Redesign LatentLabels This way LatentLabels not only store actually t…

…he h_i (latent labels) but as well as in a general way store the actual y_i (labels).
shogun-toolbox · Aug 15, 2012 · 5689042 · 5689042
1 parent f7116ba
commit 5689042
Show file tree

Hide file tree

Showing 11 changed files with 126 additions and 38 deletions.
diff --git a/examples/undocumented/libshogun/classifier_latent_svm.cpp b/examples/undocumented/libshogun/classifier_latent_svm.cpp
@@ -102,6 +102,8 @@ static void read_dataset(char* fname, CLatentFeatures*& feats, CLatentLabels*& l
 	labels = new CLatentLabels(num_examples);
 	SG_REF(labels);
 
+	CBinaryLabels* ys = new CBinaryLabels(num_examples);
+
 	feats = new CLatentFeatures(num_examples);
 	SG_REF(feats);
 
@@ -122,7 +124,7 @@ static void read_dataset(char* fname, CLatentFeatures*& feats, CLatentLabels*& l
 		label = (atoi(last_pchar) % 2 == 0) ? 1 : -1;
 		pchar++;
 
-		if (labels->set_label(i, label) == false)
+		if (ys->set_label(i, label) == false)
 			SG_SERROR("Couldn't set label for element %d\n", i);
 
 		last_pchar = pchar;
@@ -166,6 +168,8 @@ static void read_dataset(char* fname, CLatentFeatures*& feats, CLatentLabels*& l
 	}
 	fclose(fd);
 
+	labels->set_labels(ys);
+
 	SG_SDONE();
 }
 

diff --git a/src/shogun/classifier/svm/SVMOcas.cpp b/src/shogun/classifier/svm/SVMOcas.cpp
@@ -58,7 +58,7 @@ bool CSVMOcas::train_machine(CFeatures* data)
 	SG_DEBUG("use_bias = %i\n", get_bias_enabled()) ;
 
 	ASSERT(m_labels);
-  ASSERT(m_labels->get_label_type() == LT_BINARY || m_labels->get_label_type() == LT_LATENT);
+  ASSERT(m_labels->get_label_type() == LT_BINARY);
 	if (data)
 	{
 		if (!data->has_property(FP_DOT))

diff --git a/src/shogun/labels/LatentLabels.cpp b/src/shogun/labels/LatentLabels.cpp
@@ -13,28 +13,41 @@
 using namespace shogun;
 
 CLatentLabels::CLatentLabels()
-	: CBinaryLabels()
+	: CLabels()
 {
 	init();
 }
 
-CLatentLabels::CLatentLabels(int32_t num_labels)
-	: CBinaryLabels(num_labels)
+CLatentLabels::CLatentLabels(int32_t num_samples)
+	: CLabels()
 {
 	init();
-	m_latent_labels = new CDynamicObjectArray(num_labels);
+	m_latent_labels = new CDynamicObjectArray(num_samples);
+	SG_REF(m_latent_labels);
+}
+
+CLatentLabels::CLatentLabels(CLabels* labels)
+	: CLabels()
+{
+	init();
+	m_labels = labels;
+	SG_REF(m_labels);
+	m_latent_labels = new CDynamicObjectArray(m_labels->get_num_labels());
 	SG_REF(m_latent_labels);
 }
 
 CLatentLabels::~CLatentLabels()
 {
 	SG_UNREF(m_latent_labels);
+	SG_UNREF(m_labels);
 }
 
 void CLatentLabels::init()
 {
-	SG_ADD((CSGObject**) &m_latent_labels, "m_labels", "The labels", MS_NOT_AVAILABLE);
+	SG_ADD((CSGObject**) &m_latent_labels, "m_latent_labels", "The latent labels", MS_NOT_AVAILABLE);
+	SG_ADD((CSGObject**) &m_labels, "m_labels", "The labels", MS_NOT_AVAILABLE);
 	m_latent_labels = NULL;
+	m_labels = NULL;
 }
 
 CDynamicObjectArray* CLatentLabels::get_latent_labels() const
@@ -87,3 +100,27 @@ CLatentLabels* CLatentLabels::obtain_from_generic(CLabels* base_labels)
 	return NULL;
 }
 
