Rename config.h to config_platform.h

Add bin5..bin10 q & u discretizers algos
Fix trouble in computing states
Update mdlp to 2.0.0

.gitmodules

@@ -10,10 +10,12 @@
 [submodule "lib/libxlsxwriter"]
 	path = lib/libxlsxwriter
 	url = https://github.com/jmcnamara/libxlsxwriter.git
-[submodule "lib/mdlp"]
-	path = lib/mdlp
-	url = https://github.com/rmontanana/mdlp
-        update = merge
 [submodule "lib/folding"]
 	path = lib/folding
 	url = https://github.com/rmontanana/folding
+[submodule "lib/Files"]
+	path = lib/Files
+	url = https://github.com/rmontanana/ArffFiles
+[submodule "lib/mdlp"]
+	path = lib/mdlp
+	url = https://github.com/rmontanana/mdlp

CMakeLists.txt

@@ -1,7 +1,7 @@
 cmake_minimum_required(VERSION 3.20)
 
 project(Platform
-  VERSION 1.0.4
+  VERSION 1.1.0
   DESCRIPTION "Platform to run Experiments with classifiers."
   HOMEPAGE_URL "https://github.com/rmontanana/platform"
   LANGUAGES CXX
@@ -21,7 +21,7 @@ set(CMAKE_CXX_EXTENSIONS                  OFF)
 set(CMAKE_EXPORT_COMPILE_COMMANDS          ON)
 set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} ${TORCH_CXX_FLAGS}")
 SET(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -pthread")
-set(CMAKE_CXX_FLAGS_RELEASE "${CMAKE_CXX_FLAGS_RELEASE} -O3")
+set(CMAKE_CXX_FLAGS_RELEASE "${CMAKE_CXX_FLAGS_RELEASE} -Ofast")
 set(CMAKE_CXX_FLAGS_DEBUG " ${CMAKE_CXX_FLAGS} -fprofile-arcs -ftest-coverage -O0 -g")
 
 # Options
@@ -46,7 +46,7 @@ if(Boost_FOUND)
 endif()
 
 # Python
-find_package(Python3 3.11...3.11.9 COMPONENTS Interpreter Development REQUIRED)
+find_package(Python3 3.11 COMPONENTS Interpreter Development REQUIRED)
 message("Python3_LIBRARIES=${Python3_LIBRARIES}")
 
 # CMakes modules
@@ -88,10 +88,9 @@ message(STATUS "Bayesnet_INCLUDE_DIRS=${Bayesnet_INCLUDE_DIRS}")
 ## Configure test data path
 cmake_path(SET TEST_DATA_PATH "${CMAKE_CURRENT_SOURCE_DIR}/tests/data")
 configure_file(src/common/SourceData.h.in "${CMAKE_BINARY_DIR}/configured_files/include/SourceData.h")
-add_subdirectory(lib/Files)
 add_subdirectory(config)
 add_subdirectory(src)
-add_subdirectory(sample)
+# add_subdirectory(sample)
 file(GLOB Platform_SOURCES CONFIGURE_DEPENDS ${Platform_SOURCE_DIR}/src/*.cpp)
 
 # Testing

Makefile

@@ -6,7 +6,6 @@ f_release = build_release
 f_debug = build_debug
 app_targets = b_best b_list b_main b_manage b_grid
 test_targets = unit_tests_platform
-n_procs = -j 16
 
 define ClearTests
 	@for t in $(test_targets); do \
@@ -56,10 +55,10 @@ dependency: ## Create a dependency graph diagram of the project (build/dependenc
 	cd $(f_debug) && cmake .. --graphviz=dependency.dot && dot -Tpng dependency.dot -o dependency.png
 
 buildd: ## Build the debug targets
-	cmake --build $(f_debug) -t $(app_targets) PlatformSample $(n_procs)
+	cmake --build $(f_debug) -t $(app_targets) PlatformSample --parallel
 
 buildr: ## Build the release targets
-	cmake --build $(f_release) -t $(app_targets) $(n_procs)
+	cmake --build $(f_release) -t $(app_targets) --parallel
 
 clean: ## Clean the tests info
 	@echo ">>> Cleaning Debug Platform tests...";
@@ -87,7 +86,7 @@ opt = ""
 test: ## Run tests (opt="-s") to verbose output the tests, (opt="-c='Test Maximum Spanning Tree'") to run only that section
 	@echo ">>> Running Platform tests...";
 	@$(MAKE) clean
-	@cmake --build $(f_debug) -t $(test_targets) $(n_procs)
+	@cmake --build $(f_debug) -t $(test_targets) --parallel
 	@for t in $(test_targets); do \
 		if [ -f $(f_debug)/tests/$$t ]; then \
 			cd $(f_debug)/tests ; \

README.md

@@ -20,11 +20,18 @@ In Linux sometimes the library libstdc++ is mistaken from the miniconda installa
 libstdc++.so.6: version `GLIBCXX_3.4.32' not found (required by b_xxxx)
 ```
 
-The solution is to erase the libstdc++ library from the miniconda installation:
+The solution is to erase the libstdc++ library from the miniconda installation and no further compilation is needed.
 
 ### MPI
 
-In Linux just install openmpi & openmpi-devel packages. Only if cmake can't find openmpi installation (like in Oracle Linux) set the following variable:
+In Linux just install openmpi & openmpi-devel packages.
+
+```bash
+source /etc/profile.d/modules.sh
+module load mpi/openmpi-x86_64
+```
+
+If cmake can't find openmpi installation (like in Oracle Linux) set the following variable:
 
 ```bash
 export MPI_HOME="/usr/lib64/openmpi"

config/CMakeLists.txt

@@ -1,4 +1,4 @@
 configure_file(
   "config.h.in"
-  "${CMAKE_BINARY_DIR}/configured_files/include/config.h" ESCAPE_QUOTES
+  "${CMAKE_BINARY_DIR}/configured_files/include/config_platform.h" ESCAPE_QUOTES
 )

gitmodules

@@ -1,8 +1,3 @@
-[submodule "lib/mdlp"]
-	path = lib/mdlp
-	url = https://github.com/rmontanana/mdlp
-	main = main
-	update = merge
 [submodule "lib/catch2"]
 	path = lib/catch2
         main  = v2.x

lib/Files/ArffFiles.cc

@@ -1,176 +0,0 @@
-#include "ArffFiles.h"
-#include <fstream>
-#include <sstream>
-#include <map>
-#include <cctype> // std::isdigit
-#include <algorithm> // std::all_of
-#include <iostream>
-
-ArffFiles::ArffFiles() = default;
-
-std::vector<std::string> ArffFiles::getLines() const
-{
-    return lines;
-}
-
-unsigned long int ArffFiles::getSize() const
-{
-    return lines.size();
-}
-
-std::vector<std::pair<std::string, std::string>> ArffFiles::getAttributes() const
-{
-    return attributes;
-}
-
-std::string ArffFiles::getClassName() const
-{
-    return className;
-}
-
-std::string ArffFiles::getClassType() const
-{
-    return classType;
-}
-
-std::vector<std::vector<float>>& ArffFiles::getX()
-{
-    return X;
-}
-
-std::vector<int>& ArffFiles::getY()
-{
-    return y;
-}
-
-void ArffFiles::loadCommon(std::string fileName)
-{
-    std::ifstream file(fileName);
-    if (!file.is_open()) {
-        throw std::invalid_argument("Unable to open file");
-    }
-    std::string line;
-    std::string keyword;
-    std::string attribute;
-    std::string type;
-    std::string type_w;
-    while (getline(file, line)) {
-        if (line.empty() || line[0] == '%' || line == "\r" || line == " ") {
-            continue;
-        }
-        if (line.find("@attribute") != std::string::npos || line.find("@ATTRIBUTE") != std::string::npos) {
-            std::stringstream ss(line);
-            ss >> keyword >> attribute;
-            type = "";
-            while (ss >> type_w)
-                type += type_w + " ";
-            attributes.emplace_back(trim(attribute), trim(type));
-            continue;
-        }
-        if (line[0] == '@') {
-            continue;
-        }
-        lines.push_back(line);
-    }
-    file.close();
-    if (attributes.empty())
-        throw std::invalid_argument("No attributes found");
-}
-
-void ArffFiles::load(const std::string& fileName, bool classLast)
-{
-    int labelIndex;
-    loadCommon(fileName);
-    if (classLast) {
-        className = std::get<0>(attributes.back());
-        classType = std::get<1>(attributes.back());
-        attributes.pop_back();
-        labelIndex = static_cast<int>(attributes.size());
-    } else {
-        className = std::get<0>(attributes.front());
-        classType = std::get<1>(attributes.front());
-        attributes.erase(attributes.begin());
-        labelIndex = 0;
-    }
-    generateDataset(labelIndex);
-}
-void ArffFiles::load(const std::string& fileName, const std::string& name)
-{
-    int labelIndex;
-    loadCommon(fileName);
-    bool found = false;
-    for (int i = 0; i < attributes.size(); ++i) {
-        if (attributes[i].first == name) {
-            className = std::get<0>(attributes[i]);
-            classType = std::get<1>(attributes[i]);
-            attributes.erase(attributes.begin() + i);
-            labelIndex = i;
-            found = true;
-            break;
-        }
-    }
-    if (!found) {
-        throw std::invalid_argument("Class name not found");
-    }
-    generateDataset(labelIndex);
-}
-
-void ArffFiles::generateDataset(int labelIndex)
-{
-    X = std::vector<std::vector<float>>(attributes.size(), std::vector<float>(lines.size()));
-    auto yy = std::vector<std::string>(lines.size(), "");
-    auto removeLines = std::vector<int>(); // Lines with missing values
-    for (size_t i = 0; i < lines.size(); i++) {
-        std::stringstream ss(lines[i]);
-        std::string value;
-        int pos = 0;
-        int xIndex = 0;
-        while (getline(ss, value, ',')) {
-            if (pos++ == labelIndex) {
-                yy[i] = value;
-            } else {
-                if (value == "?") {
-                    X[xIndex++][i] = -1;
-                    removeLines.push_back(i);
-                } else
-                    X[xIndex++][i] = stof(value);
-            }
-        }
-    }
-    for (auto i : removeLines) {
-        yy.erase(yy.begin() + i);
-        for (auto& x : X) {
-            x.erase(x.begin() + i);
-        }
-    }
-    y = factorize(yy);
-}
-
-std::string ArffFiles::trim(const std::string& source)
-{
-    std::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);
-    return s;
-}
-
-std::vector<int> ArffFiles::factorize(const std::vector<std::string>& labels_t)
-{
-    std::vector<int> yy;
-    labels.clear();
-    yy.reserve(labels_t.size());
-    std::map<std::string, int> labelMap;
-    int i = 0;
-    for (const std::string& label : labels_t) {
-        if (labelMap.find(label) == labelMap.end()) {
-            labelMap[label] = i++;
-            bool allDigits = std::all_of(label.begin(), label.end(), isdigit);
-            if (allDigits)
-                labels.push_back("Class " + label);
-            else
-                labels.push_back(label);
-        }
-        yy.push_back(labelMap[label]);
-    }
-    return yy;
-}

lib/Files/ArffFiles.h

@@ -1,34 +0,0 @@
-#ifndef ARFFFILES_H
-#define ARFFFILES_H
-
-#include <string>
-#include <vector>
-
-class ArffFiles {
-public:
-    ArffFiles();
-    void load(const std::string&, bool = true);
-    void load(const std::string&, const std::string&);
-    std::vector<std::string> getLines() const;
-    unsigned long int getSize() const;
-    std::string getClassName() const;
-    std::string getClassType() const;
-    std::vector<std::string> getLabels() const { return labels; }
-    static std::string trim(const std::string&);
-    std::vector<std::vector<float>>& getX();
-    std::vector<int>& getY();
-    std::vector<std::pair<std::string, std::string>> getAttributes() const;
-    std::vector<int> factorize(const std::vector<std::string>& labels_t);
-private:
-    std::vector<std::string> lines;
-    std::vector<std::pair<std::string, std::string>> attributes;
-    std::string className;
-    std::string classType;
-    std::vector<std::vector<float>> X;
-    std::vector<int> y;
-    std::vector<std::string> labels;
-    void generateDataset(int);
-    void loadCommon(std::string);
-};
-
-#endif

lib/Files/CMakeLists.txt

@@ -1 +0,0 @@
-add_library(ArffFiles ArffFiles.cc)

sample/CMakeLists.txt

@@ -3,7 +3,7 @@ include_directories(
     ${Platform_SOURCE_DIR}/src/main
     ${Python3_INCLUDE_DIRS}
     ${Platform_SOURCE_DIR}/lib/Files
-    ${Platform_SOURCE_DIR}/lib/mdlp
+    ${Platform_SOURCE_DIR}/lib/mdlp/src
     ${Platform_SOURCE_DIR}/lib/argparse/include
     ${Platform_SOURCE_DIR}/lib/folding
     ${Platform_SOURCE_DIR}/lib/json/include
@@ -12,4 +12,4 @@ include_directories(
     ${Bayesnet_INCLUDE_DIRS}
 )
 add_executable(PlatformSample sample.cpp ${Platform_SOURCE_DIR}/src/main/Models.cpp) 
-target_link_libraries(PlatformSample "${PyClassifiers}" "${BayesNet}" ArffFiles mdlp ${Python3_LIBRARIES} "${TORCH_LIBRARIES}" ${LIBTORCH_PYTHON} Boost::python Boost::numpy)
+target_link_libraries(PlatformSample "${PyClassifiers}" "${BayesNet}" fimdlp ${Python3_LIBRARIES} "${TORCH_LIBRARIES}" ${LIBTORCH_PYTHON} Boost::python Boost::numpy)

sample/sample.cpp

@@ -5,13 +5,13 @@
 #include <torch/torch.h>
 #include <argparse/argparse.hpp>
 #include <nlohmann/json.hpp>
-#include <ArffFiles.h>
-#include <CPPFImdlp.h>
+#include <ArffFiles.hpp>
+#include <fimdlp/CPPFImdlp.h>
 #include <folding.hpp>
 #include <bayesnet/utils/BayesMetrics.h>
 #include "Models.h"
 #include "modelRegister.h"
-#include "config.h"
+#include "config_platform.h"
 
 const std::string PATH = { platform_data_path.begin(), platform_data_path.end() };
 
@@ -79,11 +79,11 @@ int main(int argc, char** argv)
         }
         throw runtime_error("file must be one of {diabetes, ecoli, glass, iris, kdd_JapaneseVowels, letter, liver-disorders, mfeat-factors}");
             }
-    );
+        );
     program.add_argument("-p", "--path")
         .help(" folder where the data files are located, default")
         .default_value(std::string{ PATH }
-    );
+        );
     program.add_argument("-m", "--model")
         .help("Model to use " + platform::Models::instance()->toString())
         .action([](const std::string& value) {
@@ -93,7 +93,7 @@ int main(int argc, char** argv)
         }
         throw runtime_error("Model must be one of " + platform::Models::instance()->toString());
             }
-    );
+        );
     program.add_argument("--discretize").help("Discretize input dataset").default_value(false).implicit_value(true);
     program.add_argument("--dumpcpt").help("Dump CPT Tables").default_value(false).implicit_value(true);
     program.add_argument("--stratified").help("If Stratified KFold is to be done").default_value(false).implicit_value(true);
@@ -112,129 +112,130 @@ int main(int argc, char** argv)
         catch (...) {
             throw runtime_error("Number of folds must be an integer");
         }});
-    program.add_argument("-s", "--seed").help("Random seed").default_value(-1).scan<'i', int>();
-    bool class_last, stratified, tensors, dump_cpt;
-    std::string model_name, file_name, path, complete_file_name;
-    int nFolds, seed;
-    try {
-        program.parse_args(argc, argv);
-        file_name = program.get<std::string>("dataset");
-        path = program.get<std::string>("path");
-        model_name = program.get<std::string>("model");
-        complete_file_name = path + file_name + ".arff";
-        stratified = program.get<bool>("stratified");
-        tensors = program.get<bool>("tensors");
-        nFolds = program.get<int>("folds");
-        seed = program.get<int>("seed");
-        dump_cpt = program.get<bool>("dumpcpt");
-        class_last = datasets[file_name];
-        if (!file_exists(complete_file_name)) {
-            throw runtime_error("Data File " + path + file_name + ".arff" + " does not exist");
+        program.add_argument("-s", "--seed").help("Random seed").default_value(-1).scan<'i', int>();
+        bool class_last, stratified, tensors, dump_cpt;
+        std::string model_name, file_name, path, complete_file_name;
+        int nFolds, seed;
+        try {
+            program.parse_args(argc, argv);
+            file_name = program.get<std::string>("dataset");
+            path = program.get<std::string>("path");
+            model_name = program.get<std::string>("model");
+            complete_file_name = path + file_name + ".arff";
+            stratified = program.get<bool>("stratified");
+            tensors = program.get<bool>("tensors");
+            nFolds = program.get<int>("folds");
+            seed = program.get<int>("seed");
+            dump_cpt = program.get<bool>("dumpcpt");
+            class_last = datasets[file_name];
+            if (!file_exists(complete_file_name)) {
+                throw runtime_error("Data File " + path + file_name + ".arff" + " does not exist");
+            }
+        }
+        catch (const exception& err) {
+            cerr << err.what() << std::endl;
+            cerr << program;
+            exit(1);
         }
-    }
-    catch (const exception& err) {
-        cerr << err.what() << std::endl;
-        cerr << program;
-        exit(1);
-    }
 
-    /*
-    * Begin Processing
-    */
-    auto handler = ArffFiles();
-    handler.load(complete_file_name, class_last);
-    // Get Dataset X, y
-    std::vector<mdlp::samples_t>& X = handler.getX();
-    mdlp::labels_t& y = handler.getY();
-    // Get className & Features
-    auto className = handler.getClassName();
-    std::vector<std::string> features;
-    auto attributes = handler.getAttributes();
-    transform(attributes.begin(), attributes.end(), back_inserter(features),
-        [](const pair<std::string, std::string>& item) { return item.first; });
-    // Discretize Dataset
-    auto [Xd, maxes] = discretize(X, y, features);
-    maxes[className] = *max_element(y.begin(), y.end()) + 1;
-    map<std::string, std::vector<int>> states;
-    for (auto feature : features) {
-        states[feature] = std::vector<int>(maxes[feature]);
-    }
-    states[className] = std::vector<int>(maxes[className]);
-    auto clf = platform::Models::instance()->create(model_name);
-    clf->fit(Xd, y, features, className, states);
-    if (dump_cpt) {
-        std::cout << "--- CPT Tables ---" << std::endl;
-        clf->dump_cpt();
-    }
-    auto lines = clf->show();
-    for (auto line : lines) {
-        std::cout << line << std::endl;
-    }
-    std::cout << "--- Topological Order ---" << std::endl;
-    auto order = clf->topological_order();
-    for (auto name : order) {
-        std::cout << name << ", ";
-    }
-    std::cout << "end." << std::endl;
-    auto score = clf->score(Xd, y);
-    std::cout << "Score: " << score << std::endl;
-    auto graph = clf->graph();
-    auto dot_file = model_name + "_" + file_name;
-    ofstream file(dot_file + ".dot");
-    file << graph;
-    file.close();
-    std::cout << "Graph saved in " << model_name << "_" << file_name << ".dot" << std::endl;
-    std::cout << "dot -Tpng -o " + dot_file + ".png " + dot_file + ".dot " << std::endl;
-    std::string stratified_string = stratified ? " Stratified" : "";
-    std::cout << nFolds << " Folds" << stratified_string << " Cross validation" << std::endl;
-    std::cout << "==========================================" << std::endl;
-    torch::Tensor Xt = torch::zeros({ static_cast<int>(Xd.size()), static_cast<int>(Xd[0].size()) }, torch::kInt32);
-    torch::Tensor yt = torch::tensor(y, torch::kInt32);
-    for (int i = 0; i < features.size(); ++i) {
-        Xt.index_put_({ i, "..." }, torch::tensor(Xd[i], torch::kInt32));
-    }
-    float total_score = 0, total_score_train = 0, score_train, score_test;
-    folding::Fold* fold;
-    double nodes = 0.0;
-    if (stratified)
-        fold = new folding::StratifiedKFold(nFolds, y, seed);
-    else
-        fold = new folding::KFold(nFolds, y.size(), seed);
-    for (auto i = 0; i < nFolds; ++i) {
-        auto [train, test] = fold->getFold(i);
-        std::cout << "Fold: " << i + 1 << std::endl;
-        if (tensors) {
-            auto ttrain = torch::tensor(train, torch::kInt64);
-            auto ttest = torch::tensor(test, torch::kInt64);
-            torch::Tensor Xtraint = torch::index_select(Xt, 1, ttrain);
-            torch::Tensor ytraint = yt.index({ ttrain });
-            torch::Tensor Xtestt = torch::index_select(Xt, 1, ttest);
-            torch::Tensor ytestt = yt.index({ ttest });
-            clf->fit(Xtraint, ytraint, features, className, states);
-            auto temp = clf->predict(Xtraint);
-            score_train = clf->score(Xtraint, ytraint);
-            score_test = clf->score(Xtestt, ytestt);
-        } else {
-            auto [Xtrain, ytrain] = extract_indices(train, Xd, y);
-            auto [Xtest, ytest] = extract_indices(test, Xd, y);
-            clf->fit(Xtrain, ytrain, features, className, states);
-            std::cout << "Nodes: " << clf->getNumberOfNodes() << std::endl;
-            nodes += clf->getNumberOfNodes();
-            score_train = clf->score(Xtrain, ytrain);
-            score_test = clf->score(Xtest, ytest);
+        /*
+        * Begin Processing
+        */
+        auto handler = ArffFiles();
+        handler.load(complete_file_name, class_last);
+        // Get Dataset X, y
+        std::vector<mdlp::samples_t>& X = handler.getX();
+        mdlp::labels_t& y = handler.getY();
+        // Get className & Features
+        auto className = handler.getClassName();
+        std::vector<std::string> features;
+        auto attributes = handler.getAttributes();
+        transform(attributes.begin(), attributes.end(), back_inserter(features),
+            [](const pair<std::string, std::string>& item) { return item.first; });
+        // Discretize Dataset
+        auto [Xd, maxes] = discretize(X, y, features);
+        maxes[className] = *max_element(y.begin(), y.end()) + 1;
+        map<std::string, std::vector<int>> states;
+        for (auto feature : features) {
+            states[feature] = std::vector<int>(maxes[feature]);
         }
+        states[className] = std::vector<int>(maxes[className]);
+        auto clf = platform::Models::instance()->create(model_name);
+        bayesnet::Smoothing_t smoothing = bayesnet::Smoothing_t::ORIGINAL;
+        clf->fit(Xd, y, features, className, states, smoothing);
         if (dump_cpt) {
             std::cout << "--- CPT Tables ---" << std::endl;
             clf->dump_cpt();
         }
-        total_score_train += score_train;
-        total_score += score_test;
-        std::cout << "Score Train: " << score_train << std::endl;
-        std::cout << "Score Test : " << score_test << std::endl;
-        std::cout << "-------------------------------------------------------------------------------" << std::endl;
-    }
-    std::cout << "Nodes: " << nodes / nFolds << std::endl;
-    std::cout << "**********************************************************************************" << std::endl;
-    std::cout << "Average Score Train: " << total_score_train / nFolds << std::endl;
-    std::cout << "Average Score Test : " << total_score / nFolds << std::endl;return 0;
+        auto lines = clf->show();
+        for (auto line : lines) {
+            std::cout << line << std::endl;
+        }
+        std::cout << "--- Topological Order ---" << std::endl;
+        auto order = clf->topological_order();
+        for (auto name : order) {
+            std::cout << name << ", ";
+        }
+        std::cout << "end." << std::endl;
+        auto score = clf->score(Xd, y);
+        std::cout << "Score: " << score << std::endl;
+        auto graph = clf->graph();
+        auto dot_file = model_name + "_" + file_name;
+        ofstream file(dot_file + ".dot");
+        file << graph;
+        file.close();
+        std::cout << "Graph saved in " << model_name << "_" << file_name << ".dot" << std::endl;
+        std::cout << "dot -Tpng -o " + dot_file + ".png " + dot_file + ".dot " << std::endl;
+        std::string stratified_string = stratified ? " Stratified" : "";
+        std::cout << nFolds << " Folds" << stratified_string << " Cross validation" << std::endl;
+        std::cout << "==========================================" << std::endl;
+        torch::Tensor Xt = torch::zeros({ static_cast<int>(Xd.size()), static_cast<int>(Xd[0].size()) }, torch::kInt32);
+        torch::Tensor yt = torch::tensor(y, torch::kInt32);
+        for (int i = 0; i < features.size(); ++i) {
+            Xt.index_put_({ i, "..." }, torch::tensor(Xd[i], torch::kInt32));
+        }
+        float total_score = 0, total_score_train = 0, score_train, score_test;
+        folding::Fold* fold;
+        double nodes = 0.0;
+        if (stratified)
+            fold = new folding::StratifiedKFold(nFolds, y, seed);
+        else
+            fold = new folding::KFold(nFolds, y.size(), seed);
+        for (auto i = 0; i < nFolds; ++i) {
+            auto [train, test] = fold->getFold(i);
+            std::cout << "Fold: " << i + 1 << std::endl;
+            if (tensors) {
+                auto ttrain = torch::tensor(train, torch::kInt64);
+                auto ttest = torch::tensor(test, torch::kInt64);
+                torch::Tensor Xtraint = torch::index_select(Xt, 1, ttrain);
+                torch::Tensor ytraint = yt.index({ ttrain });
+                torch::Tensor Xtestt = torch::index_select(Xt, 1, ttest);
+                torch::Tensor ytestt = yt.index({ ttest });
+                clf->fit(Xtraint, ytraint, features, className, states, smoothing);
+                auto temp = clf->predict(Xtraint);
+                score_train = clf->score(Xtraint, ytraint);
+                score_test = clf->score(Xtestt, ytestt);
+            } else {
+                auto [Xtrain, ytrain] = extract_indices(train, Xd, y);
+                auto [Xtest, ytest] = extract_indices(test, Xd, y);
+                clf->fit(Xtrain, ytrain, features, className, states, smoothing);
+                std::cout << "Nodes: " << clf->getNumberOfNodes() << std::endl;
+                nodes += clf->getNumberOfNodes();
+                score_train = clf->score(Xtrain, ytrain);
+                score_test = clf->score(Xtest, ytest);
+            }
+            if (dump_cpt) {
+                std::cout << "--- CPT Tables ---" << std::endl;
+                clf->dump_cpt();
+            }
+            total_score_train += score_train;
+            total_score += score_test;
+            std::cout << "Score Train: " << score_train << std::endl;
+            std::cout << "Score Test : " << score_test << std::endl;
+            std::cout << "-------------------------------------------------------------------------------" << std::endl;
+        }
+        std::cout << "Nodes: " << nodes / nFolds << std::endl;
+        std::cout << "**********************************************************************************" << std::endl;
+        std::cout << "Average Score Train: " << total_score_train / nFolds << std::endl;
+        std::cout << "Average Score Test : " << total_score / nFolds << std::endl;return 0;
 }

