Fix ECOC Random coding bug.

shogun-toolbox · May 6, 2012 · 485f94a · 485f94a
1 parent 786359f
commit 485f94a
Show file tree

Hide file tree

Showing 4 changed files with 22 additions and 8 deletions.
diff --git a/examples/undocumented/libshogun/classifier_multiclass_ecoc_random.cpp b/examples/undocumented/libshogun/classifier_multiclass_ecoc_random.cpp
@@ -4,6 +4,7 @@
 #include <shogun/features/StreamingSimpleFeatures.h>
 #include <shogun/features/SimpleFeatures.h>
 #include <shogun/multiclass/ecoc/ECOCStrategy.h>
+#include <shogun/multiclass/ecoc/ECOCRandomDenseEncoder.h>
 #include <shogun/multiclass/ecoc/ECOCRandomSparseEncoder.h>
 #include <shogun/multiclass/ecoc/ECOCHDDecoder.h>
 #include <shogun/machine/LinearMulticlassMachine.h>
@@ -83,7 +84,7 @@ int main(int argc, char** argv)
 
 	// Create a multiclass svm classifier that consists of several of the previous one
 	CLinearMulticlassMachine* mc_svm = new CLinearMulticlassMachine(
-			new CECOCStrategy(new CECOCRandomSparseEncoder(), new CECOCHDDecoder()), (CDotFeatures*) features, svm, labels);
+			new CECOCStrategy(new CECOCRandomDenseEncoder(), new CECOCHDDecoder()), (CDotFeatures*) features, svm, labels);
 	SG_REF(mc_svm);
 
 	// Train the multiclass machine using the data passed in the constructor

diff --git a/src/shogun/lib/SGMatrix.h b/src/shogun/lib/SGMatrix.h
@@ -40,6 +40,17 @@ template<class T> class SGMatrix
 			: matrix(orig.matrix), num_rows(orig.num_rows),
 			num_cols(orig.num_cols), do_free(orig.do_free) { }
 
+        /** operator= */
+        SGMatrix<T> &operator=(const SGMatrix<T> &rhs)
+        {
+            free_matrix();
+            matrix = rhs.matrix;
+            num_rows = rhs.num_rows;
+            num_cols = rhs.num_cols;
+            do_free = rhs.do_free;
+            return *this;
+        }
+
 		/** empty destructor */
 		virtual ~SGMatrix()
 		{

diff --git a/src/shogun/multiclass/ecoc/ECOCRandomSparseEncoder.cpp b/src/shogun/multiclass/ecoc/ECOCRandomSparseEncoder.cpp
@@ -51,6 +51,7 @@ SGMatrix<int32_t> CECOCRandomSparseEncoder::create_codebook(int32_t num_classes)
     if (codelen <= 0)
         codelen = get_default_code_length(num_classes);
 
+
     SGMatrix<int32_t> best_codebook(codelen, num_classes, true);
     int32_t best_dist = 0;
 
@@ -111,16 +112,18 @@ SGMatrix<int32_t> CECOCRandomSparseEncoder::create_codebook(int32_t num_classes)
             if (min_dist > best_dist)
             {
                 best_dist = min_dist;
-                std::copy(best_codebook.matrix, best_codebook.matrix + codelen*num_classes,
-                        codebook.matrix);
+                std::copy(codebook.matrix, codebook.matrix + codelen*num_classes,
+                        best_codebook.matrix);
             }
         }
 
-        n_iter++;
-        if (best_dist > 0 && n_iter >= m_maxiter)
+        if (++n_iter >= m_maxiter)
             break;
     }
 
+    if (best_dist <= 0)
+        SG_ERROR("Failed to generate ECOC codebook within max number of iterations (%d)", m_maxiter);
+
     codebook.destroy_matrix();
     return best_codebook;
 }
diff --git a/src/shogun/multiclass/ecoc/ECOCStrategy.cpp b/src/shogun/multiclass/ecoc/ECOCStrategy.cpp
@@ -44,13 +44,12 @@ void CECOCStrategy::train_start(CLabels *orig_labels, CLabels *train_labels)
 {
     CMulticlassStrategy::train_start(orig_labels, train_labels);
 
-    m_codebook.free_matrix();
     m_codebook = m_encoder->create_codebook(m_num_classes);
 }
 
 bool CECOCStrategy::train_has_more()
 {
-    return m_train_iter < m_codebook.num_cols;
+    return m_train_iter < m_codebook.num_rows;
 }
 
 SGVector<int32_t> CECOCStrategy::train_prepare_next()
@@ -60,7 +59,7 @@ SGVector<int32_t> CECOCStrategy::train_prepare_next()
     for (int32_t i=0; i < m_orig_labels->get_num_labels(); ++i)
     {
         int32_t label = m_orig_labels->get_int_label(i);
-        switch (m_codebook(label, m_train_iter))
+        switch (m_codebook(m_train_iter, label))
         {
         case -1:
             m_train_labels->set_label(i, -1);