Refactor NearestCentroid class (#5053)

* Add NonParametricMachine class (#5055) * add nonparametric machine * fix notebooks * Refactor NearestCentroid class
shogun-toolbox · Dec 8, 2020 · fdf8555 · fdf8555
1 parent bcb14d9
commit fdf8555
Show file tree

Hide file tree

Showing 5 changed files with 43 additions and 38 deletions.
diff --git a/doc/ipython-notebooks/classification/Classification.ipynb b/doc/ipython-notebooks/classification/Classification.ipynb
@@ -441,7 +441,7 @@
     "distances_linear.init(shogun_feats_linear, shogun_feats_linear)\n",
     "knn_linear = sg.create_machine(\"KNN\", k=number_of_neighbors, distance=distances_linear, \n",
     "                               labels=shogun_labels_linear)\n",
-    "knn_linear.train()\n",
+    "knn_linear.train(shogun_feats_linear)\n",
     "classifiers_linear.append(knn_linear)\n",
     "classifiers_names.append(\"Nearest Neighbors\")\n",
     "fadings.append(False)\n",
@@ -455,7 +455,7 @@
     "distances_non_linear.init(shogun_feats_non_linear, shogun_feats_non_linear)\n",
     "knn_non_linear = sg.create_machine(\"KNN\", k=number_of_neighbors, distance=distances_non_linear, \n",
     "                                   labels=shogun_labels_non_linear)\n",
-    "knn_non_linear.train()\n",
+    "knn_non_linear.train(shogun_feats_non_linear)\n",
     "classifiers_non_linear.append(knn_non_linear)\n",
     "\n",
     "plt.subplot(122)\n",

diff --git a/examples/meta/src/multiclass/lmnn.sg.in b/examples/meta/src/multiclass/lmnn.sg.in
@@ -20,7 +20,7 @@ Machine knn = create_machine("KNN", k=k,distance=lmnn_distance,labels=labels_tra
 #![create_instance]
 
 #![train_and_apply]
-knn.train()
+knn.train(features_train)
 Labels labels_predict = knn.apply(features_test)
 RealVector output = labels_predict.get_real_vector("labels")
 #![train_and_apply]

diff --git a/src/shogun/classifier/NearestCentroid.cpp b/src/shogun/classifier/NearestCentroid.cpp
@@ -17,43 +17,24 @@ namespace shogun{
 
 	NearestCentroid::NearestCentroid() : DistanceMachine()
 	{
-		init();
 	}
 
-	NearestCentroid::NearestCentroid(const std::shared_ptr<Distance>& d, const std::shared_ptr<Labels>& trainlab) : DistanceMachine()
+	NearestCentroid::NearestCentroid(const std::shared_ptr<Distance>& d) : DistanceMachine()
 	{
-		init();
 		ASSERT(d)
-		ASSERT(trainlab)
 		set_distance(d);
-		set_labels(trainlab);
 	}
 
 	NearestCentroid::~NearestCentroid()
 	{
 	}
 
-	void NearestCentroid::init()
-	{
-		m_shrinking=0;
-		m_is_trained=false;
-	}
-
-
 	bool NearestCentroid::train_machine(std::shared_ptr<Features> data)
 	{
-		ASSERT(m_labels)
-		ASSERT(distance)
-		if (data)
-		{
-			if (m_labels->get_num_labels() != data->get_num_vectors())
-				error("Number of training vectors does not match number of labels");
-			distance->init(data, data);
-		}
-		else
-		{
-			data = distance->get_lhs();
-		}
+		require(distance, "Distance not set");
+		require(m_labels->get_num_labels() == data->get_num_vectors(),
+			"Number of training vectors does not match number of labels");
+		distance->init(data, data);
 
 		auto multiclass_labels = m_labels->as<MulticlassLabels>();
 		auto dense_data = data->as<DenseFeatures<float64_t>>();
@@ -83,7 +64,7 @@ namespace shogun{
 		linalg::scale(centroids, centroids, scale);
 
 		auto centroids_feats = std::make_shared<DenseFeatures<float64_t>>(centroids);
-
+		m_centroids = centroids_feats;
 		m_is_trained=true;
 		distance->init(centroids_feats, distance->get_rhs());
 

diff --git a/src/shogun/classifier/NearestCentroid.h b/src/shogun/classifier/NearestCentroid.h
@@ -45,7 +45,7 @@ class NearestCentroid : public DistanceMachine{
 	 * @param distance distance
 	 * @param trainlab labels for training
 	 */
-	NearestCentroid(const std::shared_ptr<Distance>& distance, const std::shared_ptr<Labels>& trainlab);
+	NearestCentroid(const std::shared_ptr<Distance>& distance);
 
 	/** Destructor
 	 */
@@ -92,26 +92,19 @@ class NearestCentroid : public DistanceMachine{
 	 */
 	bool train_machine(std::shared_ptr<Features> data=NULL) override;
 
-	/** Stores feature data of underlying model.
-	 *
-	 * Sets centroids as lhs
-	 */
-
-private:
-	void init();
 
 protected:
 	///	number of classes (i.e. number of values labels can take)
 	int32_t m_num_classes;
 
 	///	Shrinking parameter
-	float64_t m_shrinking;
+	float64_t m_shrinking = 0;
 
 	///	The centroids of the trained features
 	std::shared_ptr<DenseFeatures<float64_t>> m_centroids;
 
 	///	Tells if the classifier has been trained or not
-	bool m_is_trained;
+	bool m_is_trained = false;
 };
 
 }

diff --git a/tests/unit/classifier/NearestCentroid_unittest.cc b/tests/unit/classifier/NearestCentroid_unittest.cc
@@ -0,0 +1,31 @@
+/*
+ * This software is distributed under BSD 3-clause license (see LICENSE file).
+ *
+ * Authors: Yuhui Liu
+ */
+#include <gtest/gtest.h>
+#include <shogun/classifier/NearestCentroid.h>
+#include <shogun/distance/EuclideanDistance.h>
+#include <shogun/labels/MulticlassLabels.h>
+
+using namespace shogun;
+TEST(NearestCentroid, fit_and_predict)
+{
+	SGMatrix<float64_t> X{{-10, -1}, {-2, -1}, {-3, -2},
+	                      {1, 1},    {2, 1},   {3, 2}};
+	SGVector<float64_t> y{0, 0, 0, 1, 1, 1};
+
+	auto train_data = std::make_shared<DenseFeatures<float64_t>>(X);
+	auto train_labels = std::make_shared<MulticlassLabels>(y);
+	auto distance = std::make_shared<EuclideanDistance>();
+
+	SGMatrix<float64_t> t{{3, 2}, {-10, -1}, {-100, 100}};
+	auto test_data = std::make_shared<DenseFeatures<float64_t>>(t);
+	auto clf = std::make_shared<NearestCentroid>(distance);
+	clf->train(train_data, train_labels);
+	auto result_labels = clf->apply(test_data);
+	auto result = result_labels->as<MulticlassLabels>()->get_labels();
+	EXPECT_EQ(result[0], 1);
+	EXPECT_EQ(result[1], 0);
+	EXPECT_EQ(result[2], 0);
+}