Update Discretizer version

* Add better check in testKBins.py

* Add Discretizer base class for Both discretizers

* Refactor order of constructors init

.devcontainer/Dockerfile

@@ -1,18 +1,16 @@
 FROM mcr.microsoft.com/devcontainers/cpp:0-ubuntu-22.04
 
-ARG REINSTALL_CMAKE_VERSION_FROM_SOURCE="3.22.2"
-
-# Optionally install the cmake for vcpkg
-COPY ./reinstall-cmake.sh /tmp/
-
-RUN if [ "${REINSTALL_CMAKE_VERSION_FROM_SOURCE}" != "none" ]; then \
-        chmod +x /tmp/reinstall-cmake.sh && /tmp/reinstall-cmake.sh ${REINSTALL_CMAKE_VERSION_FROM_SOURCE}; \
-    fi \
-    && rm -f /tmp/reinstall-cmake.sh
+RUN apt-get update && export DEBIAN_FRONTEND=noninteractive \
+    && apt-get -y install --no-install-recommends \
+    python3 \
+    python3-pip \
+    lcov \
+    cmake \
+    && apt-get autoremove -y && apt-get clean -y && rm -rf /var/lib/apt/lists/*
 
+RUN pip3 install --no-cache-dir \
+    cpplint \
+    cmake-format\
+    gcovr
 # [Optional] Uncomment this section to install additional vcpkg ports.
 # RUN su vscode -c "${VCPKG_ROOT}/vcpkg install <your-port-name-here>"
-
-# [Optional] Uncomment this section to install additional packages.
-# RUN apt-get update && export DEBIAN_FRONTEND=noninteractive \
-#     && apt-get -y install --no-install-recommends <your-package-list-here>

.github/workflows/build.yml

@@ -13,11 +13,11 @@ jobs:
     env:
       BUILD_WRAPPER_OUT_DIR: build_wrapper_output_directory # Directory where build-wrapper output will be placed
     steps:
-      - uses: actions/checkout@v3.2.0
+      - uses: actions/checkout@v4.1.6
         with:
           fetch-depth: 0 # Shallow clones should be disabled for a better relevancy of analysis
       - name: Install sonar-scanner and build-wrapper
-        uses: SonarSource/sonarcloud-github-c-cpp@v1
+        uses: SonarSource/sonarcloud-github-c-cpp@v2
       - name: Install lcov & gcovr
         run: |
           sudo apt-get -y install lcov
@@ -30,8 +30,7 @@ jobs:
           make
           ctest -C Release --output-on-failure --test-dir tests
           cd ..
-          # gcovr -f CPPFImdlp.cpp -f Metrics.cpp --merge-mode-functions=separate --txt --sonarqube=coverage.xml
-          gcovr -f CPPFImdlp.cpp -f Metrics.cpp  --txt --sonarqube=coverage.xml
+          gcovr -f CPPFImdlp.cpp -f Metrics.cpp  -f BinDisc.cpp --txt --sonarqube=coverage.xml
       - name: Run sonar-scanner
         env:
           GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}

.gitignore

@@ -31,6 +31,8 @@
 *.out
 *.app
 **/build
+build_Debug
+build_Release
 **/lcoverage
 .idea
 cmake-*

.vscode/launch.json

@@ -8,15 +8,10 @@
             "name": "C++ Launch config",
             "type": "cppdbg",
             "request": "launch",
-            "program": "${workspaceFolder}/build/sample/sample",
-            "cwd": "${workspaceFolder}/build/sample",
-            "args": [
-                "-f",
-                "glass"
-            ],
-            "targetArchitecture": "arm64",
+            "program": "${workspaceFolder}/tests/build/BinDisc_unittest",
+            "cwd": "${workspaceFolder}/tests/build",
+            "args": [],
             "launchCompleteCommand": "exec-run",
-            "preLaunchTask": "CMake: build",
             "stopAtEntry": false,
             "linux": {
                 "MIMode": "gdb",

.vscode/settings.json

@@ -5,5 +5,89 @@
     },
     "C_Cpp.default.configurationProvider": "ms-vscode.cmake-tools",
     "cmake.configureOnOpen": true,
-    "sonarlint.pathToCompileCommands": "${workspaceFolder}/build/compile_commands.json"
+    "sonarlint.pathToCompileCommands": "${workspaceFolder}/build/compile_commands.json",
+    "files.associations": {
+        "*.rmd": "markdown",
+        "*.py": "python",
+        "vector": "cpp",
+        "__bit_reference": "cpp",
+        "__bits": "cpp",
+        "__config": "cpp",
+        "__debug": "cpp",
+        "__errc": "cpp",
+        "__hash_table": "cpp",
+        "__locale": "cpp",
+        "__mutex_base": "cpp",
+        "__node_handle": "cpp",
+        "__nullptr": "cpp",
+        "__split_buffer": "cpp",
+        "__string": "cpp",
+        "__threading_support": "cpp",
+        "__tuple": "cpp",
+        "array": "cpp",
+        "atomic": "cpp",
+        "bitset": "cpp",
+        "cctype": "cpp",
+        "chrono": "cpp",
+        "clocale": "cpp",
+        "cmath": "cpp",
+        "compare": "cpp",
+        "complex": "cpp",
+        "concepts": "cpp",
+        "cstdarg": "cpp",
+        "cstddef": "cpp",
+        "cstdint": "cpp",
+        "cstdio": "cpp",
+        "cstdlib": "cpp",
+        "cstring": "cpp",
+        "ctime": "cpp",
+        "cwchar": "cpp",
+        "cwctype": "cpp",
+        "exception": "cpp",
+        "initializer_list": "cpp",
+        "ios": "cpp",
+        "iosfwd": "cpp",
+        "istream": "cpp",
+        "limits": "cpp",
+        "locale": "cpp",
+        "memory": "cpp",
+        "mutex": "cpp",
+        "new": "cpp",
+        "optional": "cpp",
+        "ostream": "cpp",
+        "ratio": "cpp",
+        "sstream": "cpp",
+        "stdexcept": "cpp",
+        "streambuf": "cpp",
+        "string": "cpp",
+        "string_view": "cpp",
+        "system_error": "cpp",
+        "tuple": "cpp",
+        "type_traits": "cpp",
+        "typeinfo": "cpp",
+        "unordered_map": "cpp",
+        "variant": "cpp",
+        "algorithm": "cpp",
+        "iostream": "cpp",
+        "iomanip": "cpp",
+        "numeric": "cpp",
+        "set": "cpp",
+        "__tree": "cpp",
+        "deque": "cpp",
+        "list": "cpp",
+        "map": "cpp",
+        "unordered_set": "cpp",
+        "any": "cpp",
+        "condition_variable": "cpp",
+        "forward_list": "cpp",
+        "fstream": "cpp",
+        "stack": "cpp",
+        "thread": "cpp",
+        "__memory": "cpp",
+        "filesystem": "cpp",
+        "*.toml": "toml",
+        "utility": "cpp",
+        "span": "cpp",
+        "*.tcc": "cpp"
+    }
 }

