Update classifier and cluster meta examples

shogun-toolbox · May 31, 2018 · 246819f · 246819f
1 parent 15624ba
commit 246819f
Show file tree

Hide file tree

Showing 8 changed files with 29 additions and 29 deletions.
diff --git a/examples/meta/src/binary/averaged_perceptron.sg b/examples/meta/src/binary/averaged_perceptron.sg
@@ -1,7 +1,7 @@
-CSVFile f_feats_train("../../data/classifier_binary_2d_linear_features_train.dat")
-CSVFile f_feats_test("../../data/classifier_binary_2d_linear_features_test.dat")
-CSVFile f_labels_train("../../data/classifier_binary_2d_linear_labels_train.dat")
-CSVFile f_labels_test("../../data/classifier_binary_2d_linear_labels_test.dat")
+File f_feats_train = csv_file("../../data/classifier_binary_2d_linear_features_train.dat")
+File f_feats_test = csv_file("../../data/classifier_binary_2d_linear_features_test.dat")
+File f_labels_train = csv_file("../../data/classifier_binary_2d_linear_labels_train.dat")
+File f_labels_test = csv_file("../../data/classifier_binary_2d_linear_labels_test.dat")
 
 #![create_features]
 Features features_train = features(f_feats_train)
@@ -11,9 +11,7 @@ Labels labels_test = labels(f_labels_test)
 #![create_features]
 
 #![set_parameters]
-real learn_rate=1.0
-int max_iter=1000
-Machine perceptron = machine("AveragedPerceptron", labels=labels_train, learn_rate=learn_rate, max_iter=max_iter)
+Machine perceptron = machine("AveragedPerceptron", labels=labels_train, learn_rate=1.0, max_iter=1000)
 #![set_parameters]
 
 #![train_and_apply]
@@ -27,6 +25,6 @@ real bias = perceptron.get_real("bias")
 #![extract_weights]
 
 #![evaluate_accuracy]
-AccuracyMeasure eval()
+Evaluation eval = evaluation("AccuracyMeasure")
 real accuracy = eval.evaluate(labels_predict, labels_test)
 #![evaluate_accuracy]
diff --git a/examples/meta/src/binary/kernel_support_vector_machine.sg b/examples/meta/src/binary/kernel_support_vector_machine.sg
@@ -30,7 +30,7 @@ real b = svm.get_real("m_bias")
 #![extract_weights_bias]
 
 #![evaluate_accuracy]
-AccuracyMeasure eval()
+Evaluation eval = evaluation("AccuracyMeasure")
 real accuracy = eval.evaluate(labels_predict, labels_test)
 #![evaluate_accuracy]
 

diff --git a/examples/meta/src/binary/linear_support_vector_machine.sg b/examples/meta/src/binary/linear_support_vector_machine.sg
@@ -1,7 +1,7 @@
-CSVFile f_feats_train("../../data/classifier_binary_2d_linear_features_train.dat")
-CSVFile f_feats_test("../../data/classifier_binary_2d_linear_features_test.dat")
-CSVFile f_labels_train("../../data/classifier_binary_2d_linear_labels_train.dat")
-CSVFile f_labels_test("../../data/classifier_binary_2d_linear_labels_test.dat")
+File f_feats_train = csv_file("../../data/classifier_binary_2d_linear_features_train.dat")
+File f_feats_test = csv_file("../../data/classifier_binary_2d_linear_features_test.dat")
+File f_labels_train = csv_file("../../data/classifier_binary_2d_linear_labels_train.dat")
+File f_labels_test = csv_file("../../data/classifier_binary_2d_linear_labels_test.dat")
 
 #![create_features]
 Features features_train = features(f_feats_train)
@@ -30,7 +30,7 @@ real b = svm.get_real("bias")
 #![extract_weights_bias]
 
 #![evaluate_accuracy]
-AccuracyMeasure eval()
+Evaluation eval = evaluation("AccuracyMeasure")
 real accuracy = eval.evaluate(labels_predict, labels_test)
 #![evaluate_accuracy]
 

diff --git a/examples/meta/src/binary/multiple_kernel_learning.sg b/examples/meta/src/binary/multiple_kernel_learning.sg
@@ -45,7 +45,7 @@ Labels labels_predict = mkl.apply()
 #![mkl_apply]
 