+int32_t CLatentLabels::get_num_labels()
+{
+	if (!m_latent_labels || !m_labels)
+		return 0;
+	int32_t num_labels = m_latent_labels->get_num_elements();
+
+	ASSERT(num_labels == m_labels->get_num_labels());
+
+	return num_labels;
+}
+
+void CLatentLabels::set_labels(CLabels* labels)
+{
+	SG_UNREF(m_labels);
+	SG_REF(labels);
+	m_labels = labels;
+}
+
+CLabels* CLatentLabels::get_labels() const
+{
+	SG_REF(m_labels);
+	return m_labels;
+}
+
diff --git a/src/shogun/labels/LatentLabels.h b/src/shogun/labels/LatentLabels.h
@@ -11,7 +11,7 @@
 #ifndef __LATENTLABELS_H__
 #define __LATENTLABELS_H__
 
-#include <shogun/labels/BinaryLabels.h>
+#include <shogun/labels/Labels.h>
 #include <shogun/lib/Data.h>
 #include <shogun/lib/DynamicObjectArray.h>
 
@@ -21,7 +21,7 @@ namespace shogun
 	 * As latent labels always depends on the given application, this class
 	 * only defines the API that the user has to implement for latent labels.
 	 */
-	class CLatentLabels : public CBinaryLabels
+	class CLatentLabels : public CLabels
 	{
 		public:
 			/** default ctor */
@@ -31,7 +31,13 @@ namespace shogun
 			 *
 			 * @param num_samples the number of labels
 			 */
-			CLatentLabels(int32_t num_labels);
+			CLatentLabels(int32_t num_samples);
+
+			/** constructor
+			 *
+			 * @param labels the (y_i) labels
+			 */
+			CLatentLabels(CLabels* labels);
 
 			/** destructor */
 			virtual ~CLatentLabels();
@@ -89,9 +95,29 @@ namespace shogun
 			 */
 			virtual const char* get_name() const { return "LatentLabels"; }
 
+			/** get the number of stored labels
+			 *
+			 * @return the number of labels
+			 */
+			virtual int32_t get_num_labels();
+
+			/** set labels
+			 *
+			 * @param labels the labels (y_i)
+			 */
+			void set_labels(CLabels* labels);
+
+			/** get the labels (y_i)
+			 *
+			 * @return the labels (y_i)
+			 */
+			CLabels* get_labels() const;
+
 		protected:
-			/** the vector of labels */
+			/** the of CData, the latent labels (h_i) */
 			CDynamicObjectArray* m_latent_labels;
+			/** the labels (y_i) */
+			CLabels* m_labels;
 
 		private:
 			/** initalize the values to default values */

diff --git a/src/shogun/latent/LatentModel.cpp b/src/shogun/latent/LatentModel.cpp
@@ -9,6 +9,7 @@
  */
 
 #include <shogun/latent/LatentModel.h>
+#include <shogun/labels/BinaryLabels.h>
 
 using namespace shogun;
 
@@ -60,13 +61,15 @@ void CLatentModel::set_features(CLatentFeatures* feats)
 void CLatentModel::argmax_h(const SGVector<float64_t>& w)
 {
 	int32_t num = get_num_vectors();
+	CBinaryLabels* y = CBinaryLabels::obtain_from_generic(m_labels->get_labels());
 	ASSERT(num > 0);
 	ASSERT(num == m_labels->get_num_labels());
+
 
 	// argmax_h only for positive examples
 	for (int32_t i = 0; i < num; ++i)
 	{
-		if (m_labels->get_label(i) == 1)
+		if (y->get_label(i) == 1)
 		{
 			// infer h and set it for the argmax_h <w,psi(x,h)>
 			CData* latent_data = infer_latent_variable(w, i);

diff --git a/src/shogun/latent/LatentSOSVM.cpp b/src/shogun/latent/LatentSOSVM.cpp
@@ -32,10 +32,9 @@ CLatentSOSVM::~CLatentSOSVM()
 	SG_UNREF(m_so_solver);
 }
 
-CLatentLabels* CLatentSOSVM::apply(CFeatures* data)
+CLatentLabels* CLatentSOSVM::apply()
 {
 