src/CMakeLists.txt

@@ -2,7 +2,7 @@ include_directories(
     ## Libs
     ${Platform_SOURCE_DIR}/lib/Files
     ${Platform_SOURCE_DIR}/lib/folding
-    ${Platform_SOURCE_DIR}/lib/mdlp
+    ${Platform_SOURCE_DIR}/lib/mdlp/src
     ${Platform_SOURCE_DIR}/lib/argparse/include
     ${Platform_SOURCE_DIR}/lib/json/include
     ${Platform_SOURCE_DIR}/lib/libxlsxwriter/include
@@ -20,50 +20,50 @@ include_directories(
 # b_best
 add_executable(
     b_best commands/b_best.cpp best/Statistics.cpp
-    best/BestResultsExcel.cpp best/BestResults.cpp
-    common/Datasets.cpp common/Dataset.cpp
-    main/Models.cpp
+    best/BestResultsExcel.cpp best/BestResultsTex.cpp best/BestResultsMd.cpp best/BestResults.cpp
+    common/Datasets.cpp common/Dataset.cpp common/Discretization.cpp 
+    main/Models.cpp main/Scores.cpp
     reports/ReportExcel.cpp reports/ReportBase.cpp reports/ExcelFile.cpp
     results/Result.cpp
 )
-target_link_libraries(b_best Boost::boost "${PyClassifiers}" "${BayesNet}" ArffFiles mdlp ${Python3_LIBRARIES} "${TORCH_LIBRARIES}" ${LIBTORCH_PYTHON} Boost::python Boost::numpy "${XLSXWRITER_LIB}")
+target_link_libraries(b_best Boost::boost "${PyClassifiers}" "${BayesNet}" fimdlp ${Python3_LIBRARIES} "${TORCH_LIBRARIES}" ${LIBTORCH_PYTHON} Boost::python Boost::numpy "${XLSXWRITER_LIB}")
 
 # b_grid
 set(grid_sources GridSearch.cpp GridData.cpp)
 list(TRANSFORM grid_sources PREPEND grid/)
 add_executable(b_grid commands/b_grid.cpp ${grid_sources} 
-    common/Datasets.cpp common/Dataset.cpp
+    common/Datasets.cpp common/Dataset.cpp common/Discretization.cpp
     main/HyperParameters.cpp main/Models.cpp 
 )
-target_link_libraries(b_grid ${MPI_CXX_LIBRARIES} "${PyClassifiers}" "${BayesNet}" ArffFiles mdlp ${Python3_LIBRARIES} "${TORCH_LIBRARIES}" ${LIBTORCH_PYTHON} Boost::python Boost::numpy)
+target_link_libraries(b_grid ${MPI_CXX_LIBRARIES} "${PyClassifiers}" "${BayesNet}" fimdlp ${Python3_LIBRARIES} "${TORCH_LIBRARIES}" ${LIBTORCH_PYTHON} Boost::python Boost::numpy)
 
 # b_list
 add_executable(b_list commands/b_list.cpp
-    common/Datasets.cpp common/Dataset.cpp
+    common/Datasets.cpp common/Dataset.cpp common/Discretization.cpp
     main/Models.cpp main/Scores.cpp
     reports/ReportExcel.cpp reports/ExcelFile.cpp reports/ReportBase.cpp reports/DatasetsExcel.cpp reports/DatasetsConsole.cpp reports/ReportsPaged.cpp
     results/Result.cpp results/ResultsDatasetExcel.cpp results/ResultsDataset.cpp results/ResultsDatasetConsole.cpp
 )
-target_link_libraries(b_list "${PyClassifiers}" "${BayesNet}" ArffFiles mdlp ${Python3_LIBRARIES} "${TORCH_LIBRARIES}" ${LIBTORCH_PYTHON} Boost::python Boost::numpy "${XLSXWRITER_LIB}")
+target_link_libraries(b_list "${PyClassifiers}" "${BayesNet}" fimdlp ${Python3_LIBRARIES} "${TORCH_LIBRARIES}" ${LIBTORCH_PYTHON} Boost::python Boost::numpy "${XLSXWRITER_LIB}")
 
 # b_main
 set(main_sources Experiment.cpp Models.cpp HyperParameters.cpp Scores.cpp)
 list(TRANSFORM main_sources PREPEND main/)
 add_executable(b_main commands/b_main.cpp ${main_sources} 
-    common/Datasets.cpp common/Dataset.cpp 
+    common/Datasets.cpp common/Dataset.cpp common/Discretization.cpp
     reports/ReportConsole.cpp reports/ReportBase.cpp 
     results/Result.cpp
 )
-target_link_libraries(b_main "${PyClassifiers}" "${BayesNet}" ArffFiles mdlp ${Python3_LIBRARIES} "${TORCH_LIBRARIES}" ${LIBTORCH_PYTHON} Boost::python Boost::numpy)
+target_link_libraries(b_main "${PyClassifiers}" "${BayesNet}" fimdlp ${Python3_LIBRARIES} "${TORCH_LIBRARIES}" ${LIBTORCH_PYTHON} Boost::python Boost::numpy)
 
 # b_manage
-set(manage_sources ManageScreen.cpp CommandParser.cpp ResultsManager.cpp)
+set(manage_sources ManageScreen.cpp OptionsMenu.cpp ResultsManager.cpp)
 list(TRANSFORM manage_sources PREPEND manage/)
 add_executable(
     b_manage commands/b_manage.cpp ${manage_sources} 
-    common/Datasets.cpp common/Dataset.cpp
+    common/Datasets.cpp common/Dataset.cpp common/Discretization.cpp
     reports/ReportConsole.cpp reports/ReportExcel.cpp reports/ReportExcelCompared.cpp reports/ReportBase.cpp reports/ExcelFile.cpp reports/DatasetsConsole.cpp reports/ReportsPaged.cpp
     results/Result.cpp results/ResultsDataset.cpp results/ResultsDatasetConsole.cpp
     main/Scores.cpp
 )
-target_link_libraries(b_manage "${TORCH_LIBRARIES}" "${XLSXWRITER_LIB}" ArffFiles mdlp "${BayesNet}")
+target_link_libraries(b_manage "${TORCH_LIBRARIES}" "${XLSXWRITER_LIB}" fimdlp "${BayesNet}")

src/best/BestResults.cpp

@@ -6,8 +6,12 @@
 #include <algorithm>
 #include "common/Colors.h"
 #include "common/CLocale.h"
+#include "common/Paths.h"
+#include "common/Utils.h" // compute_std
 #include "results/Result.h"
 #include "BestResultsExcel.h"
+#include "BestResultsTex.h"
+#include "BestResultsMd.h"
 #include "best/Statistics.h"
 #include "BestResults.h"
 
@@ -51,7 +55,7 @@ namespace platform {
                     }
                 }
                 if (update) {
-                    bests[datasetName] = { item.at("score").get<double>(), item.at("hyperparameters"), file };
+                    bests[datasetName] = { item.at("score").get<double>(), item.at("hyperparameters"), file, item.at("score_std").get<double>() };
                 }
             }
         }
@@ -59,16 +63,12 @@ namespace platform {
             std::cerr << Colors::MAGENTA() << "No results found for model " << model << " and score " << score << Colors::RESET() << std::endl;
             exit(1);
         }
-        std::string bestFileName = path + bestResultFile();
+        std::string bestFileName = path + Paths::bestResultsFile(score, model);
         std::ofstream file(bestFileName);
         file << bests;
         file.close();
         return bestFileName;
     }
-    std::string BestResults::bestResultFile()
-    {
-        return "best_results_" + score + "_" + model + ".json";
-    }
     std::pair<std::string, std::string> getModelScore(std::string name)
     {
         // results_accuracy_BoostAODE_MacBookpro16_2023-09-06_12:27:00_1.json
@@ -150,7 +150,7 @@ namespace platform {
     }
     void BestResults::listFile()
     {
-        std::string bestFileName = path + bestResultFile();
+        std::string bestFileName = path + Paths::bestResultsFile(score, model);
         if (FILE* fileTest = fopen(bestFileName.c_str(), "r")) {
             fclose(fileTest);
         } else {
@@ -196,7 +196,7 @@ namespace platform {
         auto maxDate = std::filesystem::file_time_type::max();
         for (const auto& model : models) {
             this->model = model;
-            std::string bestFileName = path + bestResultFile();
+            std::string bestFileName = path + Paths::bestResultsFile(score, model);
             if (FILE* fileTest = fopen(bestFileName.c_str(), "r")) {
                 fclose(fileTest);
             } else {
@@ -213,13 +213,20 @@ namespace platform {
         table["dateTable"] = ftime_to_string(maxDate);
         return table;
     }
-    void BestResults::printTableResults(std::vector<std::string> models, json table)
+
+    void BestResults::printTableResults(std::vector<std::string> models, json table, bool tex)
     {
         std::stringstream oss;
         oss << Colors::GREEN() << "Best results for " << score << " as of " << table.at("dateTable").get<std::string>() << std::endl;
         std::cout << oss.str();
         std::cout << std::string(oss.str().size() - 8, '-') << std::endl;
         std::cout << Colors::GREEN() << " #  " << std::setw(maxDatasetName + 1) << std::left << std::string("Dataset");
+        auto bestResultsTex = BestResultsTex();
+        auto bestResultsMd = BestResultsMd();
+        if (tex) {
+            bestResultsTex.results_header(models, table.at("dateTable").get<std::string>());
+            bestResultsMd.results_header(models, table.at("dateTable").get<std::string>());
+        }
         for (const auto& model : models) {
             std::cout << std::setw(maxModelName) << std::left << model << " ";
         }
@@ -230,12 +237,13 @@ namespace platform {
         }
         std::cout << std::endl;
         auto i = 0;
-        std::map<std::string, double> totals;
+        std::map<std::string, std::vector<double>> totals;
         int nDatasets = table.begin().value().size();
-        for (const auto& model : models) {
-            totals[model] = 0.0;
-        }
         auto datasets = getDatasets(table.begin().value());
+        if (tex) {
+            bestResultsTex.results_body(datasets, table);
+            bestResultsMd.results_body(datasets, table);
+        }
         for (auto const& dataset_ : datasets) {
             auto color = (i % 2) ? Colors::BLUE() : Colors::CYAN();
             std::cout << color << std::setw(3) << std::fixed << std::right << i++ << " ";
@@ -270,7 +278,7 @@ namespace platform {
                 if (value == -1) {
                     std::cout << Colors::YELLOW() << std::setw(maxModelName) << std::right << "N/A" << " ";
                 } else {
-                    totals[model] += value;
+                    totals[model].push_back(value);
                     std::cout << efectiveColor << std::setw(maxModelName) << std::setprecision(maxModelName - 2) << std::fixed << value << " ";
                 }
             }
@@ -281,19 +289,26 @@ namespace platform {
             std::cout << std::string(maxModelName, '=') << " ";
         }
         std::cout << std::endl;
-        std::cout << Colors::GREEN() << "    Totals" << std::string(maxDatasetName - 6, '.') << " ";
+        std::cout << Colors::GREEN() << "    Average" << std::string(maxDatasetName - 7, '.') << " ";
         double max_value = 0.0;
+        std::string best_model = "";
         for (const auto& total : totals) {
-            if (total.second > max_value) {
-                max_value = total.second;
+            auto actual = std::reduce(total.second.begin(), total.second.end());
+            if (actual > max_value) {
+                max_value = actual;
+                best_model = total.first;
             }
         }
+        if (tex) {
+            bestResultsTex.results_footer(totals, best_model);
+            bestResultsMd.results_footer(totals, best_model);
+        }
         for (const auto& model : models) {
-            std::string efectiveColor = Colors::GREEN();
-            if (totals[model] == max_value) {
-                efectiveColor = Colors::RED();
-            }
-            std::cout << efectiveColor << std::right << std::setw(maxModelName) << std::setprecision(maxModelName - 4) << std::fixed << totals[model] << " ";
+            std::string efectiveColor = model == best_model ? Colors::RED() : Colors::GREEN();
+            double value = std::reduce(totals[model].begin(), totals[model].end()) / nDatasets;
+            double std_value = compute_std(totals[model], value);
+            std::cout << efectiveColor << std::right << std::setw(maxModelName) << std::setprecision(maxModelName - 4) << std::fixed << value << " ";
+
         }
         std::cout << std::endl;
     }
@@ -306,26 +321,34 @@ namespace platform {
             json table = buildTableResults(models);
             std::vector<std::string> datasets = getDatasets(table.begin().value());
             BestResultsExcel excel_report(score, datasets);
-            excel_report.reportSingle(model, path + bestResultFile());
-            messageExcelFile(excel_report.getFileName());
+            excel_report.reportSingle(model, path + Paths::bestResultsFile(score, model));
+            messageOutputFile("Excel", excel_report.getFileName());
         }
     }
-    void BestResults::reportAll(bool excel)
+    void BestResults::reportAll(bool excel, bool tex)
     {
         auto models = getModels();
         // Build the table of results
         json table = buildTableResults(models);
         std::vector<std::string> datasets = getDatasets(table.begin().value());
         // Print the table of results
-        printTableResults(models, table);
+        printTableResults(models, table, tex);
         // Compute the Friedman test
         std::map<std::string, std::map<std::string, float>> ranksModels;
         if (friedman) {
             Statistics stats(models, datasets, table, significance);
             auto result = stats.friedmanTest();
-            stats.postHocHolmTest(result);
+            stats.postHocHolmTest(result, tex);
             ranksModels = stats.getRanks();
         }
+        if (tex) {
+            messageOutputFile("TeX", Paths::tex() + Paths::tex_output());
+            messageOutputFile("MarkDown", Paths::tex() + Paths::md_output());
+            if (friedman) {
+                messageOutputFile("TeX", Paths::tex() + Paths::tex_post_hoc());
+                messageOutputFile("MarkDown", Paths::tex() + Paths::md_post_hoc());
+            }
+        }
         if (excel) {
             BestResultsExcel excel(score, datasets);
             excel.reportAll(models, table, ranksModels, friedman, significance);
@@ -346,13 +369,14 @@ namespace platform {
                     }
                 }
                 model = models.at(idx);
-                excel.reportSingle(model, path + bestResultFile());
+                excel.reportSingle(model, path + Paths::bestResultsFile(score, model));
             }
-            messageExcelFile(excel.getFileName());
+            messageOutputFile("Excel", excel.getFileName());
         }
     }
-    void BestResults::messageExcelFile(const std::string& fileName)
+    void BestResults::messageOutputFile(const std::string& title, const std::string& fileName)
     {
-        std::cout << Colors::YELLOW() << "** Excel file generated: " << fileName << Colors::RESET() << std::endl;
+        std::cout << Colors::YELLOW() << "** " << std::setw(5) << std::left << title
+            << " file generated: " << fileName << Colors::RESET() << std::endl;
     }
 }

src/best/BestResults.h

@@ -13,16 +13,15 @@ namespace platform {
         }
         std::string build();
         void reportSingle(bool excel);
-        void reportAll(bool excel);
+        void reportAll(bool excel, bool tex);
         void buildAll();
     private:
         std::vector<std::string> getModels();
         std::vector<std::string> getDatasets(json table);
         std::vector<std::string> loadResultFiles();
-        void messageExcelFile(const std::string& fileName);
+        void messageOutputFile(const std::string& title, const std::string& fileName);
         json buildTableResults(std::vector<std::string> models);
-        void printTableResults(std::vector<std::string> models, json table);
-        std::string bestResultFile();
+        void printTableResults(std::vector<std::string> models, json table, bool tex);
         json loadFile(const std::string& fileName);
         void listFile();
         std::string path;

src/best/BestResultsExcel.cpp

@@ -32,7 +32,7 @@ namespace platform {
     }
     BestResultsExcel::BestResultsExcel(const std::string& score, const std::vector<std::string>& datasets) : score(score), datasets(datasets)
     {
-        file_name = "BestResults.xlsx";
+        file_name = Paths::bestResultsExcel(score);
         workbook = workbook_new(getFileName().c_str());
         setProperties("Best Results");
         int maxDatasetName = (*max_element(datasets.begin(), datasets.end(), [](const std::string& a, const std::string& b) { return a.size() < b.size(); })).size();
@@ -64,19 +64,21 @@ namespace platform {
         json data = loadResultData(fileName);
 
         std::string title = "Best results for " + model;
-        worksheet_merge_range(worksheet, 0, 0, 0, 4, title.c_str(), styles["headerFirst"]);
+        worksheet_merge_range(worksheet, 0, 0, 0, 5, title.c_str(), styles["headerFirst"]);
         // Body header
         row = 3;
         int col = 1;
-        writeString(row, 0, "Nº", "bodyHeader");
+        writeString(row, 0, "#", "bodyHeader");
         writeString(row, 1, "Dataset", "bodyHeader");
         writeString(row, 2, "Score", "bodyHeader");
         writeString(row, 3, "File", "bodyHeader");
         writeString(row, 4, "Hyperparameters", "bodyHeader");
+        writeString(row, 5, "F", "bodyHeader");
         auto i = 0;
         std::string hyperparameters;
         int hypSize = 22;
         std::map<std::string, std::string> files; // map of files imported and their tabs
+        int numLines = data.size();
         for (auto const& item : data.items()) {
             row++;
             writeInt(row, 0, i++, "ints");
@@ -104,6 +106,8 @@ namespace platform {
                 hypSize = hyperparameters.size();
             }
             writeString(row, 4, hyperparameters, "text");
+            std::string countHyperparameters = "=COUNTIF(e5:e" + std::to_string(numLines + 4) + ", e" + std::to_string(row + 1) + ")";
+            worksheet_write_formula(worksheet, row, 5, countHyperparameters.c_str(), efectiveStyle("ints"));
         }
         row++;
         // Set Totals
@@ -180,7 +184,7 @@ namespace platform {
         // Body header
         row = 3;
         int col = 1;
-        writeString(row, 0, "Nº", "bodyHeader");
+        writeString(row, 0, "#", "bodyHeader");
         writeString(row, 1, "Dataset", "bodyHeader");
         for (const auto& model : models) {
             writeString(row, ++col, model.c_str(), "bodyHeader");