BinDisc.cpp

@@ -0,0 +1,99 @@
+#include <algorithm>
+#include <limits>
+#include <cmath>
+#include "BinDisc.h"
+#include <iostream>
+#include <string>
+
+namespace mdlp {
+
+    BinDisc::BinDisc(int n_bins, strategy_t strategy) :
+        Discretizer(), n_bins{ n_bins }, strategy{ strategy }
+    {
+        if (n_bins < 3) {
+            throw std::invalid_argument("n_bins must be greater than 2");
+        }
+    }
+    BinDisc::~BinDisc() = default;
+    void BinDisc::fit(samples_t& X)
+    {
+        // y is included for compatibility with the Discretizer interface
+        cutPoints.clear();
+        if (X.empty()) {
+            cutPoints.push_back(std::numeric_limits<precision_t>::max());
+            return;
+        }
+        if (strategy == strategy_t::QUANTILE) {
+            fit_quantile(X);
+        } else if (strategy == strategy_t::UNIFORM) {
+            fit_uniform(X);
+        }
+    }
+    void BinDisc::fit(samples_t& X, labels_t& y)
+    {
+        fit(X);
+    }
+    std::vector<precision_t> linspace(precision_t start, precision_t end, int num)
+    {
+        // Doesn't include end point as it is not needed
+        if (start == end) {
+            return { 0 };
+        }
+        precision_t delta = (end - start) / static_cast<precision_t>(num - 1);
+        std::vector<precision_t> linspc;
+        for (size_t i = 0; i < num - 1; ++i) {
+            precision_t val = start + delta * static_cast<precision_t>(i);
+            linspc.push_back(val);
+        }
+        return linspc;
+    }
+    size_t clip(const size_t n, size_t lower, size_t upper)
+    {
+        return std::max(lower, std::min(n, upper));
+    }
+    std::vector<precision_t> percentile(samples_t& data, std::vector<precision_t>& percentiles)
+    {
+        // Implementation taken from https://dpilger26.github.io/NumCpp/doxygen/html/percentile_8hpp_source.html
+        std::vector<precision_t> results;
+        results.reserve(percentiles.size());
+        for (auto percentile : percentiles) {
+            const size_t i = static_cast<size_t>(std::floor(static_cast<double>(data.size() - 1) * percentile / 100.));
+            const auto indexLower = clip(i, 0, data.size() - 1);
+            const double percentI = static_cast<double>(indexLower) / static_cast<double>(data.size() - 1);
+            const double fraction =
+                (percentile / 100.0 - percentI) /
+                (static_cast<double>(indexLower + 1) / static_cast<double>(data.size() - 1) - percentI);
+            const auto value = data[indexLower] + (data[indexLower + 1] - data[indexLower]) * fraction;
+            if (value != results.back())
+                results.push_back(value);
+        }
+        return results;
+    }
+    void BinDisc::fit_quantile(samples_t& X)
+    {
+        auto quantiles = linspace(0.0, 100.0, n_bins + 1);
+        auto data = X;
+        std::sort(data.begin(), data.end());
+        if (data.front() == data.back() || data.size() == 1) {
+            // if X is constant
+            cutPoints.push_back(std::numeric_limits<precision_t>::max());
+            return;
+        }
+        cutPoints = percentile(data, quantiles);
+        normalizeCutPoints();
+    }
+    void BinDisc::fit_uniform(samples_t& X)
+    {
+
+        auto minmax = std::minmax_element(X.begin(), X.end());
+        cutPoints = linspace(*minmax.first, *minmax.second, n_bins + 1);
+        normalizeCutPoints();
+    }
+    void BinDisc::normalizeCutPoints()
+    {
+        // Add max value to the end
+        cutPoints.push_back(std::numeric_limits<precision_t>::max());
+        // Remove first as it is not needed
+        cutPoints.erase(cutPoints.begin());
+    }
+}

BinDisc.h

@@ -0,0 +1,29 @@
+#ifndef BINDISC_H
+#define BINDISC_H
+
+#include "typesFImdlp.h"
+#include "Discretizer.h"
+#include <string>
+
+namespace mdlp {
+
+    enum class strategy_t {
+        UNIFORM,
+        QUANTILE
+    };
+    class BinDisc : public Discretizer {
+    public:
+        BinDisc(int n_bins = 3, strategy_t strategy = strategy_t::UNIFORM);
+        ~BinDisc();
+        // y is included for compatibility with the Discretizer interface
+        void fit(samples_t& X_, labels_t& y) override;
+        void fit(samples_t& X);
+    private:
+        void fit_uniform(samples_t&);
+        void fit_quantile(samples_t&);
+        void normalizeCutPoints();
+        int n_bins;
+        strategy_t strategy;
+    };
+}
+#endif

CMakeLists.txt

@@ -7,7 +7,7 @@ endif ()
 
 set(CMAKE_CXX_STANDARD 11)
 
-add_library(mdlp CPPFImdlp.cpp Metrics.cpp sample/sample.cpp)
+add_library(mdlp CPPFImdlp.cpp Metrics.cpp)
 add_subdirectory(sample)
 add_subdirectory(tests)
 

CPPFImdlp.cpp

@@ -3,20 +3,17 @@
 #include <set>
 #include <cmath>
 #include "CPPFImdlp.h"
-#include "Metrics.h"
 
 namespace mdlp {
 
-    CPPFImdlp::CPPFImdlp(size_t min_length_, int max_depth_, float proposed): min_length(min_length_),
+    CPPFImdlp::CPPFImdlp(size_t min_length_, int max_depth_, float proposed) :
+        Discretizer(),
+        min_length(min_length_),
         max_depth(max_depth_),
         proposed_cuts(proposed)
     {
     }
 
-    CPPFImdlp::CPPFImdlp() = default;
-
-    CPPFImdlp::~CPPFImdlp() = default;
-
     size_t CPPFImdlp::compute_max_num_cut_points() const
     {
         // Set the actual maximum number of cut points as a number or as a percentage of the number of samples
@@ -37,6 +34,7 @@ namespace mdlp {
         y = y_;
         num_cut_points = compute_max_num_cut_points();
         depth = 0;
+        discretizedData.clear();
         cutPoints.clear();
         if (X.size() != y.size()) {
             throw invalid_argument("X and y must have the same size");
@@ -177,7 +175,7 @@ namespace mdlp {
     indices_t CPPFImdlp::sortIndices(samples_t& X_, labels_t& y_)
     {
         indices_t idx(X_.size());
-        iota(idx.begin(), idx.end(), 0);
+        std::iota(idx.begin(), idx.end(), 0);
         stable_sort(idx.begin(), idx.end(), [&X_, &y_](size_t i1, size_t i2) {
             if (X_[i1] == X_[i2])
                 return y_[i1] < y_[i2];
@@ -208,4 +206,5 @@ namespace mdlp {
         }
         cutPoints.erase(cutPoints.begin() + static_cast<long>(maxEntropyIdx));
     }
+
 }

CPPFImdlp.h

@@ -2,13 +2,20 @@
 #define CPPFIMDLP_H
 
 #include "typesFImdlp.h"
-#include "Metrics.h"
 #include <limits>
 #include <utility>
 #include <string>
+#include "Metrics.h"
+#include "Discretizer.h"
 
 namespace mdlp {
-    class CPPFImdlp {
+    class CPPFImdlp : public Discretizer {
+    public:
+        CPPFImdlp() = default;
+        CPPFImdlp(size_t min_length_, int max_depth_, float proposed);
+        virtual ~CPPFImdlp() = default;
+        void fit(samples_t& X_, labels_t& y_) override;
+        inline int get_depth() const { return depth; };
     protected:
         size_t min_length = 3;
         int depth = 0;
@@ -18,26 +25,14 @@ namespace mdlp {
         samples_t X = samples_t();
         labels_t y = labels_t();
         Metrics metrics = Metrics(y, indices);
-        cutPoints_t cutPoints;
         size_t num_cut_points = numeric_limits<size_t>::max();
-
         static indices_t sortIndices(samples_t&, labels_t&);
-
         void computeCutPoints(size_t, size_t, int);
         void resizeCutPoints();
         bool mdlp(size_t, size_t, size_t);
         size_t getCandidate(size_t, size_t);
         size_t compute_max_num_cut_points() const;
         pair<precision_t, size_t> valueCutPoint(size_t, size_t, size_t);
-
-    public:
-        CPPFImdlp();
-        CPPFImdlp(size_t, int, float);
-        ~CPPFImdlp();
-        void fit(samples_t&, labels_t&);
-        inline cutPoints_t getCutPoints() const { return cutPoints; };
-        inline int get_depth() const { return depth; };
-        static inline string version() { return "1.1.2"; };
     };
 }
 #endif

