Neue Klasse NominalFeatureVector

cpp/subprojects/common/include/mlrl/common/data/types.hpp

@@ -6,6 +6,7 @@
 #include <cmath>
 #include <limits>
 
+typedef int int32;
 typedef long int int64;
 typedef unsigned char uint8;
 typedef unsigned int uint32;

cpp/subprojects/common/include/mlrl/common/input/feature_vector.hpp

@@ -6,6 +6,16 @@
 #include "mlrl/common/data/vector_sparse_array.hpp"
 #include "mlrl/common/input/missing_feature_vector.hpp"
 
+/**
+ * Defines an interface for all one-dimensional vectors that store the values of training examples for a certain
+ * feature.
+ */
+class IFeatureVector : public IOneDimensionalView {
+    public:
+
+        virtual ~IFeatureVector() override {};
+};
+
 /**
  * An one-dimensional sparse vector that stores the values of training examples for a certain feature, as well as the
  * indices of examples with missing feature values.

cpp/subprojects/common/include/mlrl/common/input/feature_vector_common.hpp

@@ -0,0 +1,59 @@
+/*
+ * @author Michael Rapp ([email protected])
+ */
+#pragma once
+
+#include "mlrl/common/data/vector_dok_binary.hpp"
+#include "mlrl/common/input/feature_vector.hpp"
+
+#include <memory>
+
+/**
+ * An abstract base class for all feature vectors that store the values of training examples for a certain feature. It
+ * allows to keep track of the indices of examples with missing feature values.
+ */
+class AbstractFeatureVector : public IFeatureVector {
+    private:
+
+        BinaryDokVector missingIndices_;
+
+    public:
+
+        virtual ~AbstractFeatureVector() override {};
+
+        /**
+         * An iterator that provides read-only access to the indices of examples with missing feature values.
+         */
+        typedef BinaryDokVector::index_const_iterator missing_index_const_iterator;
+
+        /**
+         * Returns a `missing_index_const_iterator` to the beginning of the indices of examples with missing feature
+         * values.
+         *
+         * @return A `missing_index_const_iterator` to the beginning
+         */
+        missing_index_const_iterator missing_indices_cbegin() const;
+
+        /**
+         * Returns a `missing_index_const_iterator` to the end of the indices of examples with missing feature values.
+         *
+         * @return A `missing_index_const_iterator` to the end
+         */
+        missing_index_const_iterator missing_indices_cend() const;
+
+        /**
+         * Sets whether the example at a specific index is missing a feature value or not.
+         *
+         * @param index     The index of the example
+         * @param missing   True, if the example at the given index is missing a feature value, false otherwise
+         */
+        void setMissing(uint32 index, bool missing);
+
+        /**
+         * Returns whether the example at a specific index is missing a feature value or not.
+         *
+         * @param index The index of the example
+         * @return      True, if the example at the given index is missing a feature value, false otherwise
+         */
+        bool isMissing(uint32 index) const;
+};

cpp/subprojects/common/include/mlrl/common/input/feature_vector_equal.hpp

@@ -0,0 +1,16 @@
+/*
+ * @author Michael Rapp ([email protected])
+ */
+#pragma once
+
+#include "mlrl/common/input/feature_vector_common.hpp"
+
+/**
+ * A feature vector that does not actually store any values. It is used in cases where all training examples have the
+ * same value for a certain feature.
+ */
+class EqualFeatureVector final : public AbstractFeatureVector {
+    public:
+
+        uint32 getNumElements() const override;
+};