src/best/BestResultsMd.cpp

@@ -0,0 +1,103 @@
+#include <iostream>
+#include "BestResultsMd.h"
+#include "common/Utils.h" // compute_std
+
+namespace platform {
+    using json = nlohmann::ordered_json;
+    void BestResultsMd::openMdFile(const std::string& name)
+    {
+        handler.open(name);
+        if (!handler.is_open()) {
+            std::cerr << "Error opening file " << name << std::endl;
+            exit(1);
+        }
+    }
+    void BestResultsMd::results_header(const std::vector<std::string>& models, const std::string& date)
+    {
+        this->models = models;
+        auto file_name = Paths::tex() + Paths::md_output();
+        openMdFile(file_name);
+        handler << "<!-- This file has been generated by the platform program" << std::endl;
+        handler << "  Date: " << date.c_str() << std::endl;
+        handler << "" << std::endl;
+        handler << "  Table of results" << std::endl;
+        handler << "-->" << std::endl;
+        handler << "| # | Dataset |";
+        for (const auto& model : models) {
+            handler << " " << model.c_str() << " |";
+        }
+        handler << std::endl;
+        handler << "|--: | :--- |";
+        for (const auto& model : models) {
+            handler << " :---: |";
+        }
+        handler << std::endl;
+    }
+    void BestResultsMd::results_body(const std::vector<std::string>& datasets, json& table)
+    {
+        int i = 0;
+        for (auto const& dataset : datasets) {
+            // Find out max value for this dataset
+            double max_value = 0;
+            // Find out the max value for this dataset
+            for (const auto& model : models) {
+                double value;
+                try {
+                    value = table[model].at(dataset).at(0).get<double>();
+                }
+                catch (nlohmann::json_abi_v3_11_3::detail::out_of_range err) {
+                    value = -1.0;
+                }
+                if (value > max_value) {
+                    max_value = value;
+                }
+            }
+            handler << "| " << ++i << " | " << dataset.c_str() << " | ";
+            for (const auto& model : models) {
+                double value = table[model].at(dataset).at(0).get<double>();
+                double std_value = table[model].at(dataset).at(3).get<double>();
+                const char* bold = value == max_value ? "**" : "";
+                handler << bold << std::setprecision(4) << std::fixed << value << "±" << std::setprecision(3) << std_value << bold << " | ";
+            }
+            handler << std::endl;
+        }
+    }
+    void BestResultsMd::results_footer(const std::map<std::string, std::vector<double>>& totals, const std::string& best_model)
+    {
+        handler << "| | **Average Score** | ";
+        int nDatasets = totals.begin()->second.size();
+        for (const auto& model : models) {
+            double value = std::reduce(totals.at(model).begin(), totals.at(model).end()) / nDatasets;
+            double std_value = compute_std(totals.at(model), value);
+            const char* bold = model == best_model ? "**" : "";
+            handler << bold << std::setprecision(4) << std::fixed << value << "±" << std::setprecision(3) << std::fixed << std_value << bold << " | ";
+        }
+
+        handler.close();
+    }
+    void BestResultsMd::holm_test(struct HolmResult& holmResult, const std::string& date)
+    {
+        auto file_name = Paths::tex() + Paths::md_post_hoc();
+        openMdFile(file_name);
+        handler << "<!-- This file has been generated by the platform program" << std::endl;
+        handler << "  Date: " << date.c_str() << std::endl;
+        handler << std::endl;
+        handler << "  Post-hoc handler test" << std::endl;
+        handler << "-->" << std::endl;
+        handler << "Post-hoc Holm test: H<sub>0</sub>: There is no significant differences between the control model and the other models." << std::endl << std::endl;
+        handler << "| classifier | pvalue | rank | win | tie | loss | H<sub>0</sub> |" << std::endl;
+        handler << "| :-- | --: | --: | --:| --: | --: | :--: |" << std::endl;
+        for (auto const& line : holmResult.holmLines) {
+            auto textStatus = !line.reject ? "**" : " ";
+            if (line.model == holmResult.model) {
+                handler << "| " << line.model << " | - | " << std::fixed << std::setprecision(2) << line.rank << " | - | - | - |" << std::endl;
+            } else {
+                handler << "| " << line.model << " | " << textStatus << std::scientific << std::setprecision(4) << line.pvalue << textStatus << " |";
+                handler << std::fixed << std::setprecision(2) << line.rank << " | " << line.wtl.win << " | " << line.wtl.tie << " | " << line.wtl.loss << " |";
+                handler << (line.reject ? "rejected" : "**accepted**") << " |" << std::endl;
+            }
+        }
+        handler << std::endl;
+        handler.close();
+    }
+}

src/best/BestResultsMd.h

@@ -0,0 +1,24 @@
+#ifndef BEST_RESULTS_MD_H
+#define BEST_RESULTS_MD_H
+#include <map>
+#include <vector>
+#include <nlohmann/json.hpp>
+#include "common/Paths.h"
+#include "Statistics.h"
+namespace platform {
+    using json = nlohmann::ordered_json;
+    class BestResultsMd {
+    public:
+        BestResultsMd() = default;
+        ~BestResultsMd() = default;
+        void results_header(const std::vector<std::string>& models, const std::string& date);
+        void results_body(const std::vector<std::string>& datasets, json& table);
+        void results_footer(const std::map<std::string, std::vector<double>>& totals, const std::string& best_model);
+        void holm_test(struct HolmResult& holmResult, const std::string& date);
+    private:
+        void openMdFile(const std::string& name);
+        std::ofstream handler;
+        std::vector<std::string> models;
+    };
+}
+#endif

src/best/BestResultsTex.cpp

@@ -0,0 +1,117 @@
+#include <iostream>
+#include "BestResultsTex.h"
+#include "common/Utils.h" // compute_std
+
+namespace platform {
+    using json = nlohmann::ordered_json;
+    void BestResultsTex::openTexFile(const std::string& name)
+    {
+        handler.open(name);
+        if (!handler.is_open()) {
+            std::cerr << "Error opening file " << name << std::endl;
+            exit(1);
+        }
+    }
+    void BestResultsTex::results_header(const std::vector<std::string>& models, const std::string& date)
+    {
+        this->models = models;
+        auto file_name = Paths::tex() + Paths::tex_output();
+        openTexFile(file_name);
+        handler << "%% This file has been generated by the platform program" << std::endl;
+        handler << "%% Date: " << date.c_str() << std::endl;
+        handler << "%%" << std::endl;
+        handler << "%% Table of results" << std::endl;
+        handler << "%%" << std::endl;
+        handler << "\\begin{table}[htbp] " << std::endl;
+        handler << "\\centering " << std::endl;
+        handler << "\\tiny " << std::endl;
+        handler << "\\renewcommand{\\arraystretch }{1.2} " << std::endl;
+        handler << "\\renewcommand{\\tabcolsep }{0.07cm} " << std::endl;
+        handler << "\\caption{Accuracy results(mean $\\pm$ std) for all the algorithms and datasets} " << std::endl;
+        handler << "\\label{tab:results_accuracy}" << std::endl;
+        handler << "\\begin{tabular} {{r" << std::string(models.size(), 'c').c_str() << "}}" << std::endl;
+        handler << "\\hline " << std::endl;
+        handler << "" << std::endl;
+        for (const auto& model : models) {
+            handler << "& " << model.c_str();
+        }
+        handler << "\\\\" << std::endl;
+        handler << "\\hline" << std::endl;
+    }
+    void BestResultsTex::results_body(const std::vector<std::string>& datasets, json& table)
+    {
+        int i = 0;
+        for (auto const& dataset : datasets) {
+            // Find out max value for this dataset
+            double max_value = 0;
+            // Find out the max value for this dataset
+            for (const auto& model : models) {
+                double value;
+                try {
+                    value = table[model].at(dataset).at(0).get<double>();
+                }
+                catch (nlohmann::json_abi_v3_11_3::detail::out_of_range err) {
+                    value = -1.0;
+                }
+                if (value > max_value) {
+                    max_value = value;
+                }
+            }
+            handler << ++i << " ";
+            for (const auto& model : models) {
+                double value = table[model].at(dataset).at(0).get<double>();
+                double std_value = table[model].at(dataset).at(3).get<double>();
+                const char* bold = value == max_value ? "\\bfseries" : "";
+                handler << "& " << bold << std::setprecision(4) << std::fixed << value << "$\\pm$" << std::setprecision(3) << std_value;
+            }
+            handler << "\\\\" << std::endl;
+        }
+    }
+    void BestResultsTex::results_footer(const std::map<std::string, std::vector<double>>& totals, const std::string& best_model)
+    {
+        handler << "\\hline" << std::endl;
+        handler << "Average ";
+        int nDatasets = totals.begin()->second.size();
+        for (const auto& model : models) {
+            double value = std::reduce(totals.at(model).begin(), totals.at(model).end()) / nDatasets;
+            double std_value = compute_std(totals.at(model), value);
+            const char* bold = model == best_model ? "\\bfseries" : "";
+            handler << "& " << bold << std::setprecision(4) << std::fixed << value << "$\\pm$" << std::setprecision(3) << std::fixed << std_value;
+        }
+        handler << "\\\\" << std::endl;
+        handler << "\\hline " << std::endl;
+        handler << "\\end{tabular}" << std::endl;
+        handler << "\\end{table}" << std::endl;
+        handler.close();
+    }
+    void BestResultsTex::holm_test(struct HolmResult& holmResult, const std::string& date)
+    {
+        auto file_name = Paths::tex() + Paths::tex_post_hoc();
+        openTexFile(file_name);
+        handler << "%% This file has been generated by the platform program" << std::endl;
+        handler << "%% Date: " << date.c_str() << std::endl;
+        handler << "%%" << std::endl;
+        handler << "%% Post-hoc handler test" << std::endl;
+        handler << "%%" << std::endl;
+        handler << "\\begin{table}[htbp]" << std::endl;
+        handler << "\\centering" << std::endl;
+        handler << "\\caption{Results of the post-hoc test for the mean accuracy of the algorithms.}\\label{tab:tests}" << std::endl;
+        handler << "\\begin{tabular}{lrrrrr}" << std::endl;
+        handler << "\\hline" << std::endl;
+        handler << "classifier & pvalue & rank & win & tie & loss\\\\" << std::endl;
+        handler << "\\hline" << std::endl;
+        for (auto const& line : holmResult.holmLines) {
+            auto textStatus = !line.reject ? "\\bf " : " ";
+            if (line.model == holmResult.model) {
+                handler << line.model << " & - & " << std::fixed << std::setprecision(2) << line.rank << " & - & - & - \\\\" << std::endl;
+            } else {
+                handler << line.model << " & " << textStatus << std::scientific << std::setprecision(4) << line.pvalue << " & ";
+                handler << std::fixed << std::setprecision(2) << line.rank << " & " << line.wtl.win << " & " << line.wtl.tie << " & " << line.wtl.loss << "\\\\" << std::endl;
+            }
+        }
+        handler << "\\hline " << std::endl;
+        handler << "\\end{tabular}" << std::endl;
+        handler << "\\end{table}" << std::endl;
+        handler.close();
+    }
+}

src/best/BestResultsTex.h

@@ -0,0 +1,24 @@
+#ifndef BEST_RESULTS_TEX_H
+#define BEST_RESULTS_TEX_H
+#include <map>
+#include <vector>
+#include <nlohmann/json.hpp>
+#include "common/Paths.h"
+#include "Statistics.h"
+namespace platform {
+    using json = nlohmann::ordered_json;
+    class BestResultsTex {
+    public:
+        BestResultsTex() = default;
+        ~BestResultsTex() = default;
+        void results_header(const std::vector<std::string>& models, const std::string& date);
+        void results_body(const std::vector<std::string>& datasets, json& table);
+        void results_footer(const std::map<std::string, std::vector<double>>& totals, const std::string& best_model);
+        void holm_test(struct HolmResult& holmResult, const std::string& date);
+    private:
+        void openTexFile(const std::string& name);
+        std::ofstream handler;
+        std::vector<std::string> models;
+    };
+}
+#endif

src/best/Statistics.cpp

@@ -4,6 +4,8 @@
 #include "common/Colors.h"
 #include "common/Symbols.h"
 #include "common/CLocale.h"
+#include "BestResultsTex.h"
+#include "BestResultsMd.h"
 #include "Statistics.h"
 
 
@@ -113,7 +115,7 @@ namespace platform {
         }
     }
 
-    void Statistics::postHocHolmTest(bool friedmanResult)
+    void Statistics::postHocHolmTest(bool friedmanResult, bool tex)
     {
         if (!fitted) {
             fit();
@@ -130,7 +132,7 @@ namespace platform {
                 stats[i] = 0.0;
                 continue;
             }
-            double z = abs(ranks.at(models[controlIdx]) - ranks.at(models[i])) / diff;
+            double z = std::abs(ranks.at(models[controlIdx]) - ranks.at(models[i])) / diff;
             double p_value = (long double)2 * (1 - cdf(dist, z));
             stats[i] = p_value;
         }
@@ -195,6 +197,12 @@ namespace platform {
         if (output) {
             std::cout << oss.str();
         }
+        if (tex) {
+            BestResultsTex bestResultsTex;
+            BestResultsMd bestResultsMd;
+            bestResultsTex.holm_test(holmResult, get_date() + " " + get_time());
+            bestResultsMd.holm_test(holmResult, get_date() + " " + get_time());
+        }
     }
     bool Statistics::friedmanTest()
     {

src/best/Statistics.h

@@ -34,7 +34,7 @@ namespace platform {
     public:
         Statistics(const std::vector<std::string>& models, const std::vector<std::string>& datasets, const json& data, double significance = 0.05, bool output = true);
         bool friedmanTest();
-        void postHocHolmTest(bool friedmanResult);
+        void postHocHolmTest(bool friedmanResult, bool tex=false);
         FriedmanResult& getFriedmanResult();
         HolmResult& getHolmResult();
         std::map<std::string, std::map<std::string, float>>& getRanks();

src/commands/b_best.cpp

@@ -5,26 +5,18 @@
 #include "common/Paths.h"
 #include "common/Colors.h"
 #include "best/BestResults.h"
-#include "config.h"
+#include "config_platform.h"
 
 void manageArguments(argparse::ArgumentParser& program)
 {
     program.add_argument("-m", "--model")
-        .help("Model to use: " + platform::Models::instance()->toString() + " or any")
-        .action([](const std::string& value) {
-        std::vector<std::string> valid(platform::Models::instance()->getNames());
-        valid.push_back("any");
-        static const std::vector<std::string> choices = valid;
-        if (find(choices.begin(), choices.end(), value) != choices.end()) {
-            return value;
-        }
-        throw std::runtime_error("Model must be one of " + platform::Models::instance()->toString() + " or any");
-            }
-    );
+        .help("Model to use or any")
+        .default_value("any");
     program.add_argument("-d", "--dataset").default_value("any").help("Filter results of the selected model) (any for all datasets)");
     program.add_argument("-s", "--score").default_value("accuracy").help("Filter results of the score name supplied");
     program.add_argument("--friedman").help("Friedman test").default_value(false).implicit_value(true);
     program.add_argument("--excel").help("Output to excel").default_value(false).implicit_value(true);
+    program.add_argument("--tex").help("Output result table to TeX file").default_value(false).implicit_value(true);
     program.add_argument("--level").help("significance level").default_value(0.05).scan<'g', double>().action([](const std::string& value) {
         try {
             auto k = std::stod(value);
@@ -46,7 +38,7 @@ int main(int argc, char** argv)
     argparse::ArgumentParser program("b_best", { platform_project_version.begin(), platform_project_version.end() });
     manageArguments(program);
     std::string model, dataset, score;
-    bool build, report, friedman, excel;
+    bool build, report, friedman, excel, tex;
     double level;
     try {
         program.parse_args(argc, argv);
@@ -55,6 +47,7 @@ int main(int argc, char** argv)
         score = program.get<std::string>("score");
         friedman = program.get<bool>("friedman");
         excel = program.get<bool>("excel");
+        tex = program.get<bool>("tex");
         level = program.get<double>("level");
         if (model == "" || score == "") {
             throw std::runtime_error("Model and score name must be supplied");
@@ -74,7 +67,7 @@ int main(int argc, char** argv)
     auto results = platform::BestResults(platform::Paths::results(), score, model, dataset, friedman, level);
     if (model == "any") {
         results.buildAll();
-        results.reportAll(excel);
+        results.reportAll(excel, tex);
     } else {
         std::string fileName = results.build();
         std::cout << Colors::GREEN() << fileName << " created!" << Colors::RESET() << std::endl;

src/commands/b_grid.cpp

@@ -11,7 +11,7 @@
 #include "common/Colors.h"
 #include "common/DotEnv.h"
 #include "grid/GridSearch.h"
-#include "config.h"
+#include "config_platform.h"
 
 using json = nlohmann::ordered_json;
 const int MAXL = 133;
@@ -93,8 +93,10 @@ void list_dump(std::string& model)
         if (item.first.size() > max_dataset) {
             max_dataset = item.first.size();
         }
-        if (item.second.dump().size() > max_hyper) {
-            max_hyper = item.second.dump().size();
+        for (auto const& [key, value] : item.second.items()) {
+            if (value.dump().size() > max_hyper) {
+                max_hyper = value.dump().size();
+            }
         }
     }
     std::cout << Colors::GREEN() << left << " #  " << left << setw(max_dataset) << "Dataset" << " #Com. "
@@ -106,7 +108,12 @@ void list_dump(std::string& model)
         std::cout << color;
         auto num_combinations = data.getNumCombinations(item.first);
         std::cout << setw(3) << fixed << right << ++index << left << " " << setw(max_dataset) << item.first
-            << " " << setw(5) << right << num_combinations << " " << setw(max_hyper) << left << item.second.dump() << std::endl;
+            << " " << setw(5) << right << num_combinations << " ";
+        std::string prefix = "";
+        for (auto const& [key, value] : item.second.items()) {
+            std::cout << prefix << setw(max_hyper) << std::left << value.dump() << std::endl;
+            prefix = string(11 + max_dataset, ' ');
+        }
     }
     std::cout << Colors::RESET() << std::endl;
 }

src/commands/b_list.cpp

@@ -13,7 +13,7 @@
 #include "results/ResultsDatasetConsole.h"
 #include "results/ResultsDataset.h"
 #include "results/ResultsDatasetExcel.h"
-#include "config.h"
+#include "config_platform.h"
 
 
 void list_datasets(argparse::ArgumentParser& program)
@@ -76,18 +76,8 @@ int main(int argc, char** argv)
             }
         );
     results_command.add_argument("-m", "--model")
-        .help("Model to use: " + platform::Models::instance()->toString() + " or any")
-        .default_value("any")
-        .action([](const std::string& value) {
-        std::vector<std::string> valid(platform::Models::instance()->getNames());
-        valid.push_back("any");
-        static const std::vector<std::string> choices = valid;
-        if (find(choices.begin(), choices.end(), value) != choices.end()) {
-            return value;
-        }
-        throw std::runtime_error("Model must be one of " + platform::Models::instance()->toString() + " or any");
-            }
-        );
+        .help("Model to use or any")
+        .default_value("any");
     results_command.add_argument("--excel").help("Output in Excel format").default_value(false).implicit_value(true);
     results_command.add_argument("-s", "--score").default_value("accuracy").help("Filter results of the score name supplied");
 

src/commands/b_main.cpp

@@ -7,11 +7,12 @@
 #include "common/Paths.h"
 #include "main/Models.h"
 #include "main/modelRegister.h"
-#include "config.h"
+#include "config_platform.h"
 
 
 using json = nlohmann::ordered_json;
 
+
 void manageArguments(argparse::ArgumentParser& program)
 {
     auto env = platform::DotEnv();
@@ -35,6 +36,7 @@ void manageArguments(argparse::ArgumentParser& program)
     program.add_argument("--hyperparameters").default_value("{}").help("Hyperparameters passed to the model in Experiment");
     program.add_argument("--hyper-file").default_value("").help("Hyperparameters file name." \
         "Mutually exclusive with hyperparameters. This file should contain hyperparameters for each dataset in json format.");
+    program.add_argument("--hyper-best").default_value(false).help("Use best results of the model as source of hyperparameters").implicit_value(true);
     program.add_argument("-m", "--model")
         .help("Model to use: " + platform::Models::instance()->toString())
         .action([](const std::string& value) {
@@ -47,6 +49,23 @@ void manageArguments(argparse::ArgumentParser& program)
         );
     program.add_argument("--title").default_value("").help("Experiment title");
     program.add_argument("--discretize").help("Discretize input dataset").default_value((bool)stoi(env.get("discretize"))).implicit_value(true);
+    auto valid_choices = env.valid_tokens("discretize_algo");
+    auto& disc_arg = program.add_argument("--discretize-algo").help("Algorithm to use in discretization. Valid values: " + env.valid_values("discretize_algo")).default_value(env.get("discretize_algo"));
+    for (auto choice : valid_choices) {
+        disc_arg.choices(choice);
+    }
+    valid_choices = env.valid_tokens("smooth_strat");
+    auto& smooth_arg = program.add_argument("--smooth-strat").help("Smooth strategy used in Bayes Network node initialization. Valid values: " + env.valid_values("smooth_strat")).default_value(env.get("smooth_strat"));
+    for (auto choice : valid_choices) {
+        smooth_arg.choices(choice);
+    }
+    auto& score_arg = program.add_argument("-s", "--score").help("Score to use. Valid values: " + env.valid_values("score")).default_value(env.get("score"));
+    valid_choices = env.valid_tokens("score");
+    for (auto choice : valid_choices) {
+        score_arg.choices(choice);
+    }
+    program.add_argument("--generate-fold-files").help("generate fold information in datasets_experiment folder").default_value(false).implicit_value(true);
+    program.add_argument("--graph").help("generate graphviz dot files with the model").default_value(false).implicit_value(true);
     program.add_argument("--no-train-score").help("Don't compute train score").default_value(false).implicit_value(true);
     program.add_argument("--quiet").help("Don't display detailed progress").default_value(false).implicit_value(true);
     program.add_argument("--save").help("Save result (always save if no dataset is supplied)").default_value(false).implicit_value(true);
@@ -66,16 +85,16 @@ void manageArguments(argparse::ArgumentParser& program)
             throw std::runtime_error("Number of folds must be an integer");
         }});
         auto seed_values = env.getSeeds();
-        program.add_argument("-s", "--seeds").nargs(1, 10).help("Random seeds. Set to -1 to have pseudo random").scan<'i', int>().default_value(seed_values);
+        program.add_argument("--seeds").nargs(1, 10).help("Random seeds. Set to -1 to have pseudo random").scan<'i', int>().default_value(seed_values);
 }
 