Discretizer.h

@@ -0,0 +1,31 @@
+#ifndef DISCRETIZER_H
+#define DISCRETIZER_H
+
+#include <string>
+#include <algorithm>
+#include "typesFImdlp.h"
+
+namespace mdlp {
+    class Discretizer {
+    public:
+        Discretizer() = default;
+        virtual ~Discretizer() = default;
+        virtual void fit(samples_t& X_, labels_t& y_) = 0;
+        inline cutPoints_t getCutPoints() const { return cutPoints; };
+        labels_t& transform(const samples_t& data)
+        {
+            discretizedData.clear();
+            discretizedData.reserve(data.size());
+            for (const precision_t& item : data) {
+                auto upper = std::upper_bound(cutPoints.begin(), cutPoints.end(), item);
+                discretizedData.push_back(upper - cutPoints.begin());
+            }
+            return discretizedData;
+        };
+        static inline std::string version() { return "1.2.0"; };
+    protected:
+        labels_t discretizedData = labels_t();
+        cutPoints_t cutPoints;
+    };
+}
+#endif

README.md

@@ -2,7 +2,7 @@
 [![Quality Gate Status](https://sonarcloud.io/api/project_badges/measure?project=rmontanana_mdlp&metric=alert_status)](https://sonarcloud.io/summary/new_code?id=rmontanana_mdlp)
 [![Reliability Rating](https://sonarcloud.io/api/project_badges/measure?project=rmontanana_mdlp&metric=reliability_rating)](https://sonarcloud.io/summary/new_code?id=rmontanana_mdlp)
 
-# mdlp
+# <img src="logo.png" alt="logo" width="50"/> mdlp
 
 Discretization algorithm based on the paper by Fayyad &amp; Irani [Multi-Interval Discretization of Continuous-Valued Attributes for Classification Learning](https://www.ijcai.org/Proceedings/93-2/Papers/022.pdf)
 

sonar-project.properties

@@ -3,7 +3,7 @@ sonar.organization=rmontanana
 
 # This is the name and version displayed in the SonarCloud UI.
 sonar.projectName=mdlp
-sonar.projectVersion=1.0.2
+sonar.projectVersion=1.1.3
 # sonar.test.exclusions=tests/**
 # sonar.tests=tests/
 # sonar.coverage.exclusions=tests/**,sample/**

tests/ArffFiles.cpp

@@ -63,7 +63,7 @@ void ArffFiles::load(const string& fileName, bool classLast)
             type = "";
             while (ss >> type_w)
                 type += type_w + " ";
-            attributes.emplace_back(attribute, trim(type));
+            attributes.emplace_back(trim(attribute), trim(type));
             continue;
         }
         if (line[0] == '@') {
@@ -111,8 +111,8 @@ void ArffFiles::generateDataset(bool classLast)
 string ArffFiles::trim(const string& source)
 {
     string s(source);
-    s.erase(0, s.find_first_not_of(" \n\r\t"));
-    s.erase(s.find_last_not_of(" \n\r\t") + 1);
+    s.erase(0, s.find_first_not_of(" '\n\r\t"));
+    s.erase(s.find_last_not_of(" '\n\r\t") + 1);
     return s;
 }
 

