double)RAND_MAX*scale ) + minval );
+      splits.push_back ( ( ( double ) rand() / ( double ) RAND_MAX*scale ) + minval );
     }
 
     for ( int run = 0 ; run < randomTests; run++ )
@@ -396,7 +396,7 @@ double SemSegContextTree::getBestSplit ( std::vector<NICE::MultiChannelImageT<do
 
       for ( mapit = eL.begin() ; mapit != eL.end(); mapit++ )
       {
-        double p = (double)( *mapit ).second / (double)counterL;
+        double p = ( double ) ( *mapit ).second / ( double ) counterL;
         leftent -= p * log2 ( p );
       }
 
@@ -404,13 +404,13 @@ double SemSegContextTree::getBestSplit ( std::vector<NICE::MultiChannelImageT<do
 
       for ( mapit = eR.begin() ; mapit != eR.end(); mapit++ )
       {
-        double p = (double)( *mapit ).second / (double)counterR;
+        double p = ( double ) ( *mapit ).second / ( double ) counterR;
         rightent -= p * log2 ( p );
       }
 
       //cout << "rightent: " << rightent << " leftent: " << leftent << endl;
 
-      double pl = (double)counterL / (double)( counterL + counterR );
+      double pl = ( double ) counterL / ( double ) ( counterL + counterR );
 
       double ig = globent - ( 1.0 - pl ) * rightent - pl * leftent;
 
@@ -448,7 +448,7 @@ double SemSegContextTree::getBestSplit ( std::vector<NICE::MultiChannelImageT<do
   }*/
 
 
-#ifdef DEBUG
+#ifdef DEBUGA
   cout << "globent: " << globent <<  " bestig " << bestig << " splitval: " << splitval << endl;
 
 #endif
@@ -464,7 +464,7 @@ inline double SemSegContextTree::getMeanProb ( const int &x, const int &y, const
     val += forest[tree][currentfeats.get ( x,y,tree ) ].dist[channel];
   }
 
-  return val / (double)nbTrees;
+  return val / ( double ) nbTrees;
 }
 
 void SemSegContextTree::computeIntegralImage ( const NICE::MultiChannelImageT<SparseVectorInt> &infeats, NICE::MultiChannelImageT<SparseVectorInt> &integralImage )
@@ -513,25 +513,27 @@ void SemSegContextTree::computeIntegralImage ( const NICE::MultiChannelImageT<un
   {
     firstiteration = false;
 #pragma omp parallel for
-    for ( int it = 0; it < integralMap.size(); it++ )
+    for ( int it = 0; it < ( int ) integralMap.size(); it++ )
     {
+      //cerr << "there are " << omp_get_num_threads() << " threads. " << endl;
+      //cerr << "current thread: " << omp_get_thread_num() << endl;
       int c1 = integralMap[it].first;
       int c = integralMap[it].second;
 
-      feats( 0, 0, c ) = feats( 0, 0, c1 );
+      feats ( 0, 0, c ) = feats ( 0, 0, c1 );
 
       //first column
       for ( int y = 1; y < ysize; y++ )
       {
-        feats( 0, y, c ) = feats.get ( 0, y, c1 )
-                         + feats.get ( 0, y-1, c );
+        feats ( 0, y, c ) = feats.get ( 0, y, c1 )
+                            + feats.get ( 0, y - 1, c );
       }
 
       //first row
       for ( int x = 1; x < xsize; x++ )
       {
-        feats( x, 0, c ) = feats.get ( x, 0, c1 ) 
-                         + feats.get ( x-1, 0, c );
+        feats ( x, 0, c ) = feats.get ( x, 0, c1 )
+                            + feats.get ( x - 1, 0, c );
       }
 
       //rest
@@ -539,17 +541,17 @@ void SemSegContextTree::computeIntegralImage ( const NICE::MultiChannelImageT<un
       {
         for ( int x = 1; x < xsize; x++ )
         {
-          feats ( x, y, c ) = feats ( x, y, c1 ) 
-                            + feats ( x, y - 1, c ) 
-                            + feats ( x - 1, y, c ) 
-                            - feats ( x - 1, y - 1, c );
+          feats ( x, y, c ) = feats ( x, y, c1 )
+                              + feats ( x, y - 1, c )
+                              + feats ( x - 1, y, c )
+                              - feats ( x - 1, y - 1, c );
         }
       }
     }
   }
 
   int channels = ( int ) forest[0][0].dist.size();
-  
+
 #pragma omp parallel for
   for ( int c = 0; c < channels; c++ )
   {
@@ -585,7 +587,7 @@ void SemSegContextTree::computeIntegralImage ( const NICE::MultiChannelImageT<un
 
 inline double computeWeight ( const double &d, const double &dim )
 {
-  return 1.0 / ( pow ( 2, (double)( dim - d + 1 ) ) );
+  return 1.0 / ( pow ( 2, ( double ) ( dim - d + 1 ) ) );
 }
 
 void SemSegContextTree::train ( const MultiDataset *md )
@@ -764,7 +766,7 @@ void SemSegContextTree::train ( const MultiDataset *md )
 
   channelsPerType = vector<vector<int> > ( amountTypes, vector<int>() );
 