cpp/subprojects/common/include/mlrl/common/input/feature_vector_nominal.hpp

@@ -0,0 +1,152 @@
+/*
+ * @author Michael Rapp ([email protected])
+ */
+#pragma once
+
+#include "mlrl/common/input/feature_vector_common.hpp"
+
+/**
+ * A feature vector that stores the indices of the examples that are associated with each value, except for the minority
+ * value, i.e., the least frequent value, of a nominal feature.
+ */
+class NominalFeatureVector final : public AbstractFeatureVector {
+    private:
+
+        int32* values_;
+
+        uint32* indices_;
+
+        uint32* indptr_;
+
+        const uint32 numValues_;
+
+        const int32 minorityValue_;
+
+    public:
+
+        /**
+         * @param numValues     The number of distinct values of the nominal feature, excluding the minority value
+         * @param numExamples   The total number of examples
+         * @param minorityValue The minority value, i.e., the least frequent value, of the nominal feature
+         */
+        NominalFeatureVector(uint32 numValues, uint32 numExamples, int32 minorityValue);
+
+        ~NominalFeatureVector() override;
+
+        /**
+         * An iterator that provides access to the values of the nominal feature and allows to modify them.
+         */
+        typedef int32* value_iterator;
+
+        /**
+         * An iterator that provides read-only access to the values of the nominal feature.
+         */
+        typedef const int32* value_const_iterator;
+
+        /**
+         * An iterator that provides access to the indices of the examples that are associated with each value of the
+         * nominal feature and allows to modify them.
+         */
+        typedef uint32* index_iterator;
+
+        /**
+         * An iterator that provides read-only access to the indices of the examples that are associated with each value
+         * of the nominal feature.
+         */
+        typedef const uint32* index_const_iterator;
+
+        /**
+         * An iterator that provides access to the indices that specify the first element in the array of example
+         * indices that corresponds to each value of the nominal feature.
+         */
+        typedef uint32* indptr_iterator;
+
+        /**
+         * Returns a `value_iterator` to the beginning of the values of the nominal feature.
+         *
+         * @return A `value_iterator` to the beginning
+         */
+        value_iterator values_begin();
+
+        /**
+         * Returns a `value_iterator` to the end of the values of the nominal feature.
+         *
+         * @return A `value_iterator` to the end
+         */
+        value_iterator values_end();
+
+        /**
+         * Returns a `value_const_iterator` to the beginning of the values of the nominal feature.
+         *
+         * @return A `value_const_iterator` to the beginning
+         */
+        value_const_iterator values_cbegin() const;
+
+        /**
+         * Returns a `value_const_iterator` to the end of the value of the nominal feature.
+         *
+         * @return A `value_const_iterator` to the end
+         */
+        value_const_iterator values_cend() const;
+
+        /**
+         * Returns an `index_iterator` to the beginning of the indices of the examples that are associated with a
+         * specific value of the nominal feature.
+         *
+         * @param index The index of the value
+         * @return      An `index_iterator` to the beginning
+         */
+        index_iterator indices_begin(uint32 index);
+
+        /**
+         * Returns an `index_iterator` to the end of the indices of the examples that are associated with a specific
+         * value of the nominal feature.
+         *
+         * @param index The index of the value
+         * @return      An `index_iterator` to the end
+         */
+        index_iterator indices_end(uint32 index);
+
+        /**
+         * Returns an `index_const_iterator` to the beginning of the indices of the examples that are associated with a
+         * specific value of the nominal feature.
+         *
+         * @param index The index of the value
+         * @return      An `index_const_iterator` to the beginning
+         */
+        index_const_iterator indices_cbegin(uint32 index) const;
+
+        /**
+         * Returns an `index_const_iterator` to the end of the indices of the examples that are associated with a
+         * specific value of the nominal feature.
+         *
+         * @param index The index of the value
+         * @return      An `index_const_iterator` to the end
+         */
+        index_const_iterator indices_cend(uint32 index) const;
+
+        /**
+         * Returns an `indptr_iterator` to the beginning of the indices that specify the first element in the array of
+         * example indices that corresponds to each value of the nominal feature.
+         *
+         * @return An `indptr_iterator` to the beginning
+         */
+        indptr_iterator indptr_begin();
+
+        /**
+         * Returns an `indptr_iterator` to the end of the indices that specify the first element in the array of example
+         * indices that corresponds to each value of the nominal feature.
+         *
+         * @return An `indptr_iterator` to the end
+         */
+        indptr_iterator indptr_end();
+
+        /**
+         * Returns the minority value, i.e., the least frequent value, of the nominal feature.
+         *
+         * @return The minority value
+         */
+        int32 getMinorityValue() const;
+
+        uint32 getNumElements() const override;
+};