tests/BinDisc_unittest.cpp

@@ -0,0 +1,346 @@
+#include <fstream>
+#include <string>
+#include <iostream>
+#include "gtest/gtest.h"
+#include "ArffFiles.h"
+#include "../BinDisc.h"
+
+namespace mdlp {
+    const float margin = 1e-4;
+    static std::string set_data_path()
+    {
+        std::string path = "../datasets/";
+        std::ifstream file(path + "iris.arff");
+        if (file.is_open()) {
+            file.close();
+            return path;
+        }
+        return "../../tests/datasets/";
+    }
+    const std::string data_path = set_data_path();
+    class TestBinDisc3U : public BinDisc, public testing::Test {
+    public:
+        TestBinDisc3U(int n_bins = 3) : BinDisc(n_bins, strategy_t::UNIFORM) {};
+    };
+    class TestBinDisc3Q : public BinDisc, public testing::Test {
+    public:
+        TestBinDisc3Q(int n_bins = 3) : BinDisc(n_bins, strategy_t::QUANTILE) {};
+    };
+    class TestBinDisc4U : public BinDisc, public testing::Test {
+    public:
+        TestBinDisc4U(int n_bins = 4) : BinDisc(n_bins, strategy_t::UNIFORM) {};
+    };
+    class TestBinDisc4Q : public BinDisc, public testing::Test {
+    public:
+        TestBinDisc4Q(int n_bins = 4) : BinDisc(n_bins, strategy_t::QUANTILE) {};
+    };
+    TEST_F(TestBinDisc3U, Easy3BinsUniform)
+    {
+        samples_t X = { 1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0 };
+        auto y = labels_t();
+        fit(X, y);
+        auto cuts = getCutPoints();
+        ASSERT_EQ(3, cuts.size());
+        EXPECT_NEAR(3.66667, cuts.at(0), margin);
+        EXPECT_NEAR(6.33333, cuts.at(1), margin);
+        EXPECT_EQ(numeric_limits<float>::max(), cuts.at(2));
+        auto labels = transform(X);
+        labels_t expected = { 0, 0, 0, 1, 1, 1, 2, 2, 2 };
+        EXPECT_EQ(expected, labels);
+    }
+    TEST_F(TestBinDisc3Q, Easy3BinsQuantile)
+    {
+        samples_t X = { 1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0 };
+        fit(X);
+        auto cuts = getCutPoints();
+        ASSERT_EQ(3, cuts.size());
+        EXPECT_NEAR(3.666667, cuts[0], margin);
+        EXPECT_NEAR(6.333333, cuts[1], margin);
+        EXPECT_EQ(numeric_limits<float>::max(), cuts[2]);
+        auto labels = transform(X);
+        labels_t expected = { 0, 0, 0, 1, 1, 1, 2, 2, 2 };
+        EXPECT_EQ(expected, labels);
+    }
+    TEST_F(TestBinDisc3U, X10BinsUniform)
+    {
+        samples_t X = { 1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0 };
+        fit(X);
+        auto cuts = getCutPoints();
+        ASSERT_EQ(3, cuts.size());
+        EXPECT_EQ(4.0, cuts[0]);
+        EXPECT_EQ(7.0, cuts[1]);
+        EXPECT_EQ(numeric_limits<float>::max(), cuts[2]);
+        auto labels = transform(X);
+        labels_t expected = { 0, 0, 0, 1, 1, 1, 2, 2, 2, 2 };
+        EXPECT_EQ(expected, labels);
+    }
+    TEST_F(TestBinDisc3Q, X10BinsQuantile)
+    {
+        samples_t X = { 1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0 };
+        fit(X);
+        auto cuts = getCutPoints();
+        ASSERT_EQ(3, cuts.size());
+        EXPECT_EQ(4, cuts[0]);
+        EXPECT_EQ(7, cuts[1]);
+        EXPECT_EQ(numeric_limits<float>::max(), cuts[2]);
+        auto labels = transform(X);
+        labels_t expected = { 0, 0, 0, 1, 1, 1, 2, 2, 2, 2 };
+        EXPECT_EQ(expected, labels);
+    }
+    TEST_F(TestBinDisc3U, X11BinsUniform)
+    {
+        samples_t X = { 1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0, 11.0 };
+        fit(X);
+        auto cuts = getCutPoints();
+        ASSERT_EQ(3, cuts.size());
+        EXPECT_NEAR(4.33333, cuts[0], margin);
+        EXPECT_NEAR(7.66667, cuts[1], margin);
+        EXPECT_EQ(numeric_limits<float>::max(), cuts[2]);
+        auto labels = transform(X);
+        labels_t expected = { 0, 0, 0, 0, 1, 1, 1, 2, 2, 2, 2 };
+        EXPECT_EQ(expected, labels);
+    }
+    TEST_F(TestBinDisc3U, X11BinsQuantile)
+    {
+        samples_t X = { 1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0, 11.0 };
+        fit(X);
+        auto cuts = getCutPoints();
+        ASSERT_EQ(3, cuts.size());
+        EXPECT_NEAR(4.33333, cuts[0], margin);
+        EXPECT_NEAR(7.66667, cuts[1], margin);
+        EXPECT_EQ(numeric_limits<float>::max(), cuts[2]);
+        auto labels = transform(X);
+        labels_t expected = { 0, 0, 0, 0, 1, 1, 1, 2, 2, 2, 2 };
+        EXPECT_EQ(expected, labels);
+    }
+    TEST_F(TestBinDisc3U, ConstantUniform)
+    {
+        samples_t X = { 1.0, 1.0, 1.0, 1.0, 1.0, 1.0 };
+        fit(X);
+        auto cuts = getCutPoints();
+        ASSERT_EQ(1, cuts.size());
+        EXPECT_EQ(numeric_limits<float>::max(), cuts[0]);
+        auto labels = transform(X);
+        labels_t expected = { 0, 0, 0, 0, 0, 0 };
+        EXPECT_EQ(expected, labels);
+    }
+    TEST_F(TestBinDisc3Q, ConstantQuantile)
+    {
+        samples_t X = { 1.0, 1.0, 1.0, 1.0, 1.0, 1.0 };
+        fit(X);
+        auto cuts = getCutPoints();
+        EXPECT_EQ(1, cuts.size());
+        EXPECT_EQ(numeric_limits<float>::max(), cuts[0]);
+        auto labels = transform(X);
+        labels_t expected = { 0, 0, 0, 0, 0, 0 };
+        EXPECT_EQ(expected, labels);
+    }
+    TEST_F(TestBinDisc3U, EmptyUniform)
+    {
+        samples_t X = {};
+        fit(X);
+        auto cuts = getCutPoints();
+        EXPECT_EQ(1, cuts.size());
+        EXPECT_EQ(numeric_limits<float>::max(), cuts[0]);
+    }
+    TEST_F(TestBinDisc3Q, EmptyQuantile)
+    {
+        samples_t X = {};
+        fit(X);
+        auto cuts = getCutPoints();
+        EXPECT_EQ(1, cuts.size());
+        EXPECT_EQ(numeric_limits<float>::max(), cuts[0]);
+    }
+    TEST(TestBinDisc3, ExceptionNumberBins)
+    {
+        EXPECT_THROW(BinDisc(2), std::invalid_argument);
+    }
+    TEST_F(TestBinDisc3U, EasyRepeated)
+    {
+        samples_t X = { 3.0, 1.0, 1.0, 3.0, 1.0, 1.0, 3.0, 1.0, 1.0 };
+        fit(X);
+        auto cuts = getCutPoints();
+        ASSERT_EQ(3, cuts.size());
+        EXPECT_NEAR(1.66667, cuts[0], margin);
+        EXPECT_NEAR(2.33333, cuts[1], margin);
+        EXPECT_EQ(numeric_limits<float>::max(), cuts[2]);
+        auto labels = transform(X);
+        labels_t expected = { 2, 0, 0, 2, 0, 0, 2, 0, 0 };
+        EXPECT_EQ(expected, labels);
+        ASSERT_EQ(3.0, X[0]); // X is not modified
+    }
+    TEST_F(TestBinDisc3Q, EasyRepeated)
+    {
+        samples_t X = { 3.0, 1.0, 1.0, 3.0, 1.0, 1.0, 3.0, 1.0, 1.0 };
+        fit(X);
+        auto cuts = getCutPoints();
+        EXPECT_EQ(2, cuts.size());
+        EXPECT_NEAR(1.66667, cuts[0], margin);
+        EXPECT_EQ(numeric_limits<float>::max(), cuts[1]);
+        auto labels = transform(X);
+        labels_t expected = { 1, 0, 0, 1, 0, 0, 1, 0, 0 };
+        EXPECT_EQ(expected, labels);
+        ASSERT_EQ(3.0, X[0]); // X is not modified
+    }
+    TEST_F(TestBinDisc4U, Easy4BinsUniform)
+    {
+        samples_t X = { 1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0, 11.0, 12.0 };
+        fit(X);
+        auto cuts = getCutPoints();