 int main(int argc, char** argv)
 {
     argparse::ArgumentParser program("b_main", { platform_project_version.begin(), platform_project_version.end() });
     manageArguments(program);
-    std::string file_name, model_name, title, hyperparameters_file, datasets_file;
+    std::string file_name, model_name, title, hyperparameters_file, datasets_file, discretize_algo, smooth_strat, score;
     json hyperparameters_json;
-    bool discretize_dataset, stratified, saveResults, quiet, no_train_score;
+    bool discretize_dataset, stratified, saveResults, quiet, no_train_score, generate_fold_files, graph, hyper_best;
     std::vector<int> seeds;
     std::vector<std::string> file_names;
     std::vector<std::string> filesToTest;
@@ -87,20 +106,33 @@ int main(int argc, char** argv)
         datasets_file = program.get<std::string>("datasets-file");
         model_name = program.get<std::string>("model");
         discretize_dataset = program.get<bool>("discretize");
+        discretize_algo = program.get<std::string>("discretize-algo");
+        smooth_strat = program.get<std::string>("smooth-strat");
         stratified = program.get<bool>("stratified");
         quiet = program.get<bool>("quiet");
+        graph = program.get<bool>("graph");
         n_folds = program.get<int>("folds");
+        score = program.get<std::string>("score");
         seeds = program.get<std::vector<int>>("seeds");
         auto hyperparameters = program.get<std::string>("hyperparameters");
         hyperparameters_json = json::parse(hyperparameters);
         hyperparameters_file = program.get<std::string>("hyper-file");
         no_train_score = program.get<bool>("no-train-score");
-        if (hyperparameters_file != "" && hyperparameters != "{}") {
-            throw runtime_error("hyperparameters and hyper_file are mutually exclusive");
+        hyper_best = program.get<bool>("hyper-best");
+        generate_fold_files = program.get<bool>("generate-fold-files");
+        if (hyper_best) {
+            // Build the best results file_name
+            hyperparameters_file = platform::Paths::results() + platform::Paths::bestResultsFile(score, model_name);
+            // ignore this parameter
+            hyperparameters = "{}";
+        } else {
+            if (hyperparameters_file != "" && hyperparameters != "{}") {
+                throw runtime_error("hyperparameters and hyper_file are mutually exclusive");
+            }
         }
         title = program.get<std::string>("title");
-        if (title == "" && file_name == "") {
-            throw runtime_error("title is mandatory if dataset is not provided");
+        if (title == "" && file_name == "all") {
+            throw runtime_error("title is mandatory if all datasets are to be tested");
         }
         saveResults = program.get<bool>("save");
     }
@@ -109,7 +141,7 @@ int main(int argc, char** argv)
         cerr << program;
         exit(1);
     }
-    auto datasets = platform::Datasets(discretize_dataset, platform::Paths::datasets());
+    auto datasets = platform::Datasets(false, platform::Paths::datasets());
     if (datasets_file != "") {
         ifstream catalog(datasets_file);
         if (catalog.is_open()) {
@@ -165,7 +197,7 @@ int main(int argc, char** argv)
 
     platform::HyperParameters test_hyperparams;
     if (hyperparameters_file != "") {
-        test_hyperparams = platform::HyperParameters(datasets.getNames(), hyperparameters_file);
+        test_hyperparams = platform::HyperParameters(datasets.getNames(), hyperparameters_file, hyper_best);
     } else {
         test_hyperparams = platform::HyperParameters(datasets.getNames(), hyperparameters_json);
     }
@@ -175,24 +207,28 @@ int main(int argc, char** argv)
      */
     auto env = platform::DotEnv();
     auto experiment = platform::Experiment();
-    experiment.setTitle(title).setLanguage("c++").setLanguageVersion("13.2.1");
+    experiment.setTitle(title).setLanguage("c++").setLanguageVersion("gcc 14.1.1");
+    experiment.setDiscretizationAlgorithm(discretize_algo).setSmoothSrategy(smooth_strat);
     experiment.setDiscretized(discretize_dataset).setModel(model_name).setPlatform(env.get("platform"));
-    experiment.setStratified(stratified).setNFolds(n_folds).setScoreName("accuracy");
+    experiment.setStratified(stratified).setNFolds(n_folds).setScoreName(score);
     experiment.setHyperparameters(test_hyperparams);
     for (auto seed : seeds) {
         experiment.addRandomSeed(seed);
     }
     platform::Timer timer;
     timer.start();
-    experiment.go(filesToTest, quiet, no_train_score);
+    experiment.go(filesToTest, quiet, no_train_score, generate_fold_files, graph);
     experiment.setDuration(timer.getDuration());
-    if (saveResults) {
-        experiment.saveResult();
-    }
     if (!quiet) {
         // Classification report if only one dataset is tested
         experiment.report(filesToTest.size() == 1);
     }
+    if (saveResults) {
+        experiment.saveResult();
+    }
+    if (graph) {
+        experiment.saveGraph();
+    }
     std::cout << "Done!" << std::endl;
     return 0;
 }

src/commands/b_manage.cpp

@@ -4,7 +4,10 @@
 #include <unistd.h>
 #include <argparse/argparse.hpp>
 #include "manage/ManageScreen.h"
-#include "config.h"
+#include <signal.h>
+#include "config_platform.h"
+
+platform::ManageScreen* manager = nullptr;
 
 void manageArguments(argparse::ArgumentParser& program, int argc, char** argv)
 {
@@ -42,6 +45,11 @@ std::pair<int, int> numRowsCols()
     return { ts.ws_row, ts.ws_col };
 #endif /* TIOCGSIZE */
 }
+void handleResize(int sig)
+{
+    auto [rows, cols] = numRowsCols();
+    manager->updateSize(rows, cols);
+}
 
 int main(int argc, char** argv)
 {
@@ -50,13 +58,15 @@ int main(int argc, char** argv)
     std::string model = program.get<std::string>("model");
     std::string score = program.get<std::string>("score");
     std::string platform = program.get<std::string>("platform");
-    auto complete = program.get<bool>("complete");
-    auto partial = program.get<bool>("partial");
-    auto compare = program.get<bool>("compare");
-    auto [rows, cols] = numRowsCols();
+    bool complete = program.get<bool>("complete");
+    bool partial = program.get<bool>("partial");
+    bool compare = program.get<bool>("compare");
     if (complete)
         partial = false;
-    auto manager = platform::ManageScreen(rows, cols, model, score, platform, complete, partial, compare);
-    manager.doMenu();
+    signal(SIGWINCH, handleResize);
+    auto [rows, cols] = numRowsCols();
+    manager = new platform::ManageScreen(rows, cols, model, score, platform, complete, partial, compare);
+    manager->doMenu();
+    delete manager;
     return 0;
 }

src/common/Dataset.cpp

@@ -1,24 +1,26 @@
-#include <ArffFiles.h>
+#include <ArffFiles.hpp>
 #include <fstream>
 #include "Dataset.h"
 namespace platform {
-    Dataset::Dataset(const Dataset& dataset) : path(dataset.path), name(dataset.name), className(dataset.className), n_samples(dataset.n_samples), n_features(dataset.n_features), features(dataset.features), states(dataset.states), loaded(dataset.loaded), discretize(dataset.discretize), X(dataset.X), y(dataset.y), Xv(dataset.Xv), Xd(dataset.Xd), yv(dataset.yv), fileType(dataset.fileType)
+    const std::string message_dataset_not_loaded = "Dataset not loaded.";
+    Dataset::Dataset(const Dataset& dataset) :
+        path(dataset.path), name(dataset.name), className(dataset.className), n_samples(dataset.n_samples),
+        n_features(dataset.n_features), numericFeatures(dataset.numericFeatures), features(dataset.features),
+        states(dataset.states), loaded(dataset.loaded), discretize(dataset.discretize), X(dataset.X), y(dataset.y),
+        X_train(dataset.X_train), X_test(dataset.X_test), Xv(dataset.Xv), yv(dataset.yv),
+        fileType(dataset.fileType)
     {
     }
     std::string Dataset::getName() const
     {
         return name;
     }
-    std::string Dataset::getClassName() const
-    {
-        return className;
-    }
     std::vector<std::string> Dataset::getFeatures() const
     {
         if (loaded) {
             return features;
         } else {
-            throw std::invalid_argument("Dataset not loaded.");
+            throw std::invalid_argument(message_dataset_not_loaded);
         }
     }
     int Dataset::getNFeatures() const
@@ -26,7 +28,7 @@ namespace platform {
         if (loaded) {
             return n_features;
         } else {
-            throw std::invalid_argument("Dataset not loaded.");
+            throw std::invalid_argument(message_dataset_not_loaded);
         }
     }
     int Dataset::getNSamples() const
@@ -34,7 +36,40 @@ namespace platform {
         if (loaded) {
             return n_samples;
         } else {
-            throw std::invalid_argument("Dataset not loaded.");
+            throw std::invalid_argument(message_dataset_not_loaded);
+        }
+    }
+    std::string Dataset::getClassName() const
+    {
+        return className;
+    }
+    int Dataset::getNClasses() const
+    {
+        if (loaded) {
+            return *std::max_element(yv.begin(), yv.end()) + 1;
+        } else {
+            throw std::invalid_argument(message_dataset_not_loaded);
+        }
+    }
+    std::vector<std::string> Dataset::getLabels() const
+    {
+        // Return the labels factorization result
+        if (loaded) {
+            return labels;
+        } else {
+            throw std::invalid_argument(message_dataset_not_loaded);
+        }
+    }
+    std::vector<int> Dataset::getClassesCounts() const
+    {
+        if (loaded) {
+            std::vector<int> counts(*std::max_element(yv.begin(), yv.end()) + 1);
+            for (auto y : yv) {
+                counts[y]++;
+            }
+            return counts;
+        } else {
+            throw std::invalid_argument(message_dataset_not_loaded);
         }
     }
     std::map<std::string, std::vector<int>> Dataset::getStates() const
@@ -42,7 +77,7 @@ namespace platform {
         if (loaded) {
             return states;
         } else {
-            throw std::invalid_argument("Dataset not loaded.");
+            throw std::invalid_argument(message_dataset_not_loaded);
         }
     }
     pair<std::vector<std::vector<float>>&, std::vector<int>&> Dataset::getVectors()
@@ -50,61 +85,56 @@ namespace platform {
         if (loaded) {
             return { Xv, yv };
         } else {
-            throw std::invalid_argument("Dataset not loaded.");
-        }
-    }
-    pair<std::vector<std::vector<int>>&, std::vector<int>&> Dataset::getVectorsDiscretized()
-    {
-        if (loaded) {
-            return { Xd, yv };
-        } else {
-            throw std::invalid_argument("Dataset not loaded.");
+            throw std::invalid_argument(message_dataset_not_loaded);
         }
     }
     pair<torch::Tensor&, torch::Tensor&> Dataset::getTensors()
     {
         if (loaded) {
-            buildTensors();
             return { X, y };
         } else {
-            throw std::invalid_argument("Dataset not loaded.");
+            throw std::invalid_argument(message_dataset_not_loaded);
         }
     }
     void Dataset::load_csv()
     {
         ifstream file(path + "/" + name + ".csv");
-        if (file.is_open()) {
-            std::string line;
-            getline(file, line);
-            std::vector<std::string> tokens = split(line, ',');
-            features = std::vector<std::string>(tokens.begin(), tokens.end() - 1);
-            if (className == "-1") {
-                className = tokens.back();
-            }
-            for (auto i = 0; i < features.size(); ++i) {
-                Xv.push_back(std::vector<float>());
-            }
-            while (getline(file, line)) {
-                tokens = split(line, ',');
-                for (auto i = 0; i < features.size(); ++i) {
-                    Xv[i].push_back(stof(tokens[i]));
-                }
-                yv.push_back(stoi(tokens.back()));
-            }
-            labels.clear();
-            file.close();
-        } else {
+        if (!file.is_open()) {
             throw std::invalid_argument("Unable to open dataset file.");
         }
+        labels.clear();
+        std::string line;
+        getline(file, line);
+        std::vector<std::string> tokens = split(line, ',');
+        features = std::vector<std::string>(tokens.begin(), tokens.end() - 1);
+        if (className == "-1") {
+            className = tokens.back();
+        }
+        for (auto i = 0; i < features.size(); ++i) {
+            Xv.push_back(std::vector<float>());
+        }
+        while (getline(file, line)) {
+            tokens = split(line, ',');
+            for (auto i = 0; i < features.size(); ++i) {
+                Xv[i].push_back(stof(tokens[i]));
+            }
+            auto label = trim(tokens.back());
+            if (find(labels.begin(), labels.end(), label) == labels.end()) {
+                labels.push_back(label);
+            }
+            yv.push_back(stoi(label));
+        }
+        file.close();
     }
     void Dataset::computeStates()
     {
         for (int i = 0; i < features.size(); ++i) {
-            states[features[i]] = std::vector<int>(*max_element(Xd[i].begin(), Xd[i].end()) + 1);
-            auto item = states.at(features[i]);
-            iota(begin(item), end(item), 0);
+            auto [max_value, idx] = torch::max(X_train.index({ i, "..." }), 0);
+            states[features[i]] = std::vector<int>(max_value.item<int>() + 1);
+            iota(begin(states.at(features[i])), end(states.at(features[i])), 0);
         }
-        states[className] = std::vector<int>(*max_element(yv.begin(), yv.end()) + 1);
+        auto [max_value, idx] = torch::max(y_train, 0);
+        states[className] = std::vector<int>(max_value.item<int>() + 1);
         iota(begin(states.at(className)), end(states.at(className)), 0);
     }
     void Dataset::load_arff()
@@ -141,32 +171,35 @@ namespace platform {
     void Dataset::load_rdata()
     {
         ifstream file(path + "/" + name + "_R.dat");
-        if (file.is_open()) {
-            std::string line;
-            getline(file, line);
-            line = ArffFiles::trim(line);
-            std::vector<std::string> tokens = tokenize(line);
-            transform(tokens.begin(), tokens.end() - 1, back_inserter(features), [](const auto& attribute) { return ArffFiles::trim(attribute); });
-            if (className == "-1") {
-                className = ArffFiles::trim(tokens.back());
-            }
-            for (auto i = 0; i < features.size(); ++i) {
-                Xv.push_back(std::vector<float>());
-            }
-            while (getline(file, line)) {
-                tokens = tokenize(line);
-                // We have to skip the first token, which is the instance number.
-                for (auto i = 1; i < features.size() + 1; ++i) {
-                    const float value = stof(tokens[i]);
-                    Xv[i - 1].push_back(value);
-                }
-                yv.push_back(stoi(tokens.back()));
-            }
-            labels.clear();
-            file.close();
-        } else {
+        if (!file.is_open()) {
             throw std::invalid_argument("Unable to open dataset file.");
         }
+        std::string line;
+        labels.clear();
+        getline(file, line);
+        line = ArffFiles::trim(line);
+        std::vector<std::string> tokens = tokenize(line);
+        transform(tokens.begin(), tokens.end() - 1, back_inserter(features), [](const auto& attribute) { return ArffFiles::trim(attribute); });
+        if (className == "-1") {
+            className = ArffFiles::trim(tokens.back());
+        }
+        for (auto i = 0; i < features.size(); ++i) {
+            Xv.push_back(std::vector<float>());
+        }
+        while (getline(file, line)) {
+            tokens = tokenize(line);
+            // We have to skip the first token, which is the instance number.
+            for (auto i = 1; i < features.size() + 1; ++i) {
+                const float value = stof(tokens[i]);
+                Xv[i - 1].push_back(value);
+            }
+            auto label = trim(tokens.back());
+            if (find(labels.begin(), labels.end(), label) == labels.end()) {
+                labels.push_back(label);
+            }
+            yv.push_back(stoi(label));
+        }
+        file.close();
     }
     void Dataset::load()
     {
@@ -180,39 +213,66 @@ namespace platform {
         } else if (fileType == RDATA) {
             load_rdata();
         }
-        if (discretize) {
-            Xd = discretizeDataset(Xv, yv);
-            computeStates();
-        }
         n_samples = Xv[0].size();
         n_features = Xv.size();
-        loaded = true;
-    }
-    void Dataset::buildTensors()
-    {
-        if (discretize) {
-            X = torch::zeros({ static_cast<int>(n_features), static_cast<int>(n_samples) }, torch::kInt32);
+        if (numericFeaturesIdx.size() == 0) {
+            numericFeatures = std::vector<bool>(n_features, false);
         } else {
-            X = torch::zeros({ static_cast<int>(n_features), static_cast<int>(n_samples) }, torch::kFloat32);
-        }
-        for (int i = 0; i < features.size(); ++i) {
-            if (discretize) {
-                X.index_put_({ i,  "..." }, torch::tensor(Xd[i], torch::kInt32));
+            if (numericFeaturesIdx.at(0) == -1) {
+                numericFeatures = std::vector<bool>(n_features, true);
             } else {
-                X.index_put_({ i,  "..." }, torch::tensor(Xv[i], torch::kFloat32));
+                numericFeatures = std::vector<bool>(n_features, false);
+                for (auto i : numericFeaturesIdx) {
+                    numericFeatures[i] = true;
+                }
             }
         }
-        y = torch::tensor(yv, torch::kInt32);
-    }
-    std::vector<mdlp::labels_t> Dataset::discretizeDataset(std::vector<mdlp::samples_t>& X, mdlp::labels_t& y)
-    {
-        std::vector<mdlp::labels_t> Xd;
-        auto fimdlp = mdlp::CPPFImdlp();
-        for (int i = 0; i < X.size(); i++) {
-            fimdlp.fit(X[i], y);
-            mdlp::labels_t& xd = fimdlp.transform(X[i]);
-            Xd.push_back(xd);
+        // Build Tensors
+        X = torch::zeros({ n_features, n_samples }, torch::kFloat32);
+        for (int i = 0; i < features.size(); ++i) {
+            X.index_put_({ i,  "..." }, torch::tensor(Xv[i], torch::kFloat32));
         }
-        return Xd;
+        y = torch::tensor(yv, torch::kInt32);
+        loaded = true;
+    }
+    std::tuple<torch::Tensor&, torch::Tensor&, torch::Tensor&, torch::Tensor&> Dataset::getTrainTestTensors(std::vector<int>& train, std::vector<int>& test)
+    {
+        if (!loaded) {
+            throw std::invalid_argument(message_dataset_not_loaded);
+        }
+        auto train_t = torch::tensor(train);
+        int samples_train = train.size();
+        int samples_test = test.size();
+        auto test_t = torch::tensor(test);
+        X_train = X.index({ "...", train_t });
+        y_train = y.index({ train_t });
+        X_test = X.index({ "...", test_t });
+        y_test = y.index({ test_t });
+        if (discretize) {
+            auto discretizer = Discretization::instance()->create(discretizer_algorithm);
+            auto X_train_d = torch::zeros({ n_features, samples_train }, torch::kInt32);
+            auto X_test_d = torch::zeros({ n_features, samples_test }, torch::kInt32);
+            for (auto feature = 0; feature < n_features; ++feature) {
+                if (numericFeatures[feature]) {
+                    auto feature_train = X_train.index({ feature, "..." });
+                    auto feature_test = X_test.index({ feature, "..." });
+                    auto feature_train_disc = discretizer->fit_transform_t(feature_train, y_train);
+                    auto feature_test_disc = discretizer->transform_t(feature_test);
+                    X_train_d.index_put_({ feature, "..." }, feature_train_disc);
+                    X_test_d.index_put_({ feature, "..." }, feature_test_disc);
+                } else {
+                    X_train_d.index_put_({ feature, "..." }, X_train.index({ feature, "..." }).to(torch::kInt32));
+                    X_test_d.index_put_({ feature, "..." }, X_test.index({ feature, "..." }).to(torch::kInt32));
+                }
+            }
+            X_train = X_train_d;
+            X_test = X_test_d;
+            assert(X_train.dtype() == torch::kInt32);
+            assert(X_test.dtype() == torch::kInt32);
+            computeStates();
+        }
+        assert(y_train.dtype() == torch::kInt32);
+        assert(y_test.dtype() == torch::kInt32);
+        return { X_train, X_test, y_train, y_test };
     }
 }

src/common/Dataset.h

@@ -4,25 +4,32 @@
 #include <map>
 #include <vector>
 #include <string>
-#include <CPPFImdlp.h>
+#include <tuple>
+#include <common/DiscretizationRegister.h>
 #include "Utils.h"
 #include "SourceData.h"
 namespace platform {
-
     class Dataset {
     public:
-        Dataset(const std::string& path, const std::string& name, const std::string& className, bool discretize, fileType_t fileType) : path(path), name(name), className(className), discretize(discretize), loaded(false), fileType(fileType) {};
+        Dataset(const std::string& path, const std::string& name, const std::string& className, bool discretize, fileType_t fileType, std::vector<int> numericFeaturesIdx, std::string discretizer_algo = "none") :
+            path(path), name(name), className(className), discretize(discretize),
+            loaded(false), fileType(fileType), numericFeaturesIdx(numericFeaturesIdx), discretizer_algorithm(discretizer_algo)
+        {
+        };
         explicit Dataset(const Dataset&);
         std::string getName() const;
         std::string getClassName() const;
-        std::vector<std::string> getLabels() const { return labels; }
+        int getNClasses() const;
+        std::vector<std::string> getLabels() const; // return the labels factorization result
+        std::vector<int> getClassesCounts() const;
         std::vector<string> getFeatures() const;
         std::map<std::string, std::vector<int>> getStates() const;
         std::pair<vector<std::vector<float>>&, std::vector<int>&> getVectors();
-        std::pair<vector<std::vector<int>>&, std::vector<int>&> getVectorsDiscretized();
         std::pair<torch::Tensor&, torch::Tensor&> getTensors();
+        std::tuple<torch::Tensor&, torch::Tensor&, torch::Tensor&, torch::Tensor&> getTrainTestTensors(std::vector<int>& train, std::vector<int>& test);
         int getNFeatures() const;
         int getNSamples() const;
+        std::vector<bool>& getNumericFeatures() { return numericFeatures; }
         void load();
         const bool inline isLoaded() const { return loaded; };
     private:
@@ -31,16 +38,18 @@ namespace platform {
         fileType_t fileType;
         std::string className;
         int n_samples{ 0 }, n_features{ 0 };
+        std::vector<int> numericFeaturesIdx;
+        std::string discretizer_algorithm;
+        std::vector<bool> numericFeatures; // true if feature is numeric
         std::vector<std::string> features;
         std::vector<std::string> labels;
         std::map<std::string, std::vector<int>> states;
         bool loaded;
         bool discretize;
         torch::Tensor X, y;
+        torch::Tensor X_train, X_test, y_train, y_test;
         std::vector<std::vector<float>> Xv;
-        std::vector<std::vector<int>> Xd;
         std::vector<int> yv;
-        void buildTensors();
         void load_csv();
         void load_arff();
         void load_rdata();

