Merge pull request #1344 from votjakovr/feature/gp_binary_classification

Add very basic libshogun example of GP binary classification

examples/undocumented/libshogun/Makefile

@@ -10,7 +10,7 @@ LINKFLAGS=
 CXXFLAGS=-O0 -g
 
 LIBRARY_PATH=LD_LIBRARY_PATH
-LINKFLAGS+=$(LIBS) $(LIBS_ADD) -L$(LIB_PATH) 
+LINKFLAGS+=$(LIBS) $(LIBS_ADD) -L$(LIB_PATH)
 CXXFLAGS+=-I$(INC_PATH) $(INCLUDES)
 CC=c++
 
@@ -22,6 +22,7 @@ TARGETS = basic_minimal \
 		  classifier_minimal_svm \
 		  classifier_mklmulticlass \
 		  classifier_conjugateindex \
+		  classifier_gaussian_process_binary_classification \
 		  classifier_gaussiannaivebayes \
 		  classifier_multiclasslibsvm \
 		  classifier_qda \

examples/undocumented/libshogun/classifier_gaussian_process_binary_classification.cpp

@@ -0,0 +1,145 @@
+/*
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 3 of the License, or
+ * (at your option) any later version.
+ *
+ * Written (W) 2013 Roman Votyakov
+ */
+
+#include <shogun/lib/config.h>
+
+// Eigen3 is required for working with this example
+#ifdef HAVE_EIGEN3
+
+#include <shogun/base/init.h>
+#include <shogun/labels/BinaryLabels.h>
+#include <shogun/features/DenseFeatures.h>
+#include <shogun/kernel/GaussianKernel.h>
+#include <shogun/machine/gp/LaplacianInferenceMethod.h>
+#include <shogun/machine/gp/ZeroMean.h>
+#include <shogun/machine/gp/LogitLikelihood.h>
+#include <shogun/machine/gp/ProbitLikelihood.h>
+#include <shogun/classifier/GaussianProcessBinaryClassification.h>
+
+using namespace shogun;
+
+int main(int argc, char** argv)
+{
+	init_shogun_with_defaults();
+
+	// create some toy classification data with n training examples
+	// and m testing features
+	const index_t n=10, m=25;
+
+	SGMatrix<float64_t> X_train(2, n);
+	SGVector<float64_t> y_train(n);
+	SGMatrix<float64_t> X_test(2, m);
+
+	// fill matrix of training features with random nonsense
+	X_train(0, 0)=0.0919736;
+	X_train(0, 1)=-0.3813827;
+	X_train(0, 2)=-1.8011128;
+	X_train(0, 3)=-1.4603061;
+	X_train(0, 4)=-0.1386884;
+	X_train(0, 5)=0.7827657;
+	X_train(0, 6)=-0.1369808;
+	X_train(0, 7)=0.0058596;
+	X_train(0, 8)=0.1059573;
+	X_train(0, 9)=-1.3059609;
+
+	X_train(1, 0)=1.4186892;
+	X_train(1, 1)=0.2271813;
+	X_train(1, 2)=0.3451326;
+	X_train(1, 3)=0.4495962;
+	X_train(1, 4)=1.2066144;
+	X_train(1, 5)=-0.5425118;
+	X_train(1, 6)=1.3479000;
+	X_train(1, 7)=0.7181545;
+	X_train(1, 8)=0.4036014;
+	X_train(1, 9)=0.8928408;
+
+	// fill vector of training labels with -1/+1 random values
+	y_train[0]=1.0;
+	y_train[1]=-1.0;
+	y_train[2]=-1.0;
+	y_train[3]=-1.0;
+	y_train[4]=-1.0;
+	y_train[5]=1.0;
+	y_train[6]=-1.0;
+	y_train[7]=1.0;
+	y_train[8]=1.0;
+	y_train[9]=-1.0;
+
+	// fill matrix of testing features
+	index_t i=0;
+
+	for (index_t x1=-2; x1<=2; x1++)
+		for (index_t x2=-2; x2<=2; x2++)
+		{
+			X_test(0, i)=(float64_t)x1;
+			X_test(1, i)=(float64_t)x2;
+			i++;
+		}
+
+	// convert training and testing data into shogun representation
+	CDenseFeatures<float64_t>* feat_train=new CDenseFeatures<float64_t>(X_train);
+	CBinaryLabels* lab_train=new CBinaryLabels(y_train);
+	CDenseFeatures<float64_t>* feat_test=new CDenseFeatures<float64_t>(X_test);
+
+	// create Gaussian kernel with width = 2.0
+	CGaussianKernel* kernel=new CGaussianKernel(10, 2.0);
+
+	// create zero mean function
+	CZeroMean* mean=new CZeroMean();
+
+	// you can easily switch between probit and logit likelihood models
+    // by uncommenting/commenting the following lines:
+
+	// create probit likelihood model
+	// CProbitLikelihood* lik=new CProbitLikelihood();
+
+	// create logit likelihood model
+	CLogitLikelihood* lik=new CLogitLikelihood();
+
+	// specify Laplace approximation inference method
+	CLaplacianInferenceMethod* inf=new CLaplacianInferenceMethod(kernel,
+		feat_train, mean, lab_train, lik);
+
+	// create and train GP classifier, which uses Laplace approximation
+	CGaussianProcessBinaryClassification* gpc=new CGaussianProcessBinaryClassification(inf);
+	gpc->train();
+
+	// apply binary classification to the test data and get -1/+1
+	// labels of the predictions
+	CBinaryLabels* predictions=gpc->apply_binary(feat_test);
+	predictions->get_labels().display_vector("predictions");
+
+	// get probabilities p(y*=1|x*) for each testing feature x*
+	SGVector<float64_t> p_test=gpc->get_probabilities(feat_test);
+	p_test.display_vector("predictive probability");
+
+	// get predictive mean
+	SGVector<float64_t> mu_test=gpc->get_mean_vector(feat_test);
+	mu_test.display_vector("predictive mean");
+
+	// get predictive variance
+	SGVector<float64_t> s2_test=gpc->get_variance_vector(feat_test);
+	s2_test.display_vector("predictive variance");
+
+	// free up memory
+	SG_UNREF(gpc);
+	SG_UNREF(predictions);
+
+	exit_shogun();
+	return 0;
+}
+
+#else
+
+int main(int argc, char **argv)
+{
+	return 0;
+}
+
+#endif /* HAVE_EIGEN3 */