+        EXPECT_EQ(4, cuts.size());
+        ASSERT_EQ(3.75, cuts[0]);
+        EXPECT_EQ(6.5, cuts[1]);
+        EXPECT_EQ(9.25, cuts[2]);
+        EXPECT_EQ(numeric_limits<float>::max(), cuts[3]);
+        auto labels = transform(X);
+        labels_t expected = { 0, 0, 0, 1, 1, 1, 2, 2, 2, 3, 3, 3 };
+        EXPECT_EQ(expected, labels);
+    }
+    TEST_F(TestBinDisc4Q, Easy4BinsQuantile)
+    {
+        samples_t X = { 1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0, 11.0, 12.0 };
+        fit(X);
+        auto cuts = getCutPoints();
+        EXPECT_EQ(4, cuts.size());
+        ASSERT_EQ(3.75, cuts[0]);
+        EXPECT_EQ(6.5, cuts[1]);
+        EXPECT_EQ(9.25, cuts[2]);
+        EXPECT_EQ(numeric_limits<float>::max(), cuts[3]);
+        auto labels = transform(X);
+        labels_t expected = { 0, 0, 0, 1, 1, 1, 2, 2, 2, 3, 3, 3 };
+        EXPECT_EQ(expected, labels);
+    }
+    TEST_F(TestBinDisc4U, X13BinsUniform)
+    {
+        samples_t X = { 1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0, 11.0, 12.0, 13.0 };
+        fit(X);
+        auto cuts = getCutPoints();
+        EXPECT_EQ(4, cuts.size());
+        EXPECT_EQ(4.0, cuts[0]);
+        EXPECT_EQ(7.0, cuts[1]);
+        EXPECT_EQ(10.0, cuts[2]);
+        EXPECT_EQ(numeric_limits<float>::max(), cuts[3]);
+        auto labels = transform(X);
+        labels_t expected = { 0, 0, 0, 1, 1, 1, 2, 2, 2, 3, 3, 3, 3 };
+        EXPECT_EQ(expected, labels);
+    }
+    TEST_F(TestBinDisc4Q, X13BinsQuantile)
+    {
+        samples_t X = { 1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0, 11.0, 12.0, 13.0 };
+        fit(X);
+        auto cuts = getCutPoints();
+        EXPECT_EQ(4, cuts.size());
+        EXPECT_EQ(4.0, cuts[0]);
+        EXPECT_EQ(7.0, cuts[1]);
+        EXPECT_EQ(10.0, cuts[2]);
+        EXPECT_EQ(numeric_limits<float>::max(), cuts[3]);
+        auto labels = transform(X);
+        labels_t expected = { 0, 0, 0, 1, 1, 1, 2, 2, 2, 3, 3, 3, 3 };
+        EXPECT_EQ(expected, labels);
+    }
+    TEST_F(TestBinDisc4U, X14BinsUniform)
+    {
+        samples_t X = { 1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0, 11.0, 12.0, 13.0, 14.0 };
+        fit(X);
+        auto cuts = getCutPoints();
+        EXPECT_EQ(4, cuts.size());
+        EXPECT_EQ(4.25, cuts[0]);
+        EXPECT_EQ(7.5, cuts[1]);
+        EXPECT_EQ(10.75, cuts[2]);
+        EXPECT_EQ(numeric_limits<float>::max(), cuts[3]);
+        auto labels = transform(X);
+        labels_t expected = { 0, 0, 0, 0, 1, 1, 1, 2, 2, 2, 3, 3, 3, 3 };
+        EXPECT_EQ(expected, labels);
+    }
+    TEST_F(TestBinDisc4Q, X14BinsQuantile)
+    {
+        samples_t X = { 1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0, 11.0, 12.0, 13.0, 14.0 };
+        fit(X);
+        auto cuts = getCutPoints();
+        EXPECT_EQ(4, cuts.size());
+        EXPECT_EQ(4.25, cuts[0]);
+        EXPECT_EQ(7.5, cuts[1]);
+        EXPECT_EQ(10.75, cuts[2]);
+        EXPECT_EQ(numeric_limits<float>::max(), cuts[3]);
+        auto labels = transform(X);
+        labels_t expected = { 0, 0, 0, 0, 1, 1, 1, 2, 2, 2, 3, 3, 3, 3 };
+        EXPECT_EQ(expected, labels);
+    }
+    TEST_F(TestBinDisc4U, X15BinsUniform)
+    {
+        samples_t X = { 15.0, 8.0, 12.0, 14.0, 6.0, 1.0, 13.0, 11.0, 10.0, 9.0, 7.0, 4.0, 3.0, 5.0, 2.0 };
+        fit(X);
+        auto cuts = getCutPoints();
+        EXPECT_EQ(4, cuts.size());
+        EXPECT_EQ(4.5, cuts[0]);
+        EXPECT_EQ(8, cuts[1]);
+        EXPECT_EQ(11.5, cuts[2]);
+        EXPECT_EQ(numeric_limits<float>::max(), cuts[3]);
+        auto labels = transform(X);
+        labels_t expected = { 3, 2, 3, 3, 1, 0, 3, 2, 2, 2, 1, 0, 0, 1, 0 };
+        EXPECT_EQ(expected, labels);
+    }
+    TEST_F(TestBinDisc4Q, X15BinsQuantile)
+    {
+        samples_t X = { 15.0, 13.0, 12.0, 14.0, 6.0, 1.0, 8.0, 11.0, 10.0, 9.0, 7.0, 4.0, 3.0, 5.0, 2.0 };
+        fit(X);
+        auto cuts = getCutPoints();
+        EXPECT_EQ(4, cuts.size());
+        EXPECT_EQ(4.5, cuts[0]);
+        EXPECT_EQ(8, cuts[1]);
+        EXPECT_EQ(11.5, cuts[2]);
+        EXPECT_EQ(numeric_limits<float>::max(), cuts[3]);
+        auto labels = transform(X);
+        labels_t expected = { 3, 3, 3, 3, 1, 0, 2, 2, 2, 2, 1, 0, 0, 1, 0 };
+        EXPECT_EQ(expected, labels);
+    }
+    TEST_F(TestBinDisc4U, RepeatedValuesUniform)
+    {
+        samples_t X = { 0.0, 1.0, 1.0, 1.0, 2.0, 2.0, 3.0, 3.0, 3.0, 4.0 };
+        //               0    1     2   3    4    5    6    7    8    9
+        fit(X);
+        auto cuts = getCutPoints();
+        EXPECT_EQ(4, cuts.size());
+        EXPECT_EQ(1.0, cuts[0]);
+        EXPECT_EQ(2.0, cuts[1]);
+        ASSERT_EQ(3.0, cuts[2]);
+        EXPECT_EQ(numeric_limits<float>::max(), cuts[3]);
+        auto labels = transform(X);
+        labels_t expected = { 0, 1, 1, 1, 2, 2, 3, 3, 3, 3 };
+        EXPECT_EQ(expected, labels);
+    }
+    TEST_F(TestBinDisc4Q, RepeatedValuesQuantile)
+    {
+        samples_t X = { 0.0, 1.0, 1.0, 1.0, 2.0, 2.0, 3.0, 3.0, 3.0, 4.0 };
+        //               0    1     2   3    4    5    6    7    8    9
+        fit(X);
+        auto cuts = getCutPoints();
+        ASSERT_EQ(3, cuts.size());
+        EXPECT_EQ(2.0, cuts[0]);
+        ASSERT_EQ(3.0, cuts[1]);
+        EXPECT_EQ(numeric_limits<float>::max(), cuts[2]);
+        auto labels = transform(X);
+        labels_t expected = { 0, 0, 0, 0, 1, 1, 2, 2, 2, 2 };
+        EXPECT_EQ(expected, labels);
+    }
+    TEST_F(TestBinDisc4U, irisUniform)
+    {
+        ArffFiles file;
+        file.load(data_path + "iris.arff", true);
+        vector<samples_t>& X = file.getX();
+        fit(X[0]);
+        auto Xt = transform(X[0]);
+        labels_t expected = { 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1, 0, 0, 0, 1, 1, 1, 0, 1, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 3, 2, 2, 1, 2, 1, 2, 0, 2, 0, 0, 1, 1, 1, 1, 2, 1, 1, 2, 1, 1, 1, 2, 1, 2, 2, 2, 2, 1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 1, 1, 1, 1, 1, 0, 1, 1, 1, 2, 0, 1, 2, 1, 3, 2, 2, 3, 0, 3, 2, 3, 2, 2, 2, 1, 1, 2, 2, 3, 3, 1, 2, 1, 3, 2, 2, 3, 2, 1, 2, 3, 3, 3, 2, 2, 1, 3, 2, 2, 1, 2, 2, 2, 1, 2, 2, 2, 2, 2, 2, 1 };
+        EXPECT_EQ(expected, Xt);
+    }
+    TEST_F(TestBinDisc4Q, irisQuantile)
+    {
+        ArffFiles file;
+        file.load(data_path + "iris.arff", true);
+        vector<samples_t>& X = file.getX();
+        fit(X[0]);
+        auto Xt = transform(X[0]);
+        labels_t expected = { 1, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1, 0, 0, 0, 2, 1, 1, 1, 1, 1, 1, 1, 0, 1, 0, 0, 0, 1, 1, 0, 0, 1, 1, 1, 0, 0, 1, 0, 0, 1, 0, 0, 0, 0, 1, 0, 1, 0, 1, 0, 3, 3, 3, 1, 3, 1, 2, 0, 3, 1, 0, 2, 2, 2, 1, 3, 1, 2, 2, 1, 2, 2, 2, 2, 3, 3, 3, 3, 2, 1, 1, 1, 2, 2, 1, 2, 3, 2, 1, 1, 1, 2, 2, 0, 1, 1, 1, 2, 1, 1, 2, 2, 3, 2, 3, 3, 0, 3, 3, 3, 3, 3, 3, 1, 2, 3, 3, 3, 3, 2, 3, 1, 3, 2, 3, 3, 2, 2, 3, 3, 3, 3, 3, 2, 2, 3, 2, 3, 2, 3, 3, 3, 2, 3, 3, 3, 2, 3, 2, 2 };
+        EXPECT_EQ(expected, Xt);
+    }
+}

