Changeset 1173 for trunk/yat/classifier

Timestamp:

Feb 27, 2008, 4:19:05 PM (16 years ago)

Author:

Markus Ringnér

Message:

Minor fixes

File:

• trunk/yat/classifier/NCC.h (modified) (13 diffs)

trunk/yat/classifier/NCC.h

@@ -114 +114 @@
     void predict_weighted(const MatrixLookupWeighted&, utility::Matrix&) const;   
 
-    utility::Matrix* centroids_;
+    utility::Matrix centroids_;
     bool centroids_nan_;
     Distance distance_;
@@ -129 +129 @@
   template <typename Distance>
   NCC<Distance>::NCC() 
-    : SupervisedClassifier(), centroids_(0), centroids_nan_(false)
+    : SupervisedClassifier(), centroids_nan_(false)
   {
   }
@@ -135 +135 @@
   template <typename Distance>
   NCC<Distance>::NCC(const Distance& dist) 
-    : SupervisedClassifier(), centroids_(0), centroids_nan_(false), distance_(dist)
+    : SupervisedClassifier(), centroids_nan_(false), distance_(dist)
   {
   }
@@ -143 +143 @@
   NCC<Distance>::~NCC()   
   {
-    if(centroids_)
-      delete centroids_;
   }
 
@@ -151 +149 @@
   const utility::Matrix& NCC<Distance>::centroids(void) const
   {
-    return *centroids_;
+    return centroids_;
   }
   
@@ -167 +165 @@
   void NCC<Distance>::train(const MatrixLookup& data, const Target& target)
   {   
-    if(centroids_) 
-      delete centroids_;
-    centroids_= new utility::Matrix(data.rows(), target.nof_classes());
+    centroids_.resize(data.rows(), target.nof_classes());
     for(size_t i=0; i<data.rows(); i++) {
       std::vector<statistics::Averager> class_averager;
@@ -177 +173 @@
       }
       for(size_t c=0;c<target.nof_classes();c++) {
-        (*centroids_)(i,c) = class_averager[c].mean();
+        centroids_(i,c) = class_averager[c].mean();
       }
     }
@@ -186 +182 @@
   void NCC<Distance>::train(const MatrixLookupWeighted& data, const Target& target)
   {   
-    if(centroids_) 
-      delete centroids_;
-    centroids_= new utility::Matrix(data.rows(), target.nof_classes());
+    centroids_.resize(data.rows(), target.nof_classes());
     for(size_t i=0; i<data.rows(); i++) {
       std::vector<statistics::AveragerWeighted> class_averager;
@@ -198 +192 @@
           centroids_nan_=true;
         }
-        (*centroids_)(i,c) = class_averager[c].mean();
+        centroids_(i,c) = class_averager[c].mean();
       }
     }
@@ -209 +203 @@
   {   
     utility::yat_assert<std::runtime_error>
-      (centroids_,"NCC::predict called for untrained classifier");
-    utility::yat_assert<std::runtime_error>
-      (centroids_->rows()==test.rows(),
+      (centroids_.rows()==test.rows(),
        "NCC::predict test data with incorrect number of rows");
     
-    prediction.resize(centroids_->columns(), test.columns());
+    prediction.resize(centroids_.columns(), test.columns());
 
     // If weighted training data has resulted in NaN in centroids: weighted calculations
@@ -231 +223 @@
   {   
     utility::yat_assert<std::runtime_error>
-      (centroids_,"NCC::predict called for untrained classifier");
-    utility::yat_assert<std::runtime_error>
-      (centroids_->rows()==test.rows(),
+      (centroids_.rows()==test.rows(),
        "NCC::predict test data with incorrect number of rows");
     
-    prediction.resize(centroids_->columns(), test.columns());
+    prediction.resize(centroids_.columns(), test.columns());
     predict_weighted(test,prediction);
   }
@@ -245 +235 @@
                                          utility::Matrix& prediction) const
   {
-    MatrixLookup centroids(*centroids_);
+    MatrixLookup centroids(centroids_);
     for(size_t j=0; j<test.columns();j++)
-      for(size_t k=0; k<centroids_->columns();k++) 
+      for(size_t k=0; k<centroids_.columns();k++) 
         prediction(k,j) = distance_(test.begin_column(j), test.end_column(j), 
                                     centroids.begin_column(k));
@@ -256 +246 @@
                                           utility::Matrix& prediction) const
   {
-    MatrixLookupWeighted weighted_centroids(*centroids_);
+    MatrixLookupWeighted weighted_centroids(centroids_);
     for(size_t j=0; j<test.columns();j++) 
-      for(size_t k=0; k<centroids_->columns();k++)
+      for(size_t k=0; k<centroids_.columns();k++)
         prediction(k,j) = distance_(test.begin_column(j), test.end_column(j), 
                                     weighted_centroids.begin_column(k));