Cleanup preprocessors

src/shogun/preprocessor/DependenceMaximization.cpp

@@ -60,19 +60,6 @@ CDependenceMaximization::~CDependenceMaximization()
 	SG_UNREF(m_labels_feats);
 }
 
-void CDependenceMaximization::fit(CFeatures* features)
-{
-	REQUIRE(features, "Features are not initialized!\n");
-	REQUIRE(features->get_feature_class()==C_DENSE ||
-			features->get_feature_class()==C_SPARSE,
-			"Only allowed for dense/sparse features! Provided an instance of "
-			"%s which is of class %d!\n",
-			features->get_name(), features->get_feature_class());
-	REQUIRE(features->get_feature_type()==F_DREAL, "Only allowed for "
-			"features of double type! Provided %d!\n",
-			features->get_feature_type());
-}
-
 CFeatures* CDependenceMaximization::create_transformed_copy(CFeatures* features,
 		index_t idx)
 {

src/shogun/preprocessor/KernelPCA.cpp

@@ -89,9 +89,9 @@ void CKernelPCA::fit(CFeatures* features)
 		m_target_dim = n;
 	}
 
-	SGVector<float64_t> bias_tmp = linalg::rowwise_sum(kernel_matrix);
+	auto bias_tmp = linalg::rowwise_sum(kernel_matrix);
 	linalg::scale(bias_tmp, bias_tmp, -1.0 / n);
-	float64_t s = linalg::sum(bias_tmp) / n;
+	auto s = linalg::sum(bias_tmp) / n;
 	linalg::add_scalar(bias_tmp, -s);
 
 	linalg::center_matrix(kernel_matrix);

src/shogun/preprocessor/PruneVarSubMean.cpp

@@ -5,11 +5,12 @@
  *          Sergey Lisitsyn, Bjoern Esser
  */
 
-#include <shogun/preprocessor/PruneVarSubMean.h>
-#include <shogun/preprocessor/DensePreprocessor.h>
 #include <shogun/features/Features.h>
 #include <shogun/io/SGIO.h>
 #include <shogun/mathematics/Math.h>
+#include <shogun/mathematics/linalg/LinalgNamespace.h>
+#include <shogun/preprocessor/DensePreprocessor.h>
+#include <shogun/preprocessor/PruneVarSubMean.h>
 
 using namespace shogun;
 
@@ -35,41 +36,29 @@ void CPruneVarSubMean::fit(CFeatures* features)
 	int32_t num_examples = simple_features->get_num_vectors();
 	int32_t num_features = simple_features->get_num_features();
 
-	m_mean = SGVector<float64_t>();
 	m_idx = SGVector<int32_t>();
 	m_std = SGVector<float64_t>();
 
-	m_mean.resize_vector(num_features);
-	float64_t* var = SG_MALLOC(float64_t, num_features);
-	int32_t i, j;
-
-	memset(var, 0, num_features * sizeof(float64_t));
-	m_mean.zero();
+	SGVector<float64_t> var(num_features);
 
 	auto feature_matrix = simple_features->get_feature_matrix();
 
 	// compute mean
-	for (i = 0; i < num_examples; i++)
-	{
-		for (j = 0; j < num_features; j++)
-			m_mean[j] += feature_matrix.matrix[i * num_features + j];
-	}
-
-	for (j = 0; j < num_features; j++)
-		m_mean[j] /= num_examples;
+	m_mean = linalg::rowwise_sum(feature_matrix);
+	linalg::scale(m_mean, m_mean, 1.0 / num_examples);
 
 	// compute var
-	for (i = 0; i < num_examples; i++)
+	for (auto i : range(num_examples))
 	{
-		for (j = 0; j < num_features; j++)
+		for (auto j : range(num_features))
 			var[j] += CMath::sq(
 			    m_mean[j] - feature_matrix.matrix[i * num_features + j]);
 	}
 
 	int32_t num_ok = 0;
 	int32_t* idx_ok = SG_MALLOC(int32_t, num_features);
 
-	for (j = 0; j < num_features; j++)
+	for (auto j : range(num_features))
 	{
 		var[j] /= num_examples;
 
@@ -86,15 +75,14 @@ void CPruneVarSubMean::fit(CFeatures* features)
 	SGVector<float64_t> new_mean(num_ok);
 	m_std.resize_vector(num_ok);
 
-	for (j = 0; j < num_ok; j++)
+	for (auto j : range(num_ok))
 	{
 		m_idx[j] = idx_ok[j];
 		new_mean[j] = m_mean[idx_ok[j]];
 		m_std[j] = std::sqrt(var[idx_ok[j]]);
 	}
 	m_num_idx = num_ok;
 	SG_FREE(idx_ok);
-	SG_FREE(var);
 	m_mean = new_mean;
 
 	m_initialized = true;

src/shogun/preprocessor/SortUlongString.cpp

@@ -4,10 +4,11 @@
  * Authors: Soeren Sonnenburg
  */
 
-#include <shogun/preprocessor/SortUlongString.h>
+#include <shogun/base/range.h>
 #include <shogun/features/Features.h>
 #include <shogun/features/StringFeatures.h>
 #include <shogun/mathematics/Math.h>
+#include <shogun/preprocessor/SortUlongString.h>
 
 using namespace shogun;
 
@@ -46,15 +47,14 @@ bool CSortUlongString::save(FILE* f)
 /// return pointer to feature_matrix, i.e. f->get_feature_matrix();
 bool CSortUlongString::apply_to_string_features(CFeatures* f)
 {
-	int32_t i;
 	auto sf = f->as<CStringFeatures<uint64_t>>();
-	int32_t num_vec = sf->get_num_vectors();
+	auto num_vec = sf->get_num_vectors();
 
-	for (i=0; i<num_vec; i++)
+	for (auto i : range(num_vec))
 	{
 		int32_t len=0;
 		bool free_vec;
-		uint64_t* vec = sf->get_feature_vector(i, len, free_vec);
+		auto vec = sf->get_feature_vector(i, len, free_vec);
 		ASSERT(!free_vec) // won't work with non-in-memory string features
 
 		SG_DEBUG("sorting string of length %i\n", len)
@@ -69,10 +69,9 @@ bool CSortUlongString::apply_to_string_features(CFeatures* f)
 uint64_t* CSortUlongString::apply_to_string(uint64_t* f, int32_t& len)
 {
 	uint64_t* vec=SG_MALLOC(uint64_t, len);
-	int32_t i=0;
 
-	for (i=0; i<len; i++)
-		vec[i]=f[i];
+	for (auto i : range(len))
+		vec[i] = f[i];
 
 	//CMath::qsort(vec, len);
 	CMath::radix_sort(vec, len);