-
 	return NULL;
 }
 

diff --git a/src/shogun/latent/LatentSOSVM.h b/src/shogun/latent/LatentSOSVM.h
@@ -41,7 +41,7 @@ namespace shogun
 			 * @param data (test)data to be classified
 			 * @return classified labels
 			 */
-			virtual CLatentLabels* apply(CFeatures* data);
+			virtual CLatentLabels* apply();
 
 			void set_so_solver(CLinearStructuredOutputMachine* so);
 

diff --git a/src/shogun/latent/LatentSVM.cpp b/src/shogun/latent/LatentSVM.cpp
@@ -29,13 +29,42 @@ CLatentSVM::~CLatentSVM()
 {
 }
 
+CLatentLabels* CLatentSVM::apply()
+{
+	if (!m_model)
+		SG_ERROR("LatentModel is not set!\n");
+
+	if (!features)
+		return NULL;
+
+	index_t num_examples = m_model->get_num_vectors();
+	CLatentLabels* hs = new CLatentLabels(num_examples);
+	CBinaryLabels* ys = new CBinaryLabels(num_examples);
+
+	for (index_t i = 0; i < num_examples; ++i)
+	{
+		/* find h for the example */
+		CData* h = m_model->infer_latent_variable(w, i);
+		hs->set_latent_label(i, h);
+		SGVector<float64_t> psi_feat = m_model->get_psi_feature_vector(i);
+
+		/* calculate and set y for the example */
+		float64_t y = w.dot(w.vector, psi_feat.vector, w.vlen);
+		ys->set_label(i, y);
+	}
+
+	hs->set_labels(ys);
+
+	return hs;
+}
+
 float64_t CLatentSVM::do_inner_loop(float64_t cooling_eps)
 {
-	CLatentLabels* labels = m_model->get_labels();
-	CSVMOcas svm(m_C, features, labels);
+	CLabels* ys = m_model->get_labels()->get_labels();
+	CSVMOcas svm(m_C, features, ys);
 	svm.set_epsilon(cooling_eps);
 	svm.train();
-	SG_UNREF(labels);
+	SG_UNREF(ys);
 
 	/* copy the resulting w */
 	SGVector<float64_t> cur_w = svm.get_w();

diff --git a/src/shogun/latent/LatentSVM.h b/src/shogun/latent/LatentSVM.h
@@ -44,6 +44,14 @@ namespace shogun
 
 			virtual ~CLatentSVM();
 
+			/** apply linear machine to all examples
+			 *
+			 * @return resulting labels
+			 */
+			virtual CLatentLabels* apply();
+
+			using CLinearLatentMachine::apply;
+
 			/** Returns the name of the SGSerializable instance.
 			 *
 			 * @return name of the SGSerializable

diff --git a/src/shogun/machine/LinearLatentMachine.cpp b/src/shogun/machine/LinearLatentMachine.cpp
@@ -44,10 +44,6 @@ CLinearLatentMachine::~CLinearLatentMachine()
 
 CLatentLabels* CLinearLatentMachine::apply()
 {
-	if (!features)
-		return NULL;
-
-	return NULL;
 }
 
 CLatentLabels* CLinearLatentMachine::apply(CFeatures* data)
@@ -57,22 +53,8 @@ CLatentLabels* CLinearLatentMachine::apply(CFeatures* data)
 
 	CLatentFeatures* lf = CLatentFeatures::obtain_from_generic(data);
 	m_model->set_features(lf);
-	index_t num_examples = m_model->get_num_vectors();
-	CLatentLabels* labels = new CLatentLabels(num_examples);
-
-	for (index_t i = 0; i < num_examples; ++i)
-	{
-		/* find h for the example */
-		CData* h = m_model->infer_latent_variable(w, i);
-		labels->set_latent_label(i, h);
-		SGVector<float64_t> psi_feat = m_model->get_psi_feature_vector(i);
-
-		/* calculate and set y for the example */
-		float64_t y = w.dot(w.vector, psi_feat.vector, w.vlen);
-		labels->set_label(i, y);
-	}
 
-	return labels;
+	return apply();
 }
 
 void CLinearLatentMachine::set_model(CLatentModel* latent_model)

diff --git a/src/shogun/machine/LinearLatentMachine.h b/src/shogun/machine/LinearLatentMachine.h
@@ -44,7 +44,7 @@ namespace shogun
 			 *
 			 * @return resulting labels
 			 */
-			virtual CLatentLabels* apply();
+			virtual CLatentLabels* apply() = 0;
 
 			/** apply linear machine to data
 			 *