tests/CMakeLists.txt

@@ -1,3 +1,4 @@
+cmake_minimum_required(VERSION 3.20)
 set(CMAKE_CXX_STANDARD 11)
 include(FetchContent)
 
@@ -15,15 +16,28 @@ FetchContent_MakeAvailable(googletest)
 enable_testing()
 
 add_executable(Metrics_unittest ../Metrics.cpp Metrics_unittest.cpp)
-add_executable(FImdlp_unittest ../CPPFImdlp.cpp ArffFiles.cpp ../Metrics.cpp FImdlp_unittest.cpp)
 target_link_libraries(Metrics_unittest GTest::gtest_main)
-target_link_libraries(FImdlp_unittest GTest::gtest_main)
 target_compile_options(Metrics_unittest PRIVATE --coverage)
-target_compile_options(FImdlp_unittest PRIVATE --coverage)
 target_link_options(Metrics_unittest PRIVATE --coverage)
+
+add_executable(FImdlp_unittest ../CPPFImdlp.cpp ArffFiles.cpp ../Metrics.cpp FImdlp_unittest.cpp)
+target_link_libraries(FImdlp_unittest GTest::gtest_main)
+target_compile_options(FImdlp_unittest PRIVATE --coverage)
 target_link_options(FImdlp_unittest PRIVATE --coverage)
 
+add_executable(BinDisc_unittest ../BinDisc.cpp ArffFiles.cpp BinDisc_unittest.cpp)
+target_link_libraries(BinDisc_unittest GTest::gtest_main)
+target_compile_options(BinDisc_unittest PRIVATE --coverage)
+target_link_options(BinDisc_unittest PRIVATE --coverage)
+
+add_executable(Discretizer_unittest ../BinDisc.cpp ../CPPFImdlp.cpp ArffFiles.cpp ../Metrics.cpp Discretizer_unittest.cpp)
+target_link_libraries(Discretizer_unittest GTest::gtest_main)
+target_compile_options(Discretizer_unittest PRIVATE --coverage)
+target_link_options(Discretizer_unittest PRIVATE --coverage)
+
 include(GoogleTest)
+
 gtest_discover_tests(Metrics_unittest)
 gtest_discover_tests(FImdlp_unittest)
-
+gtest_discover_tests(BinDisc_unittest)
+gtest_discover_tests(Discretizer_unittest)

tests/Discretizer_unittest.cpp

@@ -0,0 +1,74 @@
+#include <fstream>
+#include <string>
+#include <iostream>
+#include "gtest/gtest.h"
+#include "ArffFiles.h"
+#include "../Discretizer.h"
+#include "../BinDisc.h"
+#include "../CPPFImdlp.h"
+
+namespace mdlp {
+    const float margin = 1e-4;
+    static std::string set_data_path()
+    {
+        std::string path = "../datasets/";
+        std::ifstream file(path + "iris.arff");
+        if (file.is_open()) {
+            file.close();
+            return path;
+        }
+        return "../../tests/datasets/";
+    }
+    const std::string data_path = set_data_path();
+
+    TEST(Discretizer, BinIrisUniform)
+    {
+        ArffFiles file;
+        Discretizer* disc = new BinDisc(4, strategy_t::UNIFORM);
+        file.load(data_path + "iris.arff", true);
+        vector<samples_t>& X = file.getX();
+        auto y = labels_t();
+        disc->fit(X[0], y);
+        auto Xt = disc->transform(X[0]);
+        labels_t expected = { 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1, 0, 0, 0, 1, 1, 1, 0, 1, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 3, 2, 2, 1, 2, 1, 2, 0, 2, 0, 0, 1, 1, 1, 1, 2, 1, 1, 2, 1, 1, 1, 2, 1, 2, 2, 2, 2, 1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 1, 1, 1, 1, 1, 0, 1, 1, 1, 2, 0, 1, 2, 1, 3, 2, 2, 3, 0, 3, 2, 3, 2, 2, 2, 1, 1, 2, 2, 3, 3, 1, 2, 1, 3, 2, 2, 3, 2, 1, 2, 3, 3, 3, 2, 2, 1, 3, 2, 2, 1, 2, 2, 2, 1, 2, 2, 2, 2, 2, 2, 1 };
+        delete disc;
+        EXPECT_EQ(expected, Xt);
+    }
+    TEST(Discretizer, BinIrisQuantile)
+    {
+        ArffFiles file;
+        Discretizer* disc = new BinDisc(4, strategy_t::QUANTILE);
+        file.load(data_path + "iris.arff", true);
+        vector<samples_t>& X = file.getX();
+        auto y = labels_t();
+        disc->fit(X[0], y);
+        auto Xt = disc->transform(X[0]);
+        labels_t expected = { 1, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1, 0, 0, 0, 2, 1, 1, 1, 1, 1, 1, 1, 0, 1, 0, 0, 0, 1, 1, 0, 0, 1, 1, 1, 0, 0, 1, 0, 0, 1, 0, 0, 0, 0, 1, 0, 1, 0, 1, 0, 3, 3, 3, 1, 3, 1, 2, 0, 3, 1, 0, 2, 2, 2, 1, 3, 1, 2, 2, 1, 2, 2, 2, 2, 3, 3, 3, 3, 2, 1, 1, 1, 2, 2, 1, 2, 3, 2, 1, 1, 1, 2, 2, 0, 1, 1, 1, 2, 1, 1, 2, 2, 3, 2, 3, 3, 0, 3, 3, 3, 3, 3, 3, 1, 2, 3, 3, 3, 3, 2, 3, 1, 3, 2, 3, 3, 2, 2, 3, 3, 3, 3, 3, 2, 2, 3, 2, 3, 2, 3, 3, 3, 2, 3, 3, 3, 2, 3, 2, 2 };
+        delete disc;
+        EXPECT_EQ(expected, Xt);
+    }
+    TEST(Discretizer, FImdlpIris)
+    {
+        labels_t expected = {
+            5, 3, 4, 4, 5, 5, 5, 5, 2, 4, 5, 5, 3, 3, 5, 5, 5, 5, 5, 5, 5, 5,
+            5, 4, 5, 3, 5, 5, 5, 4, 4, 5, 5, 5, 4, 4, 5, 4, 3, 5, 5, 0, 4, 5,
+            5, 3, 5, 4, 5, 4, 4, 4, 4, 0, 1, 1, 4, 0, 2, 0, 0, 3, 0, 2, 2, 4,
+            3, 0, 0, 0, 4, 1, 0, 1, 2, 3, 1, 3, 2, 0, 0, 0, 0, 0, 3, 5, 4, 0,
+            3, 0, 0, 3, 0, 0, 0, 3, 2, 2, 0, 1, 4, 0, 3, 2, 3, 3, 0, 2, 0, 5,
+            4, 0, 3, 0, 1, 4, 3, 5, 0, 0, 4, 1, 1, 0, 4, 4, 1, 3, 1, 3, 1, 5,
+            1, 1, 0, 3, 5, 4, 3, 4, 4, 4, 0, 4, 4, 3, 0, 3, 5, 3
+        };
+        ArffFiles file;
+        Discretizer* disc = new CPPFImdlp();
+        file.load(data_path + "iris.arff", true);
+        vector<samples_t>& X = file.getX();
+        labels_t& y = file.getY();
+        disc->fit(X[1], y);
+        auto computed = disc->transform(X[1]);
+        delete disc;
+        EXPECT_EQ(computed.size(), expected.size());
+        for (unsigned long i = 0; i < computed.size(); i++) {
+            EXPECT_EQ(computed[i], expected[i]);
+        }
+    }
+}

tests/FImdlp_unittest.cpp

@@ -15,11 +15,11 @@ throw; \
 , etype)
 
 namespace mdlp {
-    class TestFImdlp: public CPPFImdlp, public testing::Test {
+    class TestFImdlp : public CPPFImdlp, public testing::Test {
     public:
         precision_t precision = 0.000001f;
 