cpp/subprojects/common/include/mlrl/common/input/feature_vector_numerical.hpp

@@ -0,0 +1,86 @@
+/*
+ * @author Michael Rapp ([email protected])
+ */
+#pragma once
+
+#include "mlrl/common/data/vector_sparse_array.hpp"
+#include "mlrl/common/input/feature_vector_common.hpp"
+
+/**
+ * A feature vector that stores the values of training examples for a certain numerical feature.
+ */
+class NumericalFeatureVector final : public AbstractFeatureVector {
+    private:
+
+        SparseArrayVector<float32> vector_;
+
+        const float32 sparseValue_;
+
+    public:
+
+        /**
+         * @param numElements   The number of elements in the vector
+         * @param sparseValue   The value of sparse elements not explicitly stored in the vector
+         */
+        NumericalFeatureVector(uint32 numElements, float32 sparseValue);
+
+        /**
+         * An iterator that provides access to the feature values in the vector and allows to modify them.
+         */
+        typedef SparseArrayVector<float32>::iterator iterator;
+
+        /**
+         * An iterator that provides read-only access to the feature values in the vector.
+         */
+        typedef SparseArrayVector<float32>::const_iterator const_iterator;
+
+        /**
+         * Returns an `iterator` to the beginning of the vector.
+         *
+         * @return An `iterator` to the beginning
+         */
+        iterator begin();
+
+        /**
+         * Returns an `iterator` to the end of the vector.
+         *
+         * @return An `iterator` to the end
+         */
+        iterator end();
+
+        /**
+         * Returns a `const_iterator` to the beginning of the vector.
+         *
+         * @return A `const_iterator` to the beginning
+         */
+        const_iterator cbegin() const;
+
+        /**
+         * Returns a `const_iterator` to the end of the vector.
+         *
+         * @return A `const_iterator` to the end
+         */
+        const_iterator cend() const;
+
+        /**
+         * Returns the value of sparse elements not explicitly stored in the vector.
+         *
+         * @return The value of sparse elements
+         */
+        float32 getSparseValue() const;
+
+        /**
+         * Sorts the elements in the vector in ascending order based on their values.
+         */
+        void sortByValues();
+
+        /**
+         * Sets the number of elements in the vector.
+         *
+         * @param numElements   The number of elements to be set
+         * @param freeMemory    True, if unused memory should be freed, if possible, false otherwise
+         */
+        void setNumElements(uint32 numElements, bool freeMemory);
+
+        uint32 getNumElements() const override;
+};

cpp/subprojects/common/meson.build

@@ -40,6 +40,10 @@ source_files = [
     'src/mlrl/common/input/feature_type_numerical.cpp',
     'src/mlrl/common/input/feature_type_ordinal.cpp',
     'src/mlrl/common/input/feature_vector.cpp',
+    'src/mlrl/common/input/feature_vector_common.cpp',
+    'src/mlrl/common/input/feature_vector_equal.cpp',
+    'src/mlrl/common/input/feature_vector_nominal.cpp',
+    'src/mlrl/common/input/feature_vector_numerical.cpp',
     'src/mlrl/common/input/label_matrix_c_contiguous.cpp',
     'src/mlrl/common/input/label_matrix_csc.cpp',
     'src/mlrl/common/input/label_matrix_csr.cpp',

cpp/subprojects/common/src/mlrl/common/input/feature_vector_equal.cpp

@@ -0,0 +1,5 @@
+#include "mlrl/common/input/feature_vector_equal.hpp"
+
+uint32 EqualFeatureVector::getNumElements() const {
+    return 0;
+}