 #![evaluate_accuracy]
-AccuracyMeasure eval()
+Evaluation eval = evaluation("AccuracyMeasure")
 real accuracy = eval.evaluate(labels_predict, labels_test)
 #![evaluate_accuracy]
 

diff --git a/examples/meta/src/binary/perceptron.sg b/examples/meta/src/binary/perceptron.sg
@@ -1,7 +1,7 @@
-CSVFile f_feats_train("../../data/classifier_binary_2d_linear_features_train.dat")
-CSVFile f_feats_test("../../data/classifier_binary_2d_linear_features_test.dat")
-CSVFile f_labels_train("../../data/classifier_binary_2d_linear_labels_train.dat")
-CSVFile f_labels_test("../../data/classifier_binary_2d_linear_labels_test.dat")
+File f_feats_train = csv_file("../../data/classifier_binary_2d_linear_features_train.dat")
+File f_feats_test = csv_file("../../data/classifier_binary_2d_linear_features_test.dat")
+File f_labels_train = csv_file("../../data/classifier_binary_2d_linear_labels_train.dat")
+File f_labels_test = csv_file("../../data/classifier_binary_2d_linear_labels_test.dat")
 
 #![create_features]
 Features features_train = features(f_feats_train)
@@ -11,9 +11,7 @@ Labels labels_test = labels(f_labels_test)
 #![create_features]
 
 #![create_instance]
-real learn_rate=1.0
-int max_iter=1000
-Machine perceptron = machine("Perceptron", labels=labels_train, learn_rate=learn_rate, max_iter=max_iter)
+Machine perceptron = machine("Perceptron", labels=labels_train, learn_rate=1.0, max_iter=1000)
 #![create_instance]
 
 #![train_and_apply]
@@ -27,6 +25,6 @@ real bias = perceptron.get_real("bias")
 #![extract_weights]
 
 #![evaluate_accuracy]
-AccuracyMeasure eval()
+Evaluation eval = evaluation("AccuracyMeasure")
 real accuracy = eval.evaluate(labels_predict, labels_test)
 #![evaluate_accuracy]
diff --git a/examples/meta/src/clustering/gaussian_mixture_models.sg b/examples/meta/src/clustering/gaussian_mixture_models.sg
@@ -1,4 +1,4 @@
-CSVFile f_feats_train("../../data/classifier_4class_2d_linear_features_train.dat")
+File f_feats_train = csv_file("../../data/classifier_4class_2d_linear_features_train.dat")
 
 Math:init_random(1)
 
@@ -12,7 +12,7 @@ GMM gmm(num_components)
 #![create_gmm_instance]
 
 #![train_sample]
-gmm.set_features(features_train)
+gmm.put("features", features_train)
 gmm.train_em()
 RealVector output = gmm.sample()
 #![train_sample]
@@ -25,7 +25,7 @@ gmm.train_smem()
 int component_num = 1
 RealVector nth_mean = gmm.get_nth_mean(component_num)
 RealMatrix nth_cov = gmm.get_nth_cov(component_num)
-RealVector coef = gmm.get_coef()
+RealVector coef = gmm.get_real_vector("m_coefficients")
 #![extract_params]
 
 #![cluster_output]

diff --git a/examples/meta/src/clustering/kmeans.sg b/examples/meta/src/clustering/kmeans.sg
@@ -1,4 +1,4 @@
-CSVFile f_feats_train("../../data/classifier_binary_2d_linear_features_train.dat")
+File f_feats_train = csv_file("../../data/classifier_binary_2d_linear_features_train.dat")
 Math:init_random(1)
 
 #![create_features]
@@ -23,6 +23,5 @@ RealVector r = kmeans.get_radiuses()
 #![extract_centers_and_radius]
 
 #![create_instance_mb]
-KMeansMiniBatch kmeans_mb(2, d)
-kmeans_mb.set_mb_params(4, 1000)
+KMeansMiniBatch kmeans_mb(k=2, distance=d, batch_size=4, mb_iter=1000)
 #![create_instance_mb]
diff --git a/src/shogun/clustering/KMeansMiniBatch.cpp b/src/shogun/clustering/KMeansMiniBatch.cpp
@@ -147,6 +147,11 @@ void CKMeansMiniBatch::init_mb_params()
 {
 	batch_size=-1;
 	minib_iter=-1;
+
+	SG_ADD(
+		&batch_size, "batch_size", "batch size for mini-batch KMeans",
+		MS_NOT_AVAILABLE);
+	SG_ADD(&minib_iter, "mb_iter", "number of iterations", MS_NOT_AVAILABLE);
 }
 
 bool CKMeansMiniBatch::train_machine(CFeatures* data)