-        TestFImdlp(): CPPFImdlp() {}
+        TestFImdlp() : CPPFImdlp() {}
 
         string data_path;
 
@@ -329,4 +329,26 @@ namespace mdlp {
         }
 
     }
+    TEST_F(TestFImdlp, TransformTest)
+    {
+        labels_t expected = {
+            5, 3, 4, 4, 5, 5, 5, 5, 2, 4, 5, 5, 3, 3, 5, 5, 5, 5, 5, 5, 5, 5,
+            5, 4, 5, 3, 5, 5, 5, 4, 4, 5, 5, 5, 4, 4, 5, 4, 3, 5, 5, 0, 4, 5,
+            5, 3, 5, 4, 5, 4, 4, 4, 4, 0, 1, 1, 4, 0, 2, 0, 0, 3, 0, 2, 2, 4,
+            3, 0, 0, 0, 4, 1, 0, 1, 2, 3, 1, 3, 2, 0, 0, 0, 0, 0, 3, 5, 4, 0,
+            3, 0, 0, 3, 0, 0, 0, 3, 2, 2, 0, 1, 4, 0, 3, 2, 3, 3, 0, 2, 0, 5,
+            4, 0, 3, 0, 1, 4, 3, 5, 0, 0, 4, 1, 1, 0, 4, 4, 1, 3, 1, 3, 1, 5,
+            1, 1, 0, 3, 5, 4, 3, 4, 4, 4, 0, 4, 4, 3, 0, 3, 5, 3
+        };
+        ArffFiles file;
+        file.load(data_path + "iris.arff", true);
+        vector<samples_t>& X = file.getX();
+        labels_t& y = file.getY();
+        fit(X[1], y);
+        auto computed = transform(X[1]);
+        EXPECT_EQ(computed.size(), expected.size());
+        for (unsigned long i = 0; i < computed.size(); i++) {
+            EXPECT_EQ(computed[i], expected[i]);
+        }
+    }
 }

tests/test

@@ -1,20 +1,18 @@
+#!/bin/bash
 if [ -d build ] ; then
 	rm -fr build
 fi
 if [ -d gcovr-report ] ; then
 	rm -fr gcovr-report
 fi
-cmake -S . -B build -Wno-dev 
+cmake -S . -B build -Wno-dev -DCMAKE_BUILD_TYPE=Debug -DCMAKE_CXX_FLAGS="--coverage" -DCMAKE_C_FLAGS="--coverage"
 cmake --build build
 cd build
 ctest --output-on-failure
 cd ..
-if [ ! -d gcovr-report ] ; then 
-	mkdir gcovr-report
-fi
-rm -fr gcovr-report/* 2>/dev/null
+mkdir gcovr-report
 #lcov --capture --directory ./ --output-file lcoverage/main_coverage.info
 #lcov --remove lcoverage/main_coverage.info 'v1/*' '/Applications/*' '*/tests/*' --output-file lcoverage/main_coverage.info -q
 #lcov --list lcoverage/main_coverage.info
 cd ..
-gcovr  --gcov-filter "CPPFImdlp.cpp" --gcov-filter "Metrics.cpp" --txt --sonarqube=tests/gcovr-report/coverage.xml   
+gcovr  --gcov-filter "CPPFImdlp.cpp" --gcov-filter "Metrics.cpp" --gcov-filter "BinDisc.cpp" --gcov-filter "Discretizer.h" --txt --sonarqube=tests/gcovr-report/coverage.xml --exclude-noncode-lines