src/common/Datasets.cpp

@@ -1,32 +1,70 @@
 #include <fstream>
 #include "Datasets.h"
+#include <nlohmann/json.hpp>
+
 namespace platform {
+    using json = nlohmann::ordered_json;
+    const std::string message_dataset_not_loaded = "dataset not loaded.";
+    Datasets::Datasets(bool discretize, std::string sfileType, std::string discretizer_algorithm) :
+        discretize(discretize), sfileType(sfileType), discretizer_algorithm(discretizer_algorithm)
+    {
+        if ((discretizer_algorithm == "none" || discretizer_algorithm == "") && discretize) {
+            throw std::runtime_error("Can't discretize without discretization algorithm");
+        }
+        load();
+    }
     void Datasets::load()
     {
         auto sd = SourceData(sfileType);
         fileType = sd.getFileType();
         path = sd.getPath();
         ifstream catalog(path + "all.txt");
-        if (catalog.is_open()) {
-            std::string line;
-            while (getline(catalog, line)) {
-                if (line.empty() || line[0] == '#') {
-                    continue;
-                }
-                std::vector<std::string> tokens = split(line, ',');
-                std::string name = tokens[0];
-                std::string className;
-                if (tokens.size() == 1) {
-                    className = "-1";
-                } else {
-                    className = tokens[1];
-                }
-                datasets[name] = make_unique<Dataset>(path, name, className, discretize, fileType);
-            }
-            catalog.close();
-        } else {
+        std::vector<int> numericFeaturesIdx;
+        if (!catalog.is_open()) {
             throw std::invalid_argument("Unable to open catalog file. [" + path + "all.txt" + "]");
         }
+        std::string line;
+        while (getline(catalog, line)) {
+            if (line.empty() || line[0] == '#') {
+                continue;
+            }
+            std::vector<std::string> tokens = split(line, ';');
+            std::string name = tokens[0];
+            std::string className;
+            numericFeaturesIdx.clear();
+            int size = tokens.size();
+            switch (size) {
+                case 1:
+                    className = "-1";
+                    numericFeaturesIdx.push_back(-1);
+                    break;
+                case 2:
+                    className = tokens[1];
+                    numericFeaturesIdx.push_back(-1);
+                    break;
+                case 3:
+                    {
+                        className = tokens[1];
+                        auto numericFeatures = tokens[2];
+                        if (numericFeatures == "all") {
+                            numericFeaturesIdx.push_back(-1);
+                        } else {
+                            if (numericFeatures != "none") {
+                                auto features = json::parse(numericFeatures);
+                                for (auto& f : features) {
+                                    numericFeaturesIdx.push_back(f);
+                                }
+                            }
+                        }
+                    }
+                    break;
+                default:
+                    throw std::invalid_argument("Invalid catalog file format.");
+
+            }
+            datasets[name] = make_unique<Dataset>(path, name, className, discretize, fileType, numericFeaturesIdx, discretizer_algorithm);
+        }
+        catalog.close();
     }
     std::vector<std::string> Datasets::getNames()
     {
@@ -34,102 +72,6 @@ namespace platform {
         transform(datasets.begin(), datasets.end(), back_inserter(result), [](const auto& d) { return d.first; });
         return result;
     }
-    std::vector<std::string> Datasets::getFeatures(const std::string& name) const
-    {
-        if (datasets.at(name)->isLoaded()) {
-            return datasets.at(name)->getFeatures();
-        } else {
-            throw std::invalid_argument("Dataset not loaded.");
-        }
-    }
-    std::vector<std::string> Datasets::getLabels(const std::string& name) const
-    {
-        if (datasets.at(name)->isLoaded()) {
-            return datasets.at(name)->getLabels();
-        } else {
-            throw std::invalid_argument("Dataset not loaded.");
-        }
-    }
-    map<std::string, std::vector<int>> Datasets::getStates(const std::string& name) const
-    {
-        if (datasets.at(name)->isLoaded()) {
-            return datasets.at(name)->getStates();
-        } else {
-            throw std::invalid_argument("Dataset not loaded.");
-        }
-    }
-    void Datasets::loadDataset(const std::string& name) const
-    {
-        if (datasets.at(name)->isLoaded()) {
-            return;
-        } else {
-            datasets.at(name)->load();
-        }
-    }
-    std::string Datasets::getClassName(const std::string& name) const
-    {
-        if (datasets.at(name)->isLoaded()) {
-            return datasets.at(name)->getClassName();
-        } else {
-            throw std::invalid_argument("Dataset not loaded.");
-        }
-    }
-    int Datasets::getNSamples(const std::string& name) const
-    {
-        if (datasets.at(name)->isLoaded()) {
-            return datasets.at(name)->getNSamples();
-        } else {
-            throw std::invalid_argument("Dataset not loaded.");
-        }
-    }
-    int Datasets::getNClasses(const std::string& name)
-    {
-        if (datasets.at(name)->isLoaded()) {
-            auto className = datasets.at(name)->getClassName();
-            if (discretize) {
-                auto states = getStates(name);
-                return states.at(className).size();
-            }
-            auto [Xv, yv] = getVectors(name);
-            return *std::max_element(yv.begin(), yv.end()) + 1;
-        } else {
-            throw std::invalid_argument("Dataset not loaded.");
-        }
-    }
-    std::vector<int> Datasets::getClassesCounts(const std::string& name) const
-    {
-        if (datasets.at(name)->isLoaded()) {
-            auto [Xv, yv] = datasets.at(name)->getVectors();
-            std::vector<int> counts(*std::max_element(yv.begin(), yv.end()) + 1);
-            for (auto y : yv) {
-                counts[y]++;
-            }
-            return counts;
-        } else {
-            throw std::invalid_argument("Dataset not loaded.");
-        }
-    }
-    pair<std::vector<std::vector<float>>&, std::vector<int>&> Datasets::getVectors(const std::string& name)
-    {
-        if (!datasets[name]->isLoaded()) {
-            datasets[name]->load();
-        }
-        return datasets[name]->getVectors();
-    }
-    pair<std::vector<std::vector<int>>&, std::vector<int>&> Datasets::getVectorsDiscretized(const std::string& name)
-    {
-        if (!datasets[name]->isLoaded()) {
-            datasets[name]->load();
-        }
-        return datasets[name]->getVectorsDiscretized();
-    }
-    pair<torch::Tensor&, torch::Tensor&> Datasets::getTensors(const std::string& name)
-    {
-        if (!datasets[name]->isLoaded()) {
-            datasets[name]->load();
-        }
-        return datasets[name]->getTensors();
-    }
     bool Datasets::isDataset(const std::string& name) const
     {
         return datasets.find(name) != datasets.end();

src/common/Datasets.h

@@ -4,25 +4,16 @@
 namespace platform {
     class Datasets {
     public:
-        explicit Datasets(bool discretize, std::string sfileType) : discretize(discretize), sfileType(sfileType) { load(); };
+        explicit Datasets(bool discretize, std::string sfileType, std::string discretizer_algorithm = "none");
         std::vector<std::string> getNames();
-        std::vector<std::string> getFeatures(const std::string& name) const;
-        int getNSamples(const std::string& name) const;
-        std::vector<std::string> getLabels(const std::string& name) const;
-        std::string getClassName(const std::string& name) const;
-        int getNClasses(const std::string& name);
-        std::vector<int> getClassesCounts(const std::string& name) const;
-        std::map<std::string, std::vector<int>> getStates(const std::string& name) const;
-        std::pair<std::vector<std::vector<float>>&, std::vector<int>&> getVectors(const std::string& name);
-        std::pair<std::vector<std::vector<int>>&, std::vector<int>&> getVectorsDiscretized(const std::string& name);
-        std::pair<torch::Tensor&, torch::Tensor&> getTensors(const std::string& name);
         bool isDataset(const std::string& name) const;
-        void loadDataset(const std::string& name) const;
+        Dataset& getDataset(const std::string& name) const { return *datasets.at(name); }
         std::string toString() const;
     private:
         std::string path;
         fileType_t fileType;
         std::string sfileType;
+        std::string discretizer_algorithm;
         std::map<std::string, std::unique_ptr<Dataset>> datasets;
         bool discretize;
         void load(); // Loads the list of datasets

src/common/Discretization.cpp

@@ -0,0 +1,55 @@
+#include "Discretization.h"
+
+namespace platform {
+    // Idea from: https://www.codeproject.com/Articles/567242/AplusC-2b-2bplusObjectplusFactory
+    Discretization* Discretization::factory = nullptr;
+    Discretization* Discretization::instance()
+    {
+        //manages singleton
+        if (factory == nullptr)
+            factory = new Discretization();
+        return factory;
+    }
+    void Discretization::registerFactoryFunction(const std::string& name,
+        function<mdlp::Discretizer* (void)> classFactoryFunction)
+    {
+        // register the class factory function
+        functionRegistry[name] = classFactoryFunction;
+    }
+    std::shared_ptr<mdlp::Discretizer> Discretization::create(const std::string& name)
+    {
+        mdlp::Discretizer* instance = nullptr;
+
+        // find name in the registry and call factory method.
+        auto it = functionRegistry.find(name);
+        if (it != functionRegistry.end())
+            instance = it->second();
+        // wrap instance in a shared ptr and return
+        if (instance != nullptr)
+            return std::unique_ptr<mdlp::Discretizer>(instance);
+        else
+            throw std::runtime_error("Discretizer not found: " + name);
+    }
+    std::vector<std::string> Discretization::getNames()
+    {
+        std::vector<std::string> names;
+        transform(functionRegistry.begin(), functionRegistry.end(), back_inserter(names),
+            [](const pair<std::string, function<mdlp::Discretizer* (void)>>& pair) { return pair.first; });
+        return names;
+    }
+    std::string Discretization::toString()
+    {
+        std::string result = "";
+        std::string sep = "";
+        for (const auto& pair : functionRegistry) {
+            result += sep + pair.first;
+            sep = ", ";
+        }
+        return "{" + result + "}";
+    }
+    RegistrarDiscretization::RegistrarDiscretization(const std::string& name, function<mdlp::Discretizer* (void)> classFactoryFunction)
+    {
+        // register the class factory function 
+        Discretization::instance()->registerFactoryFunction(name, classFactoryFunction);
+    }
+}

src/common/Discretization.h

@@ -0,0 +1,33 @@
+#ifndef DISCRETIZATION_H
+#define DISCRETIZATION_H
+#include <map>
+#include <memory>
+#include <string>
+#include <functional>
+#include <vector>
+#include <fimdlp/Discretizer.h>
+#include <fimdlp/BinDisc.h>
+#include <fimdlp/CPPFImdlp.h>
+namespace platform {
+    class Discretization {
+    public:
+        Discretization(Discretization&) = delete;
+        void operator=(const Discretization&) = delete;
+        // Idea from: https://www.codeproject.com/Articles/567242/AplusC-2b-2bplusObjectplusFactory
+        static Discretization* instance();
+        std::shared_ptr<mdlp::Discretizer> create(const std::string& name);
+        void registerFactoryFunction(const std::string& name,
+            function<mdlp::Discretizer* (void)> classFactoryFunction);
+        std::vector<string> getNames();
+        std::string toString();
+    private:
+        map<std::string, function<mdlp::Discretizer* (void)>> functionRegistry;
+        static Discretization* factory; //singleton
+        Discretization() {};
+    };
+    class RegistrarDiscretization {
+    public:
+        RegistrarDiscretization(const std::string& className, function<mdlp::Discretizer* (void)> classFactoryFunction);
+    };
+}
+#endif

src/common/DiscretizationRegister.h

@@ -0,0 +1,38 @@
+#ifndef DISCRETIZATIONREGISTER_H
+#define DISCRETIZATIONREGISTER_H
+#include <common/Discretization.h>
+static platform::RegistrarDiscretization registrarM("mdlp",
+    [](void) -> mdlp::Discretizer* { return new mdlp::CPPFImdlp();});
+static platform::RegistrarDiscretization registrarBU3("bin3u",
+    [](void) -> mdlp::Discretizer* { return new mdlp::BinDisc(3, mdlp::strategy_t::UNIFORM);});
+static platform::RegistrarDiscretization registrarBQ3("bin3q",
+    [](void) -> mdlp::Discretizer* { return new mdlp::BinDisc(3, mdlp::strategy_t::QUANTILE);});
+static platform::RegistrarDiscretization registrarBU4("bin4u",
+    [](void) -> mdlp::Discretizer* { return new mdlp::BinDisc(4, mdlp::strategy_t::UNIFORM);});
+static platform::RegistrarDiscretization registrarBQ4("bin4q",
+    [](void) -> mdlp::Discretizer* { return new mdlp::BinDisc(4, mdlp::strategy_t::QUANTILE);});
+static platform::RegistrarDiscretization registrarBU5("bin5u",
+    [](void) -> mdlp::Discretizer* { return new mdlp::BinDisc(5, mdlp::strategy_t::UNIFORM);});
+static platform::RegistrarDiscretization registrarBQ5("bin5q",
+    [](void) -> mdlp::Discretizer* { return new mdlp::BinDisc(5, mdlp::strategy_t::QUANTILE);});
+static platform::RegistrarDiscretization registrarBU6("bin6u",
+    [](void) -> mdlp::Discretizer* { return new mdlp::BinDisc(6, mdlp::strategy_t::UNIFORM);});
+static platform::RegistrarDiscretization registrarBQ6("bin6q",
+    [](void) -> mdlp::Discretizer* { return new mdlp::BinDisc(6, mdlp::strategy_t::QUANTILE);});
+static platform::RegistrarDiscretization registrarBU7("bin7u",
+    [](void) -> mdlp::Discretizer* { return new mdlp::BinDisc(7, mdlp::strategy_t::UNIFORM);});
+static platform::RegistrarDiscretization registrarBQ7("bin7q",
+    [](void) -> mdlp::Discretizer* { return new mdlp::BinDisc(7, mdlp::strategy_t::QUANTILE);});
+static platform::RegistrarDiscretization registrarBU8("bin8u",
+    [](void) -> mdlp::Discretizer* { return new mdlp::BinDisc(8, mdlp::strategy_t::UNIFORM);});
+static platform::RegistrarDiscretization registrarBQ8("bin8q",
+    [](void) -> mdlp::Discretizer* { return new mdlp::BinDisc(8, mdlp::strategy_t::QUANTILE);});
+static platform::RegistrarDiscretization registrarBU9("bin9u",
+    [](void) -> mdlp::Discretizer* { return new mdlp::BinDisc(9, mdlp::strategy_t::UNIFORM);});
+static platform::RegistrarDiscretization registrarBQ9("bin9q",
+    [](void) -> mdlp::Discretizer* { return new mdlp::BinDisc(9, mdlp::strategy_t::QUANTILE);});
+static platform::RegistrarDiscretization registrarBU10("bin10u",
+    [](void) -> mdlp::Discretizer* { return new mdlp::BinDisc(10, mdlp::strategy_t::UNIFORM);});
+static platform::RegistrarDiscretization registrarBQ10("bin10q",
+    [](void) -> mdlp::Discretizer* { return new mdlp::BinDisc(10, mdlp::strategy_t::QUANTILE);});
+#endif

src/common/DotEnv.h

@@ -13,14 +13,53 @@ namespace platform {
     class DotEnv {
     private:
         std::map<std::string, std::string> env;
+        std::map<std::string, std::vector<std::string>> valid;
     public:
         DotEnv(bool create = false)
         {
+            valid =
+            {
+                {"depth", {"any"}},
+                {"discretize", {"0", "1"}},
+                {"discretize_algo", {"mdlp", "bin3u", "bin3q", "bin4u", "bin4q", "bin5q", "bin5u", "bin6q", "bin6u", "bin7q", "bin7u", "bin8q", "bin8u", "bin9q", "bin9u", "bin10q", "bin10u"}},
+                {"experiment", {"discretiz", "odte", "covid", "Test"}},
+                {"fit_features", {"0", "1"}},
+                {"framework", {"bulma", "bootstrap"}},
+                {"ignore_nan", {"0", "1"}},
+                {"leaves", {"any"}},
+                {"margin", {"0.1", "0.2", "0.3"}},
+                {"model", {"any"}},
+                {"n_folds", {"5", "10"}},
+                {"nodes", {"any"}},
+                {"platform", {"any"}},
+                {"stratified", {"0", "1"}},
+                {"score", {"accuracy", "roc-auc-ovr"}},
+                {"seeds", {"any"}},
+                {"smooth_strat", {"ORIGINAL", "LAPLACE", "CESTNIK"}},
+                {"source_data", {"Arff", "Tanveer", "Surcov", "Test"}},
+            };
             if (create) {
                 // For testing purposes
                 std::ofstream file(".env");
-                file << "source_data = Test" << std::endl;
-                file << "margin = 0.1" << std::endl;
+                file << "experiment=Test" << std::endl;
+                file << "source_data=Test" << std::endl;
+                file << "margin=0.1" << std::endl;
+                file << "score=accuracy" << std::endl;
+                file << "platform=um790Linux" << std::endl;
+                file << "n_folds=5" << std::endl;
+                file << "discretize_algo=mdlp" << std::endl;
+                file << "smooth_strat=ORIGINAL" << std::endl;
+                file << "stratified=0" << std::endl;
+                file << "model=TAN" << std::endl;
+                file << "seeds=[271]" << std::endl;
+                file << "discretize=0" << std::endl;
+                file << "ignore_nan=0" << std::endl;
+                file << "nodes=Nodes" << std::endl;
+                file << "leaves=Edges" << std::endl;
+                file << "depth=States" << std::endl;
+                file << "fit_features=0" << std::endl;
+                file << "framework=bulma" << std::endl;
+                file << "margin=0.1" << std::endl;
                 file.close();
             }
             std::ifstream file(".env");
@@ -37,12 +76,62 @@ namespace platform {
                 std::istringstream iss(line);
                 std::string key, value;
                 if (std::getline(iss, key, '=') && std::getline(iss, value)) {
-                    env[trim(key)] = trim(value);
+                    key = trim(key);
+                    value = trim(value);
+                    parse(key, value);
+                    env[key] = value;
+                }
+            }
+            parseEnv();
+        }
+        void parse(const std::string& key, const std::string& value)
+        {
+            if (valid.find(key) == valid.end()) {
+                std::cerr << "Invalid key in .env: " << key << std::endl;
+                exit(1);
+            }
+            if (valid[key].front() == "any") {
+                return;
+            }
+            if (std::find(valid[key].begin(), valid[key].end(), value) == valid[key].end()) {
+                std::cerr << "Invalid value in .env: " << key << " = " << value << std::endl;
+                exit(1);
+            }
+        }
+        std::vector<std::string> valid_tokens(const std::string& key)
+        {
+            if (valid.find(key) == valid.end()) {
+                return {};
+            }
+            return valid.at(key);
+        }
+        std::string valid_values(const std::string& key)
+        {
+            std::string valid_values = "{", sep = "";
+            if (valid.find(key) == valid.end()) {
+                return "{}";
+            }
+            for (const auto& value : valid.at(key)) {
+                valid_values += sep + value;
+                sep = ", ";
+            }
+            return valid_values + "}";
+        }
+        void parseEnv()
+        {
+            for (auto& [key, values] : valid) {
+                if (env.find(key) == env.end()) {
+                    std::cerr << "Key not found in .env: " << key << ", valid values: " << valid_values(key) << std::endl;
+                    exit(1);
                 }
             }
         }
         std::string get(const std::string& key)
         {
+            if (env.find(key) == env.end()) {
+                std::cerr << "Key not found in .env: " << key << std::endl;
+                exit(1);
+            }
             return env.at(key);
         }
         std::vector<int> getSeeds()

src/common/Paths.h

@@ -6,15 +6,30 @@
 namespace platform {
     class Paths {
     public:
-        static std::string results() { return "results/"; }
-        static std::string hiddenResults() { return "hidden_results/"; }
-        static std::string excel() { return "excel/"; }
-        static std::string grid() { return "grid/"; }
+        static std::string createIfNotExists(const std::string& folder)
+        {
+            if (!std::filesystem::exists(folder)) {
+                std::filesystem::create_directory(folder);
+            }
+            return folder;
+        }
+        static std::string results() { return createIfNotExists("results/"); }
+        static std::string hiddenResults() { return createIfNotExists("hidden_results/"); }
+        static std::string excel() { return createIfNotExists("excel/"); }
+        static std::string grid() { return createIfNotExists("grid/"); }
+        static std::string graphs() { return createIfNotExists("graphs/"); }
+        static std::string tex() { return createIfNotExists("tex/"); }
         static std::string datasets()
         {
             auto env = platform::DotEnv();
             return env.get("source_data");
         }
+        static std::string experiment_file(const std::string& fileName, bool discretize, bool stratified, int seed, int nfold)
+        {
+            std::string disc = discretize ? "_disc_" : "_ndisc_";
+            std::string strat = stratified ? "strat_" : "nstrat_";
+            return "datasets_experiment/" + fileName + disc + strat + std::to_string(seed) + "_" + std::to_string(nfold) + ".json";
+        }
         static void createPath(const std::string& path)
         {
             // Create directory if it does not exist
@@ -25,6 +40,14 @@ namespace platform {
                 throw std::runtime_error("Could not create directory " + path);
             }
         }
+        static std::string bestResultsFile(const std::string& score, const std::string& model)
+        {
+            return "best_results_" + score + "_" + model + ".json";
+        }
+        static std::string bestResultsExcel(const std::string& score)
+        {
+            return "BestResults_" + score + ".xlsx";
+        }
         static std::string excelResults() { return "some_results.xlsx"; }
         static std::string grid_input(const std::string& model)
         {
@@ -34,6 +57,22 @@ namespace platform {
         {
             return grid() + "grid_" + model + "_output.json";
         }
+        static std::string tex_output()
+        {
+            return "results.tex";
+        }
+        static std::string md_output()
+        {
+            return "results.md";
+        }
+        static std::string tex_post_hoc()
+        {
+            return "post_hoc.tex";
+        }
+        static std::string md_post_hoc()
+        {
+            return "post_hoc.md";
+        }
     };
 }
 #endif

src/common/Symbols.h

@@ -12,6 +12,7 @@ namespace platform {
         inline static const std::string downward_arrow{ "\u27B4" };
         inline static const std::string up_arrow{ "\u2B06" };
         inline static const std::string down_arrow{ "\u2B07" };
+        inline static const std::string ellipsis{ "\u2026" };
         inline static const std::string equal_best{ check_mark };
         inline static const std::string better_best{ black_star };
         inline static const std::string notebook{ "\U0001F5C8" };

src/common/Utils.h

@@ -3,15 +3,16 @@
 #include <sstream>
 #include <string>
 #include <vector>
+#include <algorithm>
+#include <torch/torch.h>
 namespace platform {
-    static std::vector<std::string> split(const std::string& text, char delimiter)
+    template <typename T>
+    std::vector<T> tensorToVector(const torch::Tensor& tensor)
     {
-        std::vector<std::string> result;
-        std::stringstream ss(text);
-        std::string token;
-        while (std::getline(ss, token, delimiter)) {
-            result.push_back(token);
-        }
+        torch::Tensor contig_tensor = tensor.contiguous();
+        auto num_elements = contig_tensor.numel();
+        const T* tensor_data = contig_tensor.data_ptr<T>();
+        std::vector<T> result(tensor_data, tensor_data + num_elements);
         return result;
     }
     static std::string trim(const std::string& str)
@@ -25,5 +26,45 @@ namespace platform {
             }).base(), result.end());
         return result;
     }
+    static std::vector<std::string> split(const std::string& text, char delimiter)
+    {
+        std::vector<std::string> result;
+        std::stringstream ss(text);
+        std::string token;
+        while (std::getline(ss, token, delimiter)) {
+            result.push_back(trim(token));
+        }
+        return result;
+    }
+    inline double compute_std(std::vector<double> values, double mean)
+    {
+        // Compute standard devation of the values
+        double sum = 0.0;
+        for (const auto& value : values) {
+            sum += std::pow(value - mean, 2);
+        }
+        double variance = sum / values.size();
+        return std::sqrt(variance);
+    }
+    inline std::string get_date()
+    {
+        time_t rawtime;
+        tm* timeinfo;
+        time(&rawtime);
+        timeinfo = std::localtime(&rawtime);
+        std::ostringstream oss;
+        oss << std::put_time(timeinfo, "%Y-%m-%d");
+        return oss.str();
+    }
+    inline std::string get_time()
+    {
+        time_t rawtime;
+        tm* timeinfo;
+        time(&rawtime);
+        timeinfo = std::localtime(&rawtime);
+        std::ostringstream oss;
+        oss << std::put_time(timeinfo, "%H:%M:%S");
+        return oss.str();
+    }
 }
 #endif