-  for ( int i = 0; i < channelType.size(); i++ )
+  for ( int i = 0; i < ( int ) channelType.size(); i++ )
   {
     channelsPerType[channelType[i]].push_back ( i );
   }
@@ -785,8 +787,8 @@ void SemSegContextTree::train ( const MultiDataset *md )
 
   for ( int iCounter = 0; iCounter < imgcounter; iCounter++ )
   {
-    int xsize = (int)currentfeats[iCounter].width();
-    int ysize = (int)currentfeats[iCounter].height();
+    int xsize = ( int ) currentfeats[iCounter].width();
+    int ysize = ( int ) currentfeats[iCounter].height();
 
     for ( int x = 0; x < xsize; x++ )
     {
@@ -801,13 +803,13 @@ void SemSegContextTree::train ( const MultiDataset *md )
     }
   }
 
-  for ( int i = 0; i < (int)a.size(); i++ )
+  for ( int i = 0; i < ( int ) a.size(); i++ )
   {
-    a[i] /= (double)featcounter;
+    a[i] /= ( double ) featcounter;
   }
 
 #ifdef DEBUG
-  for ( int i = 0; i < (int)a.size(); i++ )
+  for ( int i = 0; i < ( int ) a.size(); i++ )
   {
     cout << "a[" << i << "]: " << a[i] << endl;
   }
@@ -840,7 +842,7 @@ void SemSegContextTree::train ( const MultiDataset *md )
   timer.stop();
   cerr << "preprocessing finished in: " << timer.getLastAbsolute() << " seconds" << endl;
   timer.start();
-  
+
   while ( !allleaf && depth < maxDepth )
   {
     depth++;
@@ -861,23 +863,24 @@ void SemSegContextTree::train ( const MultiDataset *md )
     {
       weight = computeWeight ( 1, maxDepth );
     }
-
-#pragma omp parallel for
+    
+ //   omp_set_dynamic(0);
+//#pragma omp parallel for
     for ( int tree = 0; tree < nbTrees; tree++ )
     {
-      int t = (int)forest[tree].size();
-      int s = startnode[tree];
+      const int t = ( int ) forest[tree].size();
+      const int s = startnode[tree];
       startnode[tree] = t;
-#pragma omp parallel for
+#pragma omp parallel for      
       for ( int i = s; i < t; i++ )
       {
         if ( !forest[tree][i].isleaf && forest[tree][i].left < 0 )
         {
           Operation *splitfeat = NULL;
           double splitval;
-          double bestig = getBestSplit(allfeats, lastfeats, labels, i, splitfeat, splitval, tree);
+          double bestig = getBestSplit ( allfeats, lastfeats, labels, i, splitfeat, splitval, tree );
 
-          for ( int ii = 0; ii < lastfeats.size(); ii++ )
+          for ( int ii = 0; ii < ( int ) lastfeats.size(); ii++ )
           {
             for ( int c = 0; c < lastfeats[ii].channels(); c++ )
             {
@@ -892,9 +895,13 @@ void SemSegContextTree::train ( const MultiDataset *md )
           if ( splitfeat != NULL )
           {
             allleaf = false;
-            int left = forest[tree].size();
-            forest[tree].push_back ( TreeNode() );
-            forest[tree].push_back ( TreeNode() );
+            int left;
+            #pragma omp critical
+            {
+              left = forest[tree].size();
+              forest[tree].push_back ( TreeNode() );
+              forest[tree].push_back ( TreeNode() );
+            }
             int right = left + 1;
             forest[tree][i].left = left;
             forest[tree][i].right = right;
@@ -931,9 +938,8 @@ void SemSegContextTree::train ( const MultiDataset *md )
                     feat.tree = &forest[tree];
                     double val = splitfeat->getVal ( feat, x, y );
 
-                    int subx = x / grid;
-                    int suby = y / grid;
-
+                    //int subx = x / grid;
+                    //int suby = y / grid;
 #pragma omp critical
                     if ( val < splitval )
                     {
@@ -1035,18 +1041,15 @@ void SemSegContextTree::train ( const MultiDataset *md )
         }
       }
     }
-
     //compute integral images
     if ( firstiteration )
     {
       for ( int i = 0; i < imgcounter; i++ )
       {
-        int pos = allfeats.size();
         allfeats[i].addChannel ( ( int ) ( classes + rawChannels ) );
       }
     }
 