tests/testKbins.py

@@ -0,0 +1,412 @@
+from scipy.io.arff import loadarff
+from sklearn.preprocessing import KBinsDiscretizer
+
+
+def test(clf, X, expected, title):
+    X = [[x] for x in X]
+    clf.fit(X)
+    computed = [int(x[0]) for x in clf.transform(X)]
+    print(f"{title}")
+    print(f"{computed=}")
+    print(f"{expected=}")
+    assert computed == expected
+    print("-" * 80)
+
+
+# Test Uniform Strategy
+clf3u = KBinsDiscretizer(
+    n_bins=3, encode="ordinal", strategy="uniform", subsample=200_000
+)
+clf3q = KBinsDiscretizer(
+    n_bins=3, encode="ordinal", strategy="quantile", subsample=200_000
+)
+clf4u = KBinsDiscretizer(
+    n_bins=4, encode="ordinal", strategy="uniform", subsample=200_000
+)
+clf4q = KBinsDiscretizer(
+    n_bins=4, encode="ordinal", strategy="quantile", subsample=200_000
+)
+#
+X = [1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0]
+labels = [0, 0, 0, 1, 1, 1, 2, 2, 2]
+test(clf3u, X, labels, title="Easy3BinsUniform")
+test(clf3q, X, labels, title="Easy3BinsQuantile")
+#
+X = [1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0]
+labels = [0, 0, 0, 1, 1, 1, 2, 2, 2, 2]
+# En C++ se obtiene el mismo resultado en ambos, no como aquí
+labels2 = [0, 0, 0, 1, 1, 1, 1, 2, 2, 2]
+test(clf3u, X, labels, title="X10BinsUniform")
+test(clf3q, X, labels2, title="X10BinsQuantile")
+#
+X = [1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0, 11.0]
+labels = [0, 0, 0, 0, 1, 1, 1, 2, 2, 2, 2]
+# En C++ se obtiene el mismo resultado en ambos, no como aquí
+# labels2 = [0, 0, 0, 1, 1, 1, 1, 2, 2, 2]
+test(clf3u, X, labels, title="X11BinsUniform")
+test(clf3q, X, labels, title="X11BinsQuantile")
+#
+X = [1.0, 1.0, 1.0, 1.0, 1.0, 1.0]
+labels = [0, 0, 0, 0, 0, 0]
+test(clf3u, X, labels, title="ConstantUniform")
+test(clf3q, X, labels, title="ConstantQuantile")
+#
+X = [3.0, 1.0, 1.0, 3.0, 1.0, 1.0, 3.0, 1.0, 1.0]
+labels = [2, 0, 0, 2, 0, 0, 2, 0, 0]
+labels2 = [1, 0, 0, 1, 0, 0, 1, 0, 0]  # igual que en C++
+test(clf3u, X, labels, title="EasyRepeatedUniform")
+test(clf3q, X, labels2, title="EasyRepeatedQuantile")
+#
+X = [1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0, 11.0, 12.0]
+labels = [0, 0, 0, 1, 1, 1, 2, 2, 2, 3, 3, 3]
+test(clf4u, X, labels, title="Easy4BinsUniform")
+test(clf4q, X, labels, title="Easy4BinsQuantile")
+#
+X = [1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0, 11.0, 12.0, 13.0]
+labels = [0, 0, 0, 1, 1, 1, 2, 2, 2, 3, 3, 3, 3]
+test(clf4u, X, labels, title="X13BinsUniform")
+test(clf4q, X, labels, title="X13BinsQuantile")
+#
+X = [1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0, 11.0, 12.0, 13.0, 14.0]
+labels = [0, 0, 0, 0, 1, 1, 1, 2, 2, 2, 3, 3, 3, 3]
+test(clf4u, X, labels, title="X14BinsUniform")
+test(clf4q, X, labels, title="X14BinsQuantile")
+#
+X1 = [15.0, 8.0, 12.0, 14.0, 6.0, 1.0, 13.0, 11.0, 10.0, 9.0, 7.0, 4.0, 3.0, 5.0, 2.0]
+X2 = [15.0, 13.0, 12.0, 14.0, 6.0, 1.0, 8.0, 11.0, 10.0, 9.0, 7.0, 4.0, 3.0, 5.0, 2.0]
+labels1 = [3, 2, 3, 3, 1, 0, 3, 2, 2, 2, 1, 0, 0, 1, 0]
+labels2 = [3, 3, 3, 3, 1, 0, 2, 2, 2, 2, 1, 0, 0, 1, 0]
+test(clf4u, X1, labels1, title="X15BinsUniform")
+test(clf4q, X2, labels2, title="X15BinsQuantile")
+#
+X = [0.0, 1.0, 1.0, 1.0, 2.0, 2.0, 3.0, 3.0, 3.0, 4.0]
+labels = [0, 1, 1, 1, 2, 2, 3, 3, 3, 3]
+test(clf4u, X, labels, title="RepeatedValuesUniform")
+test(clf4q, X, labels, title="RepeatedValuesQuantile")
+
+print(f"Uniform   {clf4u.bin_edges_=}")
+print(f"Quaintile {clf4q.bin_edges_=}")
+print("-" * 80)
+#
+data, meta = loadarff("tests/datasets/iris.arff")
+
+labelsu = [
+    0,
+    0,
+    0,
+    0,
+    0,
+    1,
+    0,
+    0,
+    0,
+    0,
+    1,
+    0,
+    0,
+    0,
+    1,
+    1,
+    1,
+    0,
+    1,
+    0,
+    1,
+    0,
+    0,
+    0,
+    0,
+    0,
+    0,
+    0,
+    0,
+    0,
+    0,
+    1,
+    0,
+    1,
+    0,
+    0,
+    1,
+    0,
+    0,
+    0,
+    0,
+    0,
+    0,
+    0,
+    0,
+    0,
+    0,
+    0,
+    1,
+    0,
+    3,
+    2,
+    2,
+    1,
+    2,
+    1,
+    2,
+    0,
+    2,
+    0,
+    0,
+    1,
+    1,
+    1,
+    1,
+    2,
+    1,
+    1,
+    2,
+    1,
+    1,
+    1,
+    2,
+    1,
+    2,
+    2,
+    2,
+    2,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    2,
+    2,
+    1,
+    1,
+    1,
+    1,
+    1,
+    0,
+    1,
+    1,
+    1,
+    2,
+    0,
+    1,
+    2,
+    1,
+    3,
+    2,
+    2,
+    3,
+    0,
+    3,
+    2,
+    3,
+    2,
+    2,
+    2,
+    1,
+    1,
+    2,
+    2,
+    3,
+    3,
+    1,
+    2,
+    1,
+    3,
+    2,
+    2,
+    3,
+    2,
+    1,
+    2,
+    3,
+    3,
+    3,
+    2,
+    2,
+    1,
+    3,
+    2,
+    2,
+    1,
+    2,
+    2,
+    2,
+    1,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2,
+    1,
+]
+labelsq = [
+    1,
+    0,
+    0,
+    0,
+    0,
+    1,
+    0,
+    0,
+    0,
+    0,
+    1,
+    0,
+    0,
+    0,
+    2,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    1,
+    0,
+    1,
+    0,
+    0,
+    0,
+    1,
+    1,
+    0,
+    0,
+    1,
+    1,
+    1,
+    0,
+    0,
+    1,
+    0,
+    0,
+    1,
+    0,
+    0,
+    0,
+    0,
+    1,
+    0,
+    1,
+    0,
+    1,
+    0,
+    3,
+    3,
+    3,
+    1,
+    3,
+    1,
+    2,
+    0,
+    3,
+    1,
+    0,
+    2,
+    2,
+    2,
+    1,
+    3,
+    1,
+    2,
+    2,
+    1,
+    2,
+    2,
+    2,
+    2,
+    3,
+    3,
+    3,
+    3,
+    2,
+    1,
+    1,
+    1,
+    2,
+    2,
+    1,
+    2,
+    3,
+    2,
+    1,
+    1,
+    1,
+    2,
+    2,
+    0,
+    1,
+    1,
+    1,
+    2,
+    1,
+    1,
+    2,
+    2,
+    3,
+    2,
+    3,
+    3,
+    0,
+    3,
+    3,
+    3,
+    3,
+    3,
+    3,
+    1,
+    2,
+    3,
+    3,
+    3,
+    3,
+    2,
+    3,
+    1,
+    3,
+    2,
+    3,
+    3,
+    2,
+    2,
+    3,
+    3,
+    3,
+    3,
+    3,
+    2,
+    2,
+    3,
+    2,
+    3,
+    2,
+    3,
+    3,
+    3,
+    2,
+    3,
+    3,
+    3,
+    2,
+    3,
+    2,
+    2,
+]
+# test(clf4u, data["sepallength"], labelsu, title="IrisUniform")
+# test(clf4q, data["sepallength"], labelsq, title="IrisQuantile")
+sepallength = [[x] for x in data["sepallength"]]
+clf4u.fit(sepallength)
+clf4q.fit(sepallength)
+computedu = clf4u.transform(sepallength)
+computedq = clf4q.transform(sepallength)
+wrongu = 0
+wrongq = 0
+for i in range(len(labelsu)):
+    if labelsu[i] != computedu[i]:
+        wrongu += 1
+    if labelsq[i] != computedq[i]:
+        wrongq += 1
+print(f"Iris sepallength diff. between BinDisc & sklearn::KBins Uniform  ={wrongu:3d}")
+print(f"Iris sepallength diff. between BinDisc & sklearn::KBins Quantile ={wrongq:3d}")

typesFImdlp.h

@@ -8,11 +8,11 @@
 using namespace std;
 namespace mdlp {
     typedef float precision_t;
-    typedef vector<precision_t> samples_t;
-    typedef vector<int> labels_t;
-    typedef vector<size_t> indices_t;
-    typedef vector<precision_t> cutPoints_t;
-    typedef map<pair<int, int>, precision_t> cacheEnt_t;
-    typedef map<tuple<int, int, int>, precision_t> cacheIg_t;
+    typedef std::vector<precision_t> samples_t;
+    typedef std::vector<int> labels_t;
+    typedef std::vector<size_t> indices_t;
+    typedef std::vector<precision_t> cutPoints_t;
+    typedef std::map<std::pair<int, int>, precision_t> cacheEnt_t;
+    typedef std::map<std::tuple<int, int, int>, precision_t> cacheIg_t;
 }
 #endif