src/grid/GridSearch.cpp

@@ -5,29 +5,11 @@
 #include "main/Models.h"
 #include "common/Paths.h"
 #include "common/Colors.h"
+#include "common/Utils.h"
 #include "GridSearch.h"
 
 namespace platform {
-    std::string get_date()
-    {
-        time_t rawtime;
-        tm* timeinfo;
-        time(&rawtime);
-        timeinfo = std::localtime(&rawtime);
-        std::ostringstream oss;
-        oss << std::put_time(timeinfo, "%Y-%m-%d");
-        return oss.str();
-    }
-    std::string get_time()
-    {
-        time_t rawtime;
-        tm* timeinfo;
-        time(&rawtime);
-        timeinfo = std::localtime(&rawtime);
-        std::ostringstream oss;
-        oss << std::put_time(timeinfo, "%H:%M:%S");
-        return oss.str();
-    }
+
     std::string get_color_rank(int rank)
     {
         auto colors = { Colors::WHITE(), Colors::RED(), Colors::GREEN(),  Colors::BLUE(), Colors::MAGENTA(), Colors::CYAN() };
@@ -103,11 +85,11 @@ namespace platform {
         std::mt19937 g{ 271 }; // Use fixed seed to obtain the same shuffle
         std::shuffle(tasks.begin(), tasks.end(), g);
         std::cout << get_color_rank(rank) << "* Number of tasks: " << tasks.size() << std::endl;
-        std::cout << "|";
+        std::cout << separator;
         for (int i = 0; i < tasks.size(); ++i) {
             std::cout << (i + 1) % 10;
         }
-        std::cout << "|" << std::endl << "|" << std::flush;
+        std::cout << separator << std::endl << separator << std::flush;
         return tasks;
     }
     void process_task_mpi_consumer(struct ConfigGrid& config, struct ConfigMPI& config_mpi, json& tasks, int n_task, Datasets& datasets, Task_Result* result)
@@ -118,17 +100,18 @@ namespace platform {
         json task = tasks[n_task];
         auto model = config.model;
         auto grid = GridData(Paths::grid_input(model));
-        auto dataset = task["dataset"].get<std::string>();
+        auto dataset_name = task["dataset"].get<std::string>();
         auto idx_dataset = task["idx_dataset"].get<int>();
         auto seed = task["seed"].get<int>();
         auto n_fold = task["fold"].get<int>();
         bool stratified = config.stratified;
         // Generate the hyperparamters combinations
-        auto combinations = grid.getGrid(dataset);
-        auto [X, y] = datasets.getTensors(dataset);
-        auto states = datasets.getStates(dataset);
-        auto features = datasets.getFeatures(dataset);
-        auto className = datasets.getClassName(dataset);
+        auto& dataset = datasets.getDataset(dataset_name);
+        auto combinations = grid.getGrid(dataset_name);
+        dataset.load();
+        auto [X, y] = dataset.getTensors();
+        auto features = dataset.getFeatures();
+        auto className = dataset.getClassName();
         //
         // Start working on task
         //
@@ -138,14 +121,11 @@ namespace platform {
         else
             fold = new folding::KFold(config.n_folds, y.size(0), seed);
         auto [train, test] = fold->getFold(n_fold);
-        auto train_t = torch::tensor(train);
-        auto test_t = torch::tensor(test);
-        auto X_train = X.index({ "...", train_t });
-        auto y_train = y.index({ train_t });
-        auto X_test = X.index({ "...", test_t });
-        auto y_test = y.index({ test_t });
+        auto [X_train, X_test, y_train, y_test] = dataset.getTrainTestTensors(train, test);
+        auto states = dataset.getStates(); // Get the states of the features Once they are discretized
         double best_fold_score = 0.0;
         int best_idx_combination = -1;
+        bayesnet::Smoothing_t smoothing = bayesnet::Smoothing_t::NONE;
         json best_fold_hyper;
         for (int idx_combination = 0; idx_combination < combinations.size(); ++idx_combination) {
             auto hyperparam_line = combinations[idx_combination];
@@ -168,10 +148,10 @@ namespace platform {
                 // Build Classifier with selected hyperparameters
                 auto clf = Models::instance()->create(config.model);
                 auto valid = clf->getValidHyperparameters();
-                hyperparameters.check(valid, dataset);
-                clf->setHyperparameters(hyperparameters.get(dataset));
+                hyperparameters.check(valid, dataset_name);
+                clf->setHyperparameters(hyperparameters.get(dataset_name));
                 // Train model
-                clf->fit(X_nested_train, y_nested_train, features, className, states);
+                clf->fit(X_nested_train, y_nested_train, features, className, states, smoothing);
                 // Test model
                 score += clf->score(X_nested_test, y_nested_test);
             }
@@ -188,9 +168,9 @@ namespace platform {
         auto hyperparameters = platform::HyperParameters(datasets.getNames(), best_fold_hyper);
         auto clf = Models::instance()->create(config.model);
         auto valid = clf->getValidHyperparameters();
-        hyperparameters.check(valid, dataset);
+        hyperparameters.check(valid, dataset_name);
         clf->setHyperparameters(best_fold_hyper);
-        clf->fit(X_train, y_train, features, className, states);
+        clf->fit(X_train, y_train, features, className, states, smoothing);
         best_fold_score = clf->score(X_test, y_test);
         // Return the result
         result->idx_dataset = task["idx_dataset"].get<int>();
@@ -373,14 +353,16 @@ namespace platform {
         MPI_Bcast(msg, tasks_size + 1, MPI_CHAR, config_mpi.manager, MPI_COMM_WORLD);
         tasks = json::parse(msg);
         delete[] msg;
-        auto datasets = Datasets(config.discretize, Paths::datasets());
+        auto env = platform::DotEnv();
+        auto datasets = Datasets(config.discretize, Paths::datasets(), env.get("discretize_algo"));
+
         if (config_mpi.rank == config_mpi.manager) {
             //
             // 2a. Producer delivers the tasks to the consumers
             //
             auto datasets_names = filterDatasets(datasets);
             json all_results = producer(datasets_names, tasks, config_mpi, MPI_Result);
-            std::cout << get_color_rank(config_mpi.rank) << "|" << std::endl;
+            std::cout << get_color_rank(config_mpi.rank) << separator << std::endl;
             //
             // 3. Manager select the bests sccores for each dataset
             //

src/grid/GridSearch.h

@@ -55,6 +55,7 @@ namespace platform {
         struct ConfigGrid config;
         json build_tasks_mpi(int rank);
         Timer timer; // used to measure the time of the whole process
+        const std::string separator = "|";
     };
 } /* namespace platform */
 #endif

src/main/Experiment.cpp

@@ -10,6 +10,7 @@ namespace platform {
     void Experiment::saveResult()
     {
         result.save();
+        std::cout << "Result saved in " << Paths::results() << result.getFilename() << std::endl;
     }
     void Experiment::report(bool classification_report)
     {
@@ -23,7 +24,24 @@ namespace platform {
     {
         std::cout << result.getJson().dump(4) << std::endl;
     }
-    void Experiment::go(std::vector<std::string> filesToProcess, bool quiet, bool no_train_score)
+    void Experiment::saveGraph()
+    {
+        std::cout << "Saving graphs..." << std::endl;
+        auto data = result.getJson();
+        for (const auto& item : data["results"]) {
+            auto graphs = item["graph"];
+            int i = 0;
+            for (const auto& graph : graphs) {
+                i++;
+                auto fileName = Paths::graphs() + result.getFilename() + "_graph_" + item["dataset"].get<std::string>() + "_" + std::to_string(i) + ".dot";
+                auto file = std::ofstream(fileName);
+                file << graph.get<std::string>();
+                file.close();
+                std::cout << "Graph saved in " << fileName << std::endl;
+            }
+        }
+    }
+    void Experiment::go(std::vector<std::string> filesToProcess, bool quiet, bool no_train_score, bool generate_fold_files, bool graph)
     {
         for (auto fileName : filesToProcess) {
             if (fileName.size() > max_name)
@@ -40,14 +58,14 @@ namespace platform {
             std::cout << " ( " << Colors::GREEN() << "b" << Colors::RESET() << " )  Scoring train dataset" << std::endl;
             std::cout << " ( " << Colors::GREEN() << "c" << Colors::RESET() << " )  Scoring test dataset" << std::endl << std::endl;
             std::cout << Colors::YELLOW() << "Note: fold number in this color means fitting had issues such as not using all features in BoostAODE classifier" << std::endl << std::endl;
-            std::cout << Colors::GREEN() << left << "  #  " << setw(max_name) << "Dataset" << " #Samp #Feat Seed Status" << std::endl;
-            std::cout << " --- " << string(max_name, '-') << " ----- ----- ---- " << string(4 + 3 * nfolds, '-') << Colors::RESET() << std::endl;
+            std::cout << Colors::GREEN() << left << "  #  " << setw(max_name) << "Dataset" << " #Samp #Feat Seed Status" << string(3 * nfolds - 2, ' ') << " Time" << std::endl;
+            std::cout << " --- " << string(max_name, '-') << " ----- ----- ---- " << string(4 + 3 * nfolds, '-') << " ----------" << Colors::RESET() << std::endl;
         }
         int num = 0;
         for (auto fileName : filesToProcess) {
             if (!quiet)
                 std::cout << " " << setw(3) << right << num++ << " " << setw(max_name) << left << fileName << right << flush;
-            cross_validation(fileName, quiet, no_train_score);
+            cross_validation(fileName, quiet, no_train_score, generate_fold_files, graph);
             if (!quiet)
                 std::cout << std::endl;
         }
@@ -67,36 +85,88 @@ namespace platform {
                 return Colors::RESET();
         }
     }
-
+    score_t Experiment::parse_score() const
+    {
+        if (result.getScoreName() == "accuracy")
+            return score_t::ACCURACY;
+        if (result.getScoreName() == "roc-auc-ovr")
+            return score_t::ROC_AUC_OVR;
+        throw std::runtime_error("Unknown score: " + result.getScoreName());
+    }
     void showProgress(int fold, const std::string& color, const std::string& phase)
     {
-        std::string prefix = phase == "a" ? "" : "\b\b\b\b";
+        std::string prefix = phase == "-" ? "" : "\b\b\b\b";
         std::cout << prefix << color << fold << Colors::RESET() << "(" << color << phase << Colors::RESET() << ")" << flush;
 
     }
-    void Experiment::cross_validation(const std::string& fileName, bool quiet, bool no_train_score)
+    void generate_files(const std::string& fileName, bool discretize, bool stratified, int seed, int nfold, torch::Tensor X_train, torch::Tensor y_train, torch::Tensor X_test, torch::Tensor y_test, std::vector<int>& train, std::vector<int>& test)
     {
-        auto datasets = Datasets(discretized, Paths::datasets());
-        // Get dataset
-        auto [X, y] = datasets.getTensors(fileName);
-        auto states = datasets.getStates(fileName);
-        auto features = datasets.getFeatures(fileName);
-        auto samples = datasets.getNSamples(fileName);
-        auto className = datasets.getClassName(fileName);
-        auto labels = datasets.getLabels(fileName);
-        int num_classes = states[className].size() == 0 ? labels.size() : states[className].size();
-        if (!quiet) {
-            std::cout << " " << setw(5) << samples << " " << setw(5) << features.size() << flush;
+        std::string file_name = Paths::experiment_file(fileName, discretize, stratified, seed, nfold);
+        auto file = std::ofstream(file_name);
+        json output;
+        output["seed"] = seed;
+        output["nfold"] = nfold;
+        output["X_train"] = json::array();
+        auto n = X_train.size(1);
+        for (int i = 0; i < X_train.size(0); i++) {
+            if (X_train.dtype() == torch::kFloat32) {
+                auto xvf_ptr = X_train.index({ i }).data_ptr<float>();
+                auto feature = std::vector<float>(xvf_ptr, xvf_ptr + n);
+                output["X_train"].push_back(feature);
+            } else {
+                auto feature = std::vector<int>(X_train.index({ i }).data_ptr<int>(), X_train.index({ i }).data_ptr<int>() + n);
+                output["X_train"].push_back(feature);
+            }
         }
+        output["y_train"] = std::vector<int>(y_train.data_ptr<int>(), y_train.data_ptr<int>() + n);
+        output["X_test"] = json::array();
+        n = X_test.size(1);
+        for (int i = 0; i < X_test.size(0); i++) {
+            if (X_train.dtype() == torch::kFloat32) {
+                auto xvf_ptr = X_test.index({ i }).data_ptr<float>();
+                auto feature = std::vector<float>(xvf_ptr, xvf_ptr + n);
+                output["X_test"].push_back(feature);
+            } else {
+                auto feature = std::vector<int>(X_test.index({ i }).data_ptr<int>(), X_test.index({ i }).data_ptr<int>() + n);
+                output["X_test"].push_back(feature);
+            }
+        }
+        output["y_test"] = std::vector<int>(y_test.data_ptr<int>(), y_test.data_ptr<int>() + n);
+        output["train"] = train;
+        output["test"] = test;
+        file << output.dump(4);
+        file.close();
+    }
+    void Experiment::cross_validation(const std::string& fileName, bool quiet, bool no_train_score, bool generate_fold_files, bool graph)
+    {
+        //
+        // Load dataset and prepare data
+        //
+        auto datasets = Datasets(discretized, Paths::datasets(), discretization_algo);
+        auto& dataset = datasets.getDataset(fileName);
+        dataset.load();
+        auto [X, y] = dataset.getTensors(); // Only need y for folding
+        auto features = dataset.getFeatures();
+        auto n_features = dataset.getNFeatures();
+        auto n_samples = dataset.getNSamples();
+        auto className = dataset.getClassName();
+        auto labels = dataset.getLabels();
+        int num_classes = dataset.getNClasses();
+        if (!quiet) {
+            std::cout << " " << setw(5) << n_samples << " " << setw(5) << n_features << flush;
+        }
+        //
         // Prepare Result
+        //
         auto partial_result = PartialResult();
-        auto [values, counts] = at::_unique(y);
-        partial_result.setSamples(X.size(1)).setFeatures(X.size(0)).setClasses(values.size(0));
+        partial_result.setSamples(n_samples).setFeatures(n_features).setClasses(num_classes);
         partial_result.setHyperparameters(hyperparameters.get(fileName));
+        //
         // Initialize results std::vectors
+        //
         int nResults = nfolds * static_cast<int>(randomSeeds.size());
-        auto accuracy_test = torch::zeros({ nResults }, torch::kFloat64);
-        auto accuracy_train = torch::zeros({ nResults }, torch::kFloat64);
+        auto score_test = torch::zeros({ nResults }, torch::kFloat64);
+        auto score_train = torch::zeros({ nResults }, torch::kFloat64);
         auto train_time = torch::zeros({ nResults }, torch::kFloat64);
         auto test_time = torch::zeros({ nResults }, torch::kFloat64);
         auto nodes = torch::zeros({ nResults }, torch::kFloat64);
@@ -105,10 +175,16 @@ namespace platform {
         json confusion_matrices = json::array();
         json confusion_matrices_train = json::array();
         std::vector<std::string> notes;
-        Timer train_timer, test_timer;
+        std::vector<std::string> graphs;
+        Timer train_timer, test_timer, seed_timer;
         int item = 0;
         bool first_seed = true;
+        //
+        // Loop over random seeds
+        //
+        auto score = parse_score();
         for (auto seed : randomSeeds) {
+            seed_timer.start();
             if (!quiet) {
                 string prefix = " ";
                 if (!first_seed) {
@@ -121,26 +197,33 @@ namespace platform {
             if (stratified)
                 fold = new folding::StratifiedKFold(nfolds, y, seed);
             else
-                fold = new folding::KFold(nfolds, y.size(0), seed);
+                fold = new folding::KFold(nfolds, n_samples, seed);
+            //
+            // Loop over folds
+            //
             for (int nfold = 0; nfold < nfolds; nfold++) {
                 auto clf = Models::instance()->create(result.getModel());
+                if (!quiet)
+                    showProgress(nfold + 1, getColor(clf->getStatus()), "-");
                 setModelVersion(clf->getVersion());
                 auto valid = clf->getValidHyperparameters();
                 hyperparameters.check(valid, fileName);
                 clf->setHyperparameters(hyperparameters.get(fileName));
+                //
                 // Split train - test dataset
+                //
                 train_timer.start();
                 auto [train, test] = fold->getFold(nfold);
-                auto train_t = torch::tensor(train);
-                auto test_t = torch::tensor(test);
-                auto X_train = X.index({ "...", train_t });
-                auto y_train = y.index({ train_t });
-                auto X_test = X.index({ "...", test_t });
-                auto y_test = y.index({ test_t });
+                auto [X_train, X_test, y_train, y_test] = dataset.getTrainTestTensors(train, test);
+                auto states = dataset.getStates(); // Get the states of the features Once they are discretized
+                if (generate_fold_files)
+                    generate_files(fileName, discretized, stratified, seed, nfold, X_train, y_train, X_test, y_test, train, test);
                 if (!quiet)
                     showProgress(nfold + 1, getColor(clf->getStatus()), "a");
+                //
                 // Train model
-                clf->fit(X_train, y_train, features, className, states);
+                //
+                clf->fit(X_train, y_train, features, className, states, smooth_type);
                 if (!quiet)
                     showProgress(nfold + 1, getColor(clf->getStatus()), "b");
                 auto clf_notes = clf->getNotes();
@@ -150,40 +233,60 @@ namespace platform {
                 edges[item] = clf->getNumberOfEdges();
                 num_states[item] = clf->getNumberOfStates();
                 train_time[item] = train_timer.getDuration();
-                double accuracy_train_value = 0.0;
+                double score_train_value = 0.0;
+                //
                 // Score train
+                //
                 if (!no_train_score) {
-                    auto y_predict = clf->predict(X_train);
-                    Scores scores(y_train, y_predict, num_classes, labels);
-                    accuracy_train_value = scores.accuracy();
+                    auto y_proba_train = clf->predict_proba(X_train);
+                    Scores scores(y_train, y_proba_train, num_classes, labels);
+                    score_train_value = score == score_t::ACCURACY ? scores.accuracy() : scores.auc();
                     confusion_matrices_train.push_back(scores.get_confusion_matrix_json(true));
                 }
+                //
                 // Test model
+                //
                 if (!quiet)
                     showProgress(nfold + 1, getColor(clf->getStatus()), "c");
                 test_timer.start();
-                auto y_predict = clf->predict(X_test);
-                Scores scores(y_test, y_predict, num_classes, labels);
-                auto accuracy_test_value = scores.accuracy();
+                // auto y_predict = clf->predict(X_test);
+                auto y_proba_test = clf->predict_proba(X_test);
+                Scores scores(y_test, y_proba_test, num_classes, labels);
+                auto score_test_value = score == score_t::ACCURACY ? scores.accuracy() : scores.auc();
                 test_time[item] = test_timer.getDuration();
-                accuracy_train[item] = accuracy_train_value;
-                accuracy_test[item] = accuracy_test_value;
+                score_train[item] = score_train_value;
+                score_test[item] = score_test_value;
                 confusion_matrices.push_back(scores.get_confusion_matrix_json(true));
                 if (!quiet)
                     std::cout << "\b\b\b, " << flush;
+                //
                 // Store results and times in std::vector
-                partial_result.addScoreTrain(accuracy_train_value);
-                partial_result.addScoreTest(accuracy_test_value);
+                //
+                partial_result.addScoreTrain(score_train_value);
+                partial_result.addScoreTest(score_test_value);
                 partial_result.addTimeTrain(train_time[item].item<double>());
                 partial_result.addTimeTest(test_time[item].item<double>());
                 item++;
+                if (graph) {
+                    std::string result = "";
+                    for (const auto& line : clf->graph()) {
+                        result += line + "\n";
+                    }
+                    graphs.push_back(result);
+                }
+            }
+            if (!quiet) {
+                seed_timer.stop();
+                std::cout << "end. [" << seed_timer.getDurationString() << "]" << std::endl;
             }
-            if (!quiet)
-                std::cout << "end. " << flush;
             delete fold;
         }
-        partial_result.setScoreTest(torch::mean(accuracy_test).item<double>()).setScoreTrain(torch::mean(accuracy_train).item<double>());
-        partial_result.setScoreTestStd(torch::std(accuracy_test).item<double>()).setScoreTrainStd(torch::std(accuracy_train).item<double>());
+        //
+        // Store result totals in Result
+        //
+        partial_result.setGraph(graphs);
+        partial_result.setScoreTest(torch::mean(score_test).item<double>()).setScoreTrain(torch::mean(score_train).item<double>());
+        partial_result.setScoreTestStd(torch::std(score_test).item<double>()).setScoreTrainStd(torch::std(score_train).item<double>());
         partial_result.setTrainTime(torch::mean(train_time).item<double>()).setTestTime(torch::mean(test_time).item<double>());
         partial_result.setTestTimeStd(torch::std(test_time).item<double>()).setTrainTimeStd(torch::std(train_time).item<double>());
         partial_result.setNodes(torch::mean(nodes).item<double>()).setLeaves(torch::mean(edges).item<double>()).setDepth(torch::mean(num_states).item<double>());

src/main/Experiment.h

@@ -7,10 +7,11 @@
 #include "bayesnet/BaseClassifier.h"
 #include "HyperParameters.h"
 #include "results/Result.h"
+#include "bayesnet/network/Network.h"
 