-#pragma omp parallel for
     for ( int i = 0; i < imgcounter; i++ )
     {
       computeIntegralImage ( currentfeats[i], allfeats[i], channelType.size() - classes );
@@ -1061,7 +1064,7 @@ void SemSegContextTree::train ( const MultiDataset *md )
     cout << "time for depth " << depth << ": " << timerDepth.getLastAbsolute() << endl;
 #endif
   }
-  
+
   timer.stop();
   cerr << "learning finished in: " << timer.getLastAbsolute() << " seconds" << endl;
   timer.start();
@@ -1102,7 +1105,7 @@ void SemSegContextTree::train ( const MultiDataset *md )
 #ifdef DEBUG
   for ( int tree = 0; tree < nbTrees; tree++ )
   {
-    int t = (int)forest[tree].size();
+    int t = ( int ) forest[tree].size();
 
     for ( int i = 0; i < t; i++ )
     {
@@ -1112,10 +1115,10 @@ void SemSegContextTree::train ( const MultiDataset *md )
       {
         cout <<  ", feat: " << forest[tree][i].feat->writeInfos() << " ";
         opOverview[forest[tree][i].feat->getOps() ]++;
-        contextOverview[forest[tree][i].depth][ (int)forest[tree][i].feat->getContext() ]++;
+        contextOverview[forest[tree][i].depth][ ( int ) forest[tree][i].feat->getContext() ]++;
       }
 
-      for ( int d = 0; d < (int)forest[tree][i].dist.size(); d++ )
+      for ( int d = 0; d < ( int ) forest[tree][i].dist.size(); d++ )
       {
         cout << " " << forest[tree][i].dist[d];
       }
@@ -1183,7 +1186,7 @@ void SemSegContextTree::extractBasicFeatures ( NICE::MultiChannelImageT<double>
       ImageT<double> tmp = feats[c];
       ImageT<double> tmp2 = feats[c+currentsize];
 
-      NICE::FilterT<double,double,double>::gradientStrength( tmp, tmp2 );
+      NICE::FilterT<double, double, double>::gradientStrength ( tmp, tmp2 );
     }
   }
 
@@ -1228,9 +1231,9 @@ void SemSegContextTree::extractBasicFeatures ( NICE::MultiChannelImageT<double>
         ImageT<double> dst = feats[currentChannel];
 
         // copy standard image to double image
-        for ( uint y = 0 ; y < confidenceImage.height(); y++ )
-          for ( uint x = 0 ; x < confidenceImage.width(); x++ )
-            feats ( x, y, currentChannel ) = (double) confidenceImage ( x, y );
+        for ( uint y = 0 ; y < ( uint ) confidenceImage.height(); y++ )
+          for ( uint x = 0 ; x < ( uint ) confidenceImage.width(); x++ )
+            feats ( x, y, currentChannel ) = ( double ) confidenceImage ( x, y );
       }
     }
   }
@@ -1317,9 +1320,8 @@ void SemSegContextTree::semanticseg ( CachedExample *ce, NICE::Image & segresult
 
             double val = forest[tree][t].feat->getVal ( feat, x, y );
 
-            int subx = x / grid;
-            int suby = y / grid;
-
+            //int subx = x / grid;
+            //int suby = y / grid;
             if ( val < forest[tree][t].decision )
             {
               currentfeats.set ( x, y, forest[tree][t].left, tree );
@@ -1473,7 +1475,7 @@ void SemSegContextTree::semanticseg ( CachedExample *ce, NICE::Image & segresult
     }
 #undef VISUALIZE
 #ifdef VISUALIZE
-    for ( int j = 0 ; j < (int)probabilities.numChannels; j++ )
+    for ( int j = 0 ; j < ( int ) probabilities.numChannels; j++ )
     {
       //cout << "class: " << j << endl;//" " << cn.text ( j ) << endl;
 
@@ -1587,7 +1589,7 @@ void SemSegContextTree::semanticseg ( CachedExample *ce, NICE::Image & segresult
 
 void SemSegContextTree::store ( std::ostream & os, int format ) const
 {
-  os.precision (numeric_limits<double>::digits10 + 1);
+  os.precision ( numeric_limits<double>::digits10 + 1 );
   os << nbTrees << endl;
   classnames.store ( os );
 
@@ -1624,14 +1626,14 @@ void SemSegContextTree::store ( std::ostream & os, int format ) const
   }
 
   os << channelType.size() << endl;
-  for ( int i = 0; i < channelType.size(); i++ )
+  for ( int i = 0; i < ( int ) channelType.size(); i++ )
   {
     os << channelType[i] << " ";
   }
   os << endl;
 
   os << integralMap.size() << endl;
-  for ( int i = 0; i < integralMap.size(); i++ )
+  for ( int i = 0; i < ( int ) integralMap.size(); i++ )
   {
     os << integralMap[i].first << " " << integralMap[i].second << endl;
   }

semseg/SemanticSegmentation.h

@@ -8,8 +8,6 @@
 #ifndef SEMANTICSEGMENTATIONINCLUDE
 #define SEMANTICSEGMENTATIONINCLUDE
 
-#include <objrec/nice.h>
-
 #include "vislearning/cbaselib/MultiDataset.h"
 #include "vislearning/cbaselib/LocalizationResult.h"
 #include "vislearning/cbaselib/CachedExample.h"