 namespace platform {
     using json = nlohmann::ordered_json;
-
+    enum class score_t { NONE, ACCURACY, ROC_AUC_OVR };
     class Experiment {
     public:
         Experiment() = default;
@@ -20,6 +21,25 @@ namespace platform {
         Experiment& setModelVersion(const std::string& model_version) { this->result.setModelVersion(model_version); return *this; }
         Experiment& setModel(const std::string& model) { this->result.setModel(model); return *this; }
         Experiment& setLanguage(const std::string& language) { this->result.setLanguage(language); return *this; }
+        Experiment& setDiscretizationAlgorithm(const std::string& discretization_algo)
+        {
+            this->discretization_algo = discretization_algo; this->result.setDiscretizationAlgorithm(discretization_algo); return *this;
+        }
+        Experiment& setSmoothSrategy(const std::string& smooth_strategy)
+        {
+            this->smooth_strategy = smooth_strategy; this->result.setSmoothStrategy(smooth_strategy);
+            if (smooth_strategy == "ORIGINAL")
+                smooth_type = bayesnet::Smoothing_t::ORIGINAL;
+            else if (smooth_strategy == "LAPLACE")
+                smooth_type = bayesnet::Smoothing_t::LAPLACE;
+            else if (smooth_strategy == "CESTNIK")
+                smooth_type = bayesnet::Smoothing_t::CESTNIK;
+            else {
+                std::cerr << "Experiment: Unknown smoothing strategy: " << smooth_strategy << std::endl;
+                exit(1);
+            }
+            return *this;
+        }
         Experiment& setLanguageVersion(const std::string& language_version) { this->result.setLanguageVersion(language_version); return *this; }
         Experiment& setDiscretized(bool discretized) { this->discretized = discretized; result.setDiscretized(discretized); return *this; }
         Experiment& setStratified(bool stratified) { this->stratified = stratified; result.setStratified(stratified); return *this; }
@@ -28,16 +48,21 @@ namespace platform {
         Experiment& addRandomSeed(int randomSeed) { randomSeeds.push_back(randomSeed); result.addSeed(randomSeed); return *this; }
         Experiment& setDuration(float duration) { this->result.setDuration(duration); return *this; }
         Experiment& setHyperparameters(const HyperParameters& hyperparameters_) { this->hyperparameters = hyperparameters_; return *this; }
-        void cross_validation(const std::string& fileName, bool quiet, bool no_train_score);
-        void go(std::vector<std::string> filesToProcess, bool quiet, bool no_train_score);
+        void cross_validation(const std::string& fileName, bool quiet, bool no_train_score, bool generate_fold_files, bool graph);
+        void go(std::vector<std::string> filesToProcess, bool quiet, bool no_train_score, bool generate_fold_files, bool graph);
         void saveResult();
         void show();
+        void saveGraph();
         void report(bool classification_report = false);
     private:
+        score_t parse_score() const;
         Result result;
         bool discretized{ false }, stratified{ false };
         std::vector<PartialResult> results;
         std::vector<int> randomSeeds;
+        std::string discretization_algo;
+        std::string smooth_strategy;
+        bayesnet::Smoothing_t smooth_type{ bayesnet::Smoothing_t::NONE };
         HyperParameters hyperparameters;
         int nfolds{ 0 };
         int max_name{ 7 }; // max length of dataset name for formatting (default 7)

src/main/HyperParameters.cpp

@@ -10,16 +10,9 @@ namespace platform {
         for (const auto& item : datasets) {
             hyperparameters[item] = hyperparameters_;
         }
+        normalize_nested(datasets);
     }
-    // https://www.techiedelight.com/implode-a-vector-of-strings-into-a-comma-separated-string-in-cpp/
-    std::string join(std::vector<std::string> const& strings, std::string delim)
-    {
-        std::stringstream ss;
-        std::copy(strings.begin(), strings.end(),
-            std::ostream_iterator<std::string>(ss, delim.c_str()));
-        return ss.str();
-    }
-    HyperParameters::HyperParameters(const std::vector<std::string>& datasets, const std::string& hyperparameters_file)
+    HyperParameters::HyperParameters(const std::vector<std::string>& datasets, const std::string& hyperparameters_file, bool best)
     {
         // Check if file exists
         std::ifstream file(hyperparameters_file);
@@ -28,7 +21,14 @@ namespace platform {
         }
         // Check if file is a json
         json file_hyperparameters = json::parse(file);
-        auto input_hyperparameters = file_hyperparameters["results"];
+        json input_hyperparameters;
+        if (best) {
+            for (const auto& [key, value] : file_hyperparameters.items()) {
+                input_hyperparameters[key]["hyperparameters"] = value[1];
+            }
+        } else {
+            input_hyperparameters = file_hyperparameters["results"];
+        }
         // Check if hyperparameters are valid
         for (const auto& dataset : datasets) {
             if (!input_hyperparameters.contains(dataset)) {
@@ -38,6 +38,24 @@ namespace platform {
             }
             hyperparameters[dataset] = input_hyperparameters[dataset]["hyperparameters"].get<json>();
         }
+        normalize_nested(datasets);
+    }
+    void HyperParameters::normalize_nested(const std::vector<std::string>& datasets)
+    {
+        // for (const auto& dataset : datasets) {
+        //     if (hyperparameters[dataset].contains("be_hyperparams")) {
+        //         // Odte has base estimator hyperparameters set this way
+        //         hyperparameters[dataset]["be_hyperparams"] = hyperparameters[dataset]["be_hyperparams"].dump();
+        //     }
+        // }
+    }
+    // https://www.techiedelight.com/implode-a-vector-of-strings-into-a-comma-separated-string-in-cpp/
+    std::string join(std::vector<std::string> const& strings, std::string delim)
+    {
+        std::stringstream ss;
+        std::copy(strings.begin(), strings.end(),
+            std::ostream_iterator<std::string>(ss, delim.c_str()));
+        return ss.str();
     }
     void HyperParameters::check(const std::vector<std::string>& valid, const std::string& fileName)
     {

src/main/HyperParameters.h

@@ -10,14 +10,18 @@ namespace platform {
     class HyperParameters {
     public:
         HyperParameters() = default;
+        // Constructor to use command line hyperparameters
         explicit HyperParameters(const std::vector<std::string>& datasets, const json& hyperparameters_);
-        explicit HyperParameters(const std::vector<std::string>& datasets, const std::string& hyperparameters_file);
+        // Constructor to use hyperparameters file generated by grid or by best results
+        explicit HyperParameters(const std::vector<std::string>& datasets, const std::string& hyperparameters_file, bool best = false);
         ~HyperParameters() = default;
         bool notEmpty(const std::string& key) const { return !hyperparameters.at(key).empty(); }
         void check(const std::vector<std::string>& valid, const std::string& fileName);
         json get(const std::string& fileName);
     private:
+        void normalize_nested(const std::vector<std::string>& datasets);
         std::map<std::string, json> hyperparameters;
+        bool best = false; // Used to separate grid/best hyperparameters as the format of those files are different
     };
 } /* namespace platform */
 #endif

src/main/PartialResult.h

@@ -15,6 +15,7 @@ namespace platform {
             data["times_train"] = json::array();
             data["times_test"] = json::array();
             data["notes"] = json::array();
+            data["graph"] = json::array();
             data["train_time"] = 0.0;
             data["train_time_std"] = 0.0;
             data["test_time"] = 0.0;
@@ -27,6 +28,12 @@ namespace platform {
             data["notes"].insert(data["notes"].end(), notes_.begin(), notes_.end());
             return *this;
         }
+        PartialResult& setGraph(const std::vector<std::string>& graph)
+        {
+            json graph_ = graph;
+            data["graph"].insert(data["graph"].end(), graph_.begin(), graph_.end());
+            return *this;
+        }
         PartialResult& setConfusionMatrices(const json& confusion_matrices) { data["confusion_matrices"] = confusion_matrices; return *this; }
         PartialResult& setConfusionMatricesTrain(const json& confusion_matrices) { data["confusion_matrices_train"] = confusion_matrices; return *this; }
         PartialResult& setHyperparameters(const json& hyperparameters) { data["hyperparameters"] = hyperparameters; return *this; }

src/main/RocAuc.cpp

@@ -0,0 +1,67 @@
+#include <sstream>
+#include <algorithm>
+#include <numeric>
+#include <utility>
+#include "RocAuc.h"
+namespace platform {
+    
+    double RocAuc::compute(const torch::Tensor& y_proba, const torch::Tensor& labels)
+    {
+        size_t nClasses = y_proba.size(1);
+        // In binary classification problem there's no need to calculate the average of the AUCs
+        if (nClasses == 2)
+            nClasses = 1;
+        size_t nSamples = y_proba.size(0);
+        y_test = tensorToVector(labels);
+        std::vector<double> aucScores(nClasses, 0.0);
+        for (size_t classIdx = 0; classIdx < nClasses; ++classIdx) {
+            scoresAndLabels.clear();
+            for (size_t i = 0; i < nSamples; ++i) {
+                scoresAndLabels.emplace_back(y_proba[i][classIdx].item<float>(), y_test[i] == classIdx ? 1 : 0);
+            }
+            aucScores[classIdx] = compute_common(nSamples, classIdx);
+        }
+        return std::accumulate(aucScores.begin(), aucScores.end(), 0.0) / nClasses;
+    }
+    double RocAuc::compute(const std::vector<std::vector<double>>& y_proba, const std::vector<int>& labels)
+    {
+        y_test = labels;
+        size_t nClasses = y_proba[0].size();
+        // In binary classification problem there's no need to calculate the average of the AUCs
+        if (nClasses == 2)
+            nClasses = 1;
+        size_t nSamples = y_proba.size();
+        std::vector<double> aucScores(nClasses, 0.0);
+        for (size_t classIdx = 0; classIdx < nClasses; ++classIdx) {
+            scoresAndLabels.clear();
+            for (size_t i = 0; i < nSamples; ++i) {
+                scoresAndLabels.emplace_back(y_proba[i][classIdx], labels[i] == classIdx ? 1 : 0);
+            }
+            aucScores[classIdx] = compute_common(nSamples, classIdx);
+        }
+        return std::accumulate(aucScores.begin(), aucScores.end(), 0.0) / nClasses;
+    }
+    double RocAuc::compute_common(size_t nSamples, size_t classIdx)
+    {
+        std::sort(scoresAndLabels.begin(), scoresAndLabels.end(), std::greater<>());
+        std::vector<double> tpr, fpr;
+        double tp = 0, fp = 0;
+        double totalPos = std::count(y_test.begin(), y_test.end(), classIdx);
+        double totalNeg = nSamples - totalPos;
+
+        for (const auto& [score, label] : scoresAndLabels) {
+            if (label == 1) {
+                tp += 1;
+            } else {
+                fp += 1;
+            }
+            tpr.push_back(tp / totalPos);
+            fpr.push_back(fp / totalNeg);
+        }
+        double auc = 0.0;
+        for (size_t i = 1; i < tpr.size(); ++i) {
+            auc += 0.5 * (fpr[i] - fpr[i - 1]) * (tpr[i] + tpr[i - 1]);
+        }
+        return auc;
+    }
+}

src/main/RocAuc.h

@@ -0,0 +1,21 @@
+#ifndef ROCAUC_H
+#define ROCAUC_H
+#include <torch/torch.h>
+#include <vector>
+#include <string>
+#include <nlohmann/json.hpp>
+
+namespace platform {
+    using json = nlohmann::ordered_json;
+    class RocAuc {
+    public:
+        RocAuc() = default;
+        double compute(const std::vector<std::vector<double>>& y_proba, const std::vector<int>& y_test);
+        double compute(const torch::Tensor& y_proba, const torch::Tensor& y_test);
+    private:
+        double compute_common(size_t nSamples, size_t classIdx);
+        std::vector<std::pair<double, int>> scoresAndLabels;
+        std::vector<int> y_test;
+    };
+}
+#endif

src/main/Scores.cpp

@@ -1,13 +1,15 @@
 #include <sstream>
 #include "Scores.h"
+#include "common/Utils.h" // tensorToVector
 #include "common/Colors.h"
 namespace platform {
-    Scores::Scores(torch::Tensor& y_test, torch::Tensor& y_pred, int num_classes, std::vector<std::string> labels) : num_classes(num_classes), labels(labels)
+    Scores::Scores(torch::Tensor& y_test, torch::Tensor& y_proba, int num_classes, std::vector<std::string> labels) : num_classes(num_classes), labels(labels), y_test(y_test), y_proba(y_proba)
     {
         if (labels.size() == 0) {
             init_default_labels();
         }
         total = y_test.size(0);
+        auto y_pred = y_proba.argmax(1);
         accuracy_value = (y_pred == y_test).sum().item<float>() / total;
         init_confusion_matrix();
         for (int i = 0; i < total; i++) {
@@ -16,7 +18,7 @@ namespace platform {
             confusion_matrix[actual][predicted] += 1;
         }
     }
-    Scores::Scores(json& confusion_matrix_)
+    Scores::Scores(const json& confusion_matrix_)
     {
         json values;
         total = 0;
@@ -40,7 +42,49 @@ namespace platform {
         }
         compute_accuracy_value();
     }
-    Scores Scores::create_aggregate(json& data, std::string key)
+    float Scores::auc()
+    {
+        size_t nSamples = y_test.numel();
+        if (nSamples == 0) return 0;
+        // In binary classification problem there's no need to calculate the average of the AUCs
+        auto nClasses = num_classes;
+        if (num_classes == 2)
+            nClasses = 1;
+        auto y_testv = tensorToVector<int>(y_test);
+        std::vector<double> aucScores(nClasses, 0.0);
+        std::vector<std::pair<double, int>> scoresAndLabels;
+        for (size_t classIdx = 0; classIdx < nClasses; ++classIdx) {
+            if (classIdx >= y_proba.size(1)) {
+                std::cerr << "AUC warning - class index out of range" << std::endl;
+                return 0;
+            }
+            scoresAndLabels.clear();
+            for (size_t i = 0; i < nSamples; ++i) {
+                scoresAndLabels.emplace_back(y_proba[i][classIdx].item<float>(), y_testv[i] == classIdx ? 1 : 0);
+            }
+            std::sort(scoresAndLabels.begin(), scoresAndLabels.end(), std::greater<>());
+            std::vector<double> tpr, fpr;
+            double tp = 0, fp = 0;
+            double totalPos = std::count(y_testv.begin(), y_testv.end(), classIdx);
+            double totalNeg = nSamples - totalPos;
+            for (const auto& [score, label] : scoresAndLabels) {
+                if (label == 1) {
+                    tp += 1;
+                } else {
+                    fp += 1;
+                }
+                tpr.push_back(tp / totalPos);
+                fpr.push_back(fp / totalNeg);
+            }
+            double auc = 0.0;
+            for (size_t i = 1; i < tpr.size(); ++i) {
+                auc += 0.5 * (fpr[i] - fpr[i - 1]) * (tpr[i] + tpr[i - 1]);
+            }
+            aucScores[classIdx] = auc;
+        }
+        return std::accumulate(aucScores.begin(), aucScores.end(), 0.0) / nClasses;
+    }
+    Scores Scores::create_aggregate(const json& data, const std::string key)
     {
         auto scores = Scores(data[key][0]);
         for (int i = 1; i < data[key].size(); i++) {
@@ -138,6 +182,25 @@ namespace platform {
             << std::setw(dlen) << std::right << support;
         return oss.str();
     }
+    std::tuple<float, float, float, float> Scores::compute_averages()
+    {
+        float precision_avg = 0;
+        float recall_avg = 0;
+        float precision_wavg = 0;
+        float recall_wavg = 0;
+        for (int i = 0; i < num_classes; i++) {
+            int support = confusion_matrix[i].sum().item<int>();
+            precision_avg += precision(i);
+            precision_wavg += precision(i) * support;
+            recall_avg += recall(i);
+            recall_wavg += recall(i) * support;
+        }
+        precision_wavg /= total;
+        recall_wavg /= total;
+        precision_avg /= num_classes;
+        recall_avg /= num_classes;
+        return { precision_avg, recall_avg, precision_wavg, recall_wavg };
+    }
     std::vector<std::string> Scores::classification_report(std::string color, std::string title)
     {
         std::stringstream oss;
@@ -157,21 +220,7 @@ namespace platform {
         report.push_back(" ");
         oss << classification_report_line("accuracy", 0, 0, accuracy(), total);
         report.push_back(oss.str()); oss.str("");
-        float precision_avg = 0;
-        float recall_avg = 0;
-        float precision_wavg = 0;
-        float recall_wavg = 0;
-        for (int i = 0; i < num_classes; i++) {
-            int support = confusion_matrix[i].sum().item<int>();
-            precision_avg += precision(i);
-            precision_wavg += precision(i) * support;
-            recall_avg += recall(i);
-            recall_wavg += recall(i) * support;
-        }
-        precision_wavg /= total;
-        recall_wavg /= total;
-        precision_avg /= num_classes;
-        recall_avg /= num_classes;
+        auto [precision_avg, recall_avg, precision_wavg, recall_wavg] = compute_averages();
         report.push_back(classification_report_line("macro avg", precision_avg, recall_avg, f1_macro(), total));
         report.push_back(classification_report_line("weighted avg", precision_wavg, recall_wavg, f1_weighted(), total));
         report.push_back("");
@@ -189,17 +238,33 @@ namespace platform {
         }
         return report;
     }
+    json Scores::classification_report_json(std::string title)
+    {
+        json output;
+        output["title"] = "Classification Report using " + title + " dataset";
+        output["headers"] = { " ", "precision", "recall", "f1-score", "support" };
+        output["body"] = {};
+        for (int i = 0; i < num_classes; i++) {
+            output["body"].push_back({ labels[i], precision(i), recall(i), f1_score(i), confusion_matrix[i].sum().item<int>() });
+        }
+        output["accuracy"] = { "accuracy", 0, 0, accuracy(), total };
+        auto [precision_avg, recall_avg, precision_wavg, recall_wavg] = compute_averages();
+        output["averages"] = { "macro avg", precision_avg, recall_avg, f1_macro(), total };
+        output["weighted"] = { "weighted avg", precision_wavg, recall_wavg, f1_weighted(), total };
+        output["confusion_matrix"] = get_confusion_matrix_json();
+        return output;
+    }
     json Scores::get_confusion_matrix_json(bool labels_as_keys)
     {
-        json j;
+        json output;
         for (int i = 0; i < num_classes; i++) {
             auto r_ptr = confusion_matrix[i].data_ptr<int>();
             if (labels_as_keys) {
-                j[labels[i]] = std::vector<int>(r_ptr, r_ptr + num_classes);
+                output[labels[i]] = std::vector<int>(r_ptr, r_ptr + num_classes);
             } else {
-                j[i] = std::vector<int>(r_ptr, r_ptr + num_classes);
+                output[i] = std::vector<int>(r_ptr, r_ptr + num_classes);
             }
         }
-        return j;
+        return output;
     }
 }

src/main/Scores.h

@@ -4,16 +4,16 @@
 #include <vector>
 #include <string>
 #include <nlohmann/json.hpp>
-#include <xlsxwriter.h>
 
 namespace platform {
     using json = nlohmann::ordered_json;
     class Scores {
     public:
-        Scores(torch::Tensor& y_test, torch::Tensor& y_pred, int num_classes, std::vector<std::string> labels = {});
-        explicit Scores(json& confusion_matrix_);
-        static Scores create_aggregate(json& data, std::string key);
+        Scores(torch::Tensor& y_test, torch::Tensor& y_proba, int num_classes, std::vector<std::string> labels = {});
+        explicit Scores(const json& confusion_matrix_);
+        static Scores create_aggregate(const json& data, const std::string key);
         float accuracy();
+        float auc();
         float f1_score(int num_class);
         float f1_weighted();
         float f1_macro();
@@ -21,6 +21,7 @@ namespace platform {
         float recall(int num_class);
         torch::Tensor get_confusion_matrix() { return confusion_matrix; }
         std::vector<std::string> classification_report(std::string color = "", std::string title = "");
+        json classification_report_json(std::string title = "");
         json get_confusion_matrix_json(bool labels_as_keys = false);
         void aggregate(const Scores& a);
     private:
@@ -28,11 +29,15 @@ namespace platform {
         void init_confusion_matrix();
         void init_default_labels();
         void compute_accuracy_value();
+        std::tuple<float, float, float, float> compute_averages();
         int num_classes;
         float accuracy_value;
         int total;
         std::vector<std::string> labels;
         torch::Tensor confusion_matrix; // Rows ar actual, columns are predicted
+        torch::Tensor null_t; // Covenient null tensor needed when confusion_matrix constructor is used
+        torch::Tensor& y_test = null_t; // for ROC AUC
+        torch::Tensor& y_proba = null_t; // for ROC AUC
         int label_len = 16;
         int dlen = 9;
         int ndec = 7;

src/manage/CommandParser.h

@@ -1,21 +0,0 @@
-#ifndef COMMAND_PARSER_H
-#define COMMAND_PARSER_H
-#include <string>
-#include <vector>
-#include <tuple>
-
-namespace platform {
-    class CommandParser {
-    public:
-        CommandParser() = default;
-        std::tuple<char, int, bool> parse(const std::string& color, const std::vector<std::tuple<std::string, char, bool>>& options, const char defaultCommand, const int minIndex, const int maxIndex);
-        char getCommand() const { return command; };
-        int getIndex() const { return index; };
-        std::string getErrorMessage() const { return errorMessage; };
-    private:
-        std::string errorMessage;
-        char command;
-        int index;
-    };
-} /* namespace platform */
-#endif

src/manage/ManageScreen.cpp

@@ -3,29 +3,33 @@
 #include <string>
 #include <algorithm>
 #include "folding.hpp"
-#include "common/Colors.h"
 #include "common/CLocale.h"
 #include "common/Paths.h"
-#include "CommandParser.h"
+#include "OptionsMenu.h"
 #include "ManageScreen.h"
 #include "reports/DatasetsConsole.h"
 #include "reports/ReportConsole.h"
 #include "reports/ReportExcel.h"
 #include "reports/ReportExcelCompared.h"
 #include <bayesnet/classifiers/TAN.h>
-#include "CPPFImdlp.h"
+#include <fimdlp/CPPFImdlp.h>
 
 namespace platform {
     const std::string STATUS_OK = "Ok.";
     const std::string STATUS_COLOR = Colors::GREEN();
+
     ManageScreen::ManageScreen(int rows, int cols, const std::string& model, const std::string& score, const std::string& platform, bool complete, bool partial, bool compare) :
         rows{ rows }, cols{ cols }, complete{ complete }, partial{ partial }, compare{ compare }, didExcel(false), results(ResultsManager(model, score, platform, complete, partial))
     {
         results.load();
         openExcel = false;
         workbook = NULL;
-        this->rows = std::max(0, rows - 6); // 6 is the number of lines used by the menu & header
-        cols = std::max(cols, 140);
+        maxModel = results.maxModelSize();
+        maxTitle = results.maxTitleSize();
+        header_lengths = { 3, 10, maxModel, 11, 10, 12, 2, 3, 7, maxTitle };
+        header_labels = { " #", "Date", "Model", "Score Name", "Score", "Platform", "SD", "C/P", "Time", "Title" };
+        sort_fields = { "Date", "Model", "Score", "Time" };
+        updateSize(rows, cols);
         // Initializes the paginator for each output type (experiments, datasets, result)
         for (int i = 0; i < static_cast<int>(OutputType::Count); i++) {
             paginator.push_back(Paginator(this->rows, results.size()));
@@ -36,12 +40,41 @@ namespace platform {
         subIndex = -1;
         output_type = OutputType::EXPERIMENTS;
     }
+    void ManageScreen::computeSizes()
+    {
+        int minTitle = 10;
+        // set 10 chars as minimum for Title
+        auto header_title = header_lengths[header_lengths.size() - 1];
+        min_columns = std::accumulate(header_lengths.begin(), header_lengths.end(), 0) + header_lengths.size() - header_title + minTitle;
+        maxTitle = minTitle + cols - min_columns;
+        header_lengths[header_lengths.size() - 1] = maxTitle;
+        cols = std::min(cols, min_columns + maxTitle);
+        for (auto& paginator_ : paginator) {
+            paginator_.setPageSize(rows);
+        }
+    }
+    bool ManageScreen::checkWrongColumns()
+    {
+        if (min_columns > cols) {
+            std::cerr << Colors::MAGENTA() << "Make screen bigger to fit the results! " + std::to_string(min_columns - cols) + " columns needed! " << std::endl;
+            return true;
+        }
+        return false;
+    }
+    void ManageScreen::updateSize(int rows_, int cols_)
+    {
+        rows = std::max(6, rows_ - 6); // 6 is the number of lines used by the menu & header
+        cols = cols_;
+        computeSizes();
+    }
     void ManageScreen::doMenu()
     {
         if (results.empty()) {
-            std::cout << Colors::MAGENTA() << "No results found!" << Colors::RESET() << std::endl;
+            std::cerr << Colors::MAGENTA() << "No results found!" << Colors::RESET() << std::endl;
             return;
         }
+        if (checkWrongColumns())
+            return;
         results.sortResults(sort_field, sort_type);
         list(STATUS_OK, STATUS_COLOR);
         menu();
@@ -115,7 +148,6 @@ namespace platform {
     }
     void ManageScreen::list_result(const std::string& status_message, const std::string& status_color)
     {
-
         auto data = results.at(index).getJson();
         ReportConsole report(data, compare);
         auto header_text = report.getHeader();
@@ -140,11 +172,9 @@ namespace platform {
         // Status Area
         //
         footer(status_message, status_color);
-
     }
     void ManageScreen::list_detail(const std::string& status_message, const std::string& status_color)
     {
-
         auto data = results.at(index).getJson();
         ReportConsole report(data, compare, subIndex);
         auto header_text = report.getHeader();
@@ -169,7 +199,6 @@ namespace platform {
         // Status Area
         //
         footer(status_message, status_color);
-
     }
     void ManageScreen::list_datasets(const std::string& status_message, const std::string& status_color)
     {
@@ -193,7 +222,6 @@ namespace platform {
         // Status Area
         //
         footer(status_message, status_color);
-
     }
     void ManageScreen::list_experiments(const std::string& status_message, const std::string& status_color)
     {
@@ -201,17 +229,9 @@ namespace platform {
         // header
         //
         header();
-        //
-        // Field names
-        //
-        int maxModel = results.maxModelSize();
-        int maxTitle = results.maxTitleSize();
-        std::vector<int> header_lengths = { 3, 10, maxModel, 11, 10, 12, 2, 3, 7, maxTitle };
         std::cout << Colors::RESET();
         std::string arrow_dn = Symbols::down_arrow + " ";
         std::string arrow_up = Symbols::up_arrow + " ";
-        std::vector<std::string> header_labels = { " #", "Date", "Model", "Score Name", "Score", "Platform", "SD", "C/P", "Time", "Title" };
-        std::vector<std::string> sort_fields = { "Date", "Model", "Score", "Time" };
         for (int i = 0; i < header_labels.size(); i++) {
             std::string suffix = "", color = Colors::GREEN();
             int diff = 0;
@@ -230,11 +250,15 @@ namespace platform {
         //
         // Results
         //
+        if (results.empty()) {
+            std::cout << "No results found!" << std::endl;
+            return;
+        }
         auto [index_from, index_to] = paginator[static_cast<int>(output_type)].getOffset();
         for (int i = index_from; i <= index_to; i++) {
             auto color = (i % 2) ? Colors::BLUE() : Colors::CYAN();
             std::cout << color << std::setw(3) << std::fixed << std::right << i << " ";
-            std::cout << results.at(i).to_string(maxModel) << std::endl;
+            std::cout << results.at(i).to_string(maxModel, maxTitle) << std::endl;
         }
         //
         // Status Area
@@ -289,16 +313,28 @@ namespace platform {
     }
     std::pair<std::string, std::string> ManageScreen::sortList()
     {
-        std::cout << Colors::YELLOW() << "Choose sorting field (date='d', score='s', time='t', model='m', ascending='+', descending='-'): ";
-        std::vector<std::string> fields = { "Date", "Model", "Score", "Time" };
+        std::vector<std::tuple<std::string, char, bool>>  sortOptions = {
+            {"date", 'd', false},
+            {"score", 's', false},
+            {"time", 't', false},
+            {"model", 'm', false},
+            {"ascending+", '+', false},
+            {"descending-", '-', false}
+        };
+        auto sortMenu = OptionsMenu(sortOptions, Colors::YELLOW(), Colors::RED(), cols);
         std::string invalid_option = "Invalid sorting option";
-        std::string line;
         char option;
-        getline(std::cin, line);
-        if (line.size() == 0 || line.size() > 1) {
-            return { Colors::RED(), invalid_option };
+        bool parserError = true; // force the first iteration
+        while (parserError) {
+            if (checkWrongColumns())
+                return { Colors::RED(), "Invalid column size" };
+            auto [min_index, max_index] = paginator[static_cast<int>(output_type)].getOffset();
+            std::tie(option, index, parserError) = sortMenu.parse(' ', 0, 0);
+            sortMenu.updateColumns(cols);
+            if (parserError) {
+                return { Colors::RED(), invalid_option };
+            }
         }
-        option = line[0];
         switch (option) {
             case 'd':
                 sort_field = SortField::DATE;
@@ -322,7 +358,7 @@ namespace platform {
                 return { Colors::RED(), invalid_option };
         }
         results.sortResults(sort_field, sort_type);
-        return { Colors::GREEN(), "Sorted by " + fields[static_cast<int>(sort_field)] + " " + (sort_type == SortType::ASC ? "ascending" : "descending") };
+        return { Colors::GREEN(), "Sorted by " + sort_fields[static_cast<int>(sort_field)] + " " + (sort_type == SortType::ASC ? "ascending" : "descending") };
     }
     void ManageScreen::menu()
     {
@@ -333,17 +369,17 @@ namespace platform {
         std::vector<std::tuple<std::string, char, bool>>  mainOptions = {
             {"quit", 'q', false},
             {"list", 'l', false},
-            {"delete", 'D', true},
+            {"Delete", 'D', true},
             {"datasets", 'd', false},
             {"hide", 'h', true},
             {"sort", 's', false},
             {"report", 'r', true},
             {"excel", 'e', true},
             {"title", 't', true},
-            {"set A", 'a', true},
-            {"set B", 'b', true},
+            {"set A", 'A', true},
+            {"set B", 'B', true},
             {"compare A~B", 'c', false},
-            {"Page", 'p', true},
+            {"page", 'p', true},
             {"Page+", '+', false },
             {"Page-", '-', false}
         };
@@ -352,25 +388,33 @@ namespace platform {
             {"quit", 'q', false},
             {"report", 'r', true},
             {"list", 'l', false},
-            {"excel", 'e', false},
+            {"excel", 'e', true},
             {"back", 'b', false},
-            {"Page", 'p', true},
+            {"page", 'p', true},
             {"Page+", '+', false},
             {"Page-", '-', false}
         };
-
-        auto parser = CommandParser();
         while (!finished) {
+            auto main_menu = OptionsMenu(mainOptions, Colors::IGREEN(), Colors::YELLOW(), cols);
+            auto list_menu = OptionsMenu(listOptions, Colors::IBLUE(), Colors::YELLOW(), cols);
+            OptionsMenu& menu = output_type == OutputType::EXPERIMENTS ? main_menu : list_menu;
             bool parserError = true; // force the first iteration
             while (parserError) {
+                int index_menu;
                 auto [min_index, max_index] = paginator[static_cast<int>(output_type)].getOffset();
+                std::tie(option, index_menu, parserError) = menu.parse('r', min_index, max_index);
                 if (output_type == OutputType::EXPERIMENTS) {
-                    std::tie(option, index, parserError) = parser.parse(Colors::IGREEN(), mainOptions, 'r', min_index, max_index);
+                    index = index_menu;
                 } else {
-                    std::tie(option, subIndex, parserError) = parser.parse(Colors::IBLUE(), listOptions, 'r', min_index, max_index);
+                    subIndex = index_menu;
                 }
+                if (min_columns > cols) {
+                    std::cerr << "Make screen bigger to fit the results! " + std::to_string(min_columns - cols) + " columns needed! " << std::endl;
+                    return;
+                }
+                menu.updateColumns(cols);
                 if (parserError) {
-                    list(parser.getErrorMessage(), Colors::RED());
+                    list(menu.getErrorMessage(), Colors::RED());
                 }
             }
             switch (option) {
@@ -405,7 +449,7 @@ namespace platform {
                 case 'q':
                     finished = true;
                     break;
-                case 'a':
+                case 'A':
                     if (index == index_B) {
                         list("A and B cannot be the same!", Colors::RED());
                         break;
@@ -413,7 +457,7 @@ namespace platform {
                     index_A = index;
                     list("A set to " + std::to_string(index), Colors::GREEN());
                     break;
-                case 'b': // set_b or back to list
+                case 'B': // set_b or back to list
                     if (output_type == OutputType::EXPERIMENTS) {
                         if (index == index_A) {
                             list("A and B cannot be the same!", Colors::RED());
@@ -465,6 +509,7 @@ namespace platform {
                         std::cout << "Hiding " << filename << std::endl;
                         results.hideResult(index, Paths::hiddenResults());
                         status_message = filename + " hidden! (moved to " + Paths::hiddenResults() + ")";
+                        paginator[static_cast<int>(OutputType::EXPERIMENTS)].setTotal(results.size());
                         list(status_message, Colors::YELLOW());
                     }
                     break;

src/manage/ManageScreen.h

@@ -2,6 +2,7 @@
 #define MANAGE_SCREEN_H
 #include <xlsxwriter.h>
 #include "ResultsManager.h"
+#include "common/Colors.h"
 #include "Paginator.hpp"
 
 namespace platform {
@@ -17,6 +18,7 @@ namespace platform {
         ManageScreen(int rows, int cols, const std::string& model, const std::string& score, const std::string& platform, bool complete, bool partial, bool compare);
         ~ManageScreen() = default;
         void doMenu();
+        void updateSize(int rows, int cols);
     private:
         void list(const std::string& status, const std::string& color);
         void list_experiments(const std::string& status, const std::string& color);
@@ -28,12 +30,15 @@ namespace platform {
         std::string report_compared();
         std::pair<std::string, std::string> sortList();
         std::string getVersions();
+        void computeSizes();
+        bool checkWrongColumns();
         void menu();
         void header();
         void footer(const std::string& status, const std::string& color);
         OutputType output_type;
         int rows;
         int cols;
+        int min_columns;
         int index;
         int subIndex;
         int index_A, index_B; // used for comparison of experiments
@@ -43,6 +48,10 @@ namespace platform {
         bool complete;
         bool partial;
         bool compare;
+        int maxModel, maxTitle;
+        std::vector<std::string> header_labels;
+        std::vector<int> header_lengths;
+        std::vector<std::string> sort_fields;
         SortField sort_field = SortField::DATE;
         SortType sort_type = SortType::DESC;
         std::vector<Paginator> paginator;

src/manage/OptionsMenu.cpp

@@ -1,30 +1,46 @@
-#include "CommandParser.h"
+#include "OptionsMenu.h"
 #include <iostream>
 #include <sstream>
 #include <algorithm>
-#include "common/Colors.h"
 #include "common/Utils.h"
 
 namespace platform {
-
-    std::tuple<char, int, bool> CommandParser::parse(const std::string& color, const std::vector<std::tuple<std::string, char, bool>>& options, const char defaultCommand, const int minIndex, const int maxIndex)
+    std::string OptionsMenu::to_string()
+    {
+        bool first = true;
+        std::string result = color_normal + "Options: (";
+        size_t size = 10; // Size of "Options: ("
+        for (auto& option : options) {
+            if (!first) {
+                result += ", ";
+                size += 2;
+            }
+            std::string title = std::get<0>(option);
+            auto pos = title.find(std::get<1>(option));
+            result += color_normal + title.substr(0, pos) + color_bold + title.substr(pos, 1) + color_normal + title.substr(pos + 1);
+            size += title.size();
+            first = false;
+        }
+        if (size + 3 > cols) { // 3 is the size of the "): " at the end
+            result = "";
+            first = true;
+            for (auto& option : options) {
+                if (!first) {
+                    result += color_normal + ", ";
+                }
+                result += color_bold + std::get<1>(option);
+                first = false;
+            }
+        }
+        result += "): ";
+        return result;
+    }
+    std::tuple<char, int, bool> OptionsMenu::parse(char defaultCommand, int minIndex, int maxIndex)
     {
         bool finished = false;
         while (!finished) {
-            std::stringstream oss;
+            std::cout << to_string();
             std::string line;
-            oss << color << "Options (";
-            bool first = true;
-            for (auto& option : options) {
-                if (first) {
-                    first = false;
-                } else {
-                    oss << ", ";
-                }
-                oss << std::get<char>(option) << "=" << std::get<std::string>(option);
-            }
-            oss << "): ";
-            std::cout << oss.str();
             getline(std::cin, line);
             line = trim(line);
             if (line.size() == 0) {

src/manage/OptionsMenu.h

@@ -0,0 +1,26 @@
+#ifndef OPTIONS_MENU_H
+#define OPTIONS_MENU_H
+#include <string>
+#include <vector>
+#include <tuple>
+
+namespace platform {
+    class OptionsMenu {
+    public:
+        OptionsMenu(std::vector<std::tuple<std::string, char, bool>>& options, std::string color_normal, std::string color_bold, int cols) : options(options), color_normal(color_normal), color_bold(color_bold), cols(cols) {}
+        std::string to_string();
+        std::tuple<char, int, bool> parse(char defaultCommand, int minIndex, int maxIndex);
+        char getCommand() const { return command; };
+        int getIndex() const { return index; };
+        std::string getErrorMessage() const { return errorMessage; };
+        void updateColumns(int cols) { this->cols = cols; }
+    private:
+        std::vector<std::tuple<std::string, char, bool>>& options;
+        std::string color_normal, color_bold;
+        int cols;
+        std::string errorMessage;
+        char command;
+        int index;
+    };
+} /* namespace platform */
+#endif

src/reports/DatasetsConsole.cpp

@@ -1,3 +1,4 @@
+#include <algorithm>
 #include "common/Colors.h"
 #include "common/Datasets.h"
 #include "common/Paths.h"
@@ -12,7 +13,7 @@ namespace platform {
             auto part = temp.substr(0, DatasetsConsole::BALANCE_LENGTH);
             line += part + "\n";
             body.push_back(line);
-            line = string(name_len + 22, ' ');
+            line = string(name_len + 28, ' ');
             temp = temp.substr(DatasetsConsole::BALANCE_LENGTH);
         }
         line += temp + "\n";
@@ -26,8 +27,8 @@ namespace platform {
         std::stringstream sheader;
         auto datasets_names = datasets.getNames();
         int maxName = std::max(size_t(7), (*max_element(datasets_names.begin(), datasets_names.end(), [](const std::string& a, const std::string& b) { return a.size() < b.size(); })).size());
-        std::vector<std::string> header_labels = { " #", "Dataset", "Sampl.", "Feat.", "Cls", "Balance" };
-        std::vector<int> header_lengths = { 3, maxName, 6, 5, 3, DatasetsConsole::BALANCE_LENGTH };
+        std::vector<std::string> header_labels = { " #", "Dataset", "Sampl.", "Feat.", "#Num.", "Cls", "Balance" };
+        std::vector<int> header_lengths = { 3, maxName, 6, 5, 5, 3, DatasetsConsole::BALANCE_LENGTH };
         sheader << Colors::GREEN();
         for (int i = 0; i < header_labels.size(); i++) {
             sheader << setw(header_lengths[i]) << left << header_labels[i] << " ";
@@ -41,30 +42,37 @@ namespace platform {
         sline += "\n";
         header.push_back(sline);
         int num = 0;
-        for (const auto& dataset : datasets.getNames()) {
+        for (const auto& dataset_name : datasets.getNames()) {
             std::stringstream line;
             line.imbue(loc);
             auto color = num % 2 ? Colors::CYAN() : Colors::BLUE();
             line << color << setw(3) << right << num++ << " ";
-            line << setw(maxName) << left << dataset << " ";
-            datasets.loadDataset(dataset);
-            auto nSamples = datasets.getNSamples(dataset);
+            line << setw(maxName) << left << dataset_name << " ";
+            auto& dataset = datasets.getDataset(dataset_name);
+            dataset.load();
+            auto nSamples = dataset.getNSamples();
             line << setw(6) << right << nSamples << " ";
-            line << setw(5) << right << datasets.getFeatures(dataset).size() << " ";
-            line << setw(3) << right << datasets.getNClasses(dataset) << " ";
+            auto nFeatures = dataset.getFeatures().size();
+            line << setw(5) << right << nFeatures << " ";
+            auto numericFeatures = dataset.getNumericFeatures();
+            auto num = std::count(numericFeatures.begin(), numericFeatures.end(), true);
+            line << setw(5) << right << num << " ";
+            auto nClasses = dataset.getNClasses();
+            line << setw(3) << right << nClasses << " ";
             std::string sep = "";
             oss.str("");
-            for (auto number : datasets.getClassesCounts(dataset)) {
+            for (auto number : dataset.getClassesCounts()) {
                 oss << sep << std::setprecision(2) << fixed << (float)number / nSamples * 100.0 << "% (" << number << ")";
                 sep = " / ";
             }
             split_lines(maxName, line.str(), oss.str());
             // Store data for Excel report
-            data[dataset] = json::object();
-            data[dataset]["samples"] = nSamples;
-            data[dataset]["features"] = datasets.getFeatures(dataset).size();
-            data[dataset]["classes"] = datasets.getNClasses(dataset);
-            data[dataset]["balance"] = oss.str();
+            data[dataset_name] = json::object();
+            data[dataset_name]["samples"] = nSamples;
+            data[dataset_name]["features"] = nFeatures;
+            data[dataset_name]["numericFeatures"] = num;
+            data[dataset_name]["classes"] = nClasses;
+            data[dataset_name]["balance"] = oss.str();
         }
     }
 }

src/reports/DatasetsExcel.cpp

@@ -17,11 +17,11 @@ namespace platform {
         int balanceSize = 75; // Min size of the column
         worksheet = workbook_add_worksheet(workbook, "Datasets");
         // Header
-        worksheet_merge_range(worksheet, 0, 0, 0, 5, "Datasets", styles["headerFirst"]);
+        worksheet_merge_range(worksheet, 0, 0, 0, 6, "Datasets", styles["headerFirst"]);
         // Body header
         row = 2;
         int col = 0;
-        for (const auto& name : { "Nº", "Dataset", "Samples", "Features", "Classes", "Balance" }) {
+        for (const auto& name : { "#", "Dataset", "Samples", "Features", "#Numer.", "Classes", "Balance" }) {
             writeString(row, col++, name, "bodyHeader");
         }
         // Body
@@ -34,12 +34,13 @@ namespace platform {
             writeString(row, 1, key.c_str(), "text");
             writeInt(row, 2, value["samples"], "ints");
             writeInt(row, 3, value["features"], "ints");
-            writeInt(row, 4, value["classes"], "ints");
-            writeString(row, 5, value["balance"].get<std::string>().c_str(), "text");
+            writeInt(row, 4, value["numericFeatures"], "ints");
+            writeInt(row, 5, value["classes"], "ints");
+            writeString(row, 6, value["balance"].get<std::string>().c_str(), "text");
         }
         // Format columns
         worksheet_freeze_panes(worksheet, 3, 2);
-        std::vector<int> columns_sizes = { 5, datasetNameSize, 10, 10, 10, balanceSize };
+        std::vector<int> columns_sizes = { 5, datasetNameSize, 10, 10, 10, 10, balanceSize };
         for (int i = 0; i < columns_sizes.size(); ++i) {
             worksheet_set_column(worksheet, i, i, columns_sizes.at(i), NULL);
         }

src/reports/ExcelFile.cpp

@@ -22,6 +22,27 @@ namespace platform {
         colorOdd = 0xDCE6F1;
         colorEven = 0xFDE9D9;
     }
+    lxw_worksheet* ExcelFile::createWorksheet(const std::string& name)
+    {
+        lxw_worksheet* sheet;
+        std::string suffix = "";
+        std::string efectiveName;
+        int num = 1;
+        // Create a sheet with the name of the model
+        while (true) {
+            efectiveName = name + suffix;
+            if (workbook_get_worksheet_by_name(workbook, efectiveName.c_str())) {
+                suffix = std::to_string(++num);
+            } else {
+                sheet = workbook_add_worksheet(workbook, efectiveName.c_str());
+                break;
+            }
+            if (num > 100) {
+                throw std::invalid_argument("Couldn't create sheet " + efectiveName);
+            }
+        }
+        return sheet;
+    }
 
     lxw_workbook* ExcelFile::getWorkbook()
     {
@@ -75,7 +96,7 @@ namespace platform {
     }
     void ExcelFile::boldGreen()
     {
-        boldFontColor(0x00FF00);
+        boldFontColor(0x009900);
     }
     void ExcelFile::boldRed()
     {

src/reports/ExcelFile.h

@@ -24,6 +24,7 @@ namespace platform {
         void boldBlue(); //set blue color for the bold styles
         void boldGreen(); //set green color for the bold styles
         void createStyle(const std::string& name, lxw_format* style, bool odd);
+        lxw_worksheet* createWorksheet(const std::string& name);
         void addColor(lxw_format* style, bool odd);
         lxw_format* efectiveStyle(const std::string& name);
         lxw_workbook* workbook;

src/reports/ReportBase.cpp

@@ -61,12 +61,13 @@ namespace platform {
             }
         } else {
             if (data["score_name"].get<std::string>() == "accuracy") {
-                auto dt = Datasets(false, Paths::datasets());
-                dt.loadDataset(dataset);
-                auto numClasses = dt.getNClasses(dataset);
+                auto datasets = Datasets(false, Paths::datasets());
+                auto& dt = datasets.getDataset(dataset);
+                dt.load();
+                auto numClasses = dt.getNClasses();
                 if (numClasses == 2) {
-                    std::vector<int> distribution = dt.getClassesCounts(dataset);
-                    double nSamples = dt.getNSamples(dataset);
+                    std::vector<int> distribution = dt.getClassesCounts();
+                    double nSamples = dt.getNSamples();
                     std::vector<int>::iterator maxValue = max_element(distribution.begin(), distribution.end());
                     double mark = *maxValue / nSamples * (1 + margin);
                     if (mark > 1) {

src/reports/ReportConsole.cpp

@@ -2,6 +2,7 @@
 #include <locale>
 #include "best/BestScore.h"
 #include "common/CLocale.h"
+#include "common/Timer.h"
 #include "ReportConsole.h"
 #include "main/Scores.h"
 
@@ -23,12 +24,30 @@ namespace platform {
             + " random seeds. " + data["date"].get<std::string>() + " " + data["time"].get<std::string>()
         );
         sheader << headerLine(data["title"].get<std::string>());
+        std::string discretize_algo = data.find("discretization_algorithm") != data.end() ? data["discretization_algorithm"].get<std::string>() : "ORIGINAL";
+        std::string algorithm = data["discretized"].get<bool>() ? " (" + discretize_algo + ")" : "";
+        std::string smooth = data.find("smooth_strategy") != data.end() ? data["smooth_strategy"].get<std::string>() : "ORIGINAL";
+        std::string stratified;
+        try {
+            stratified = data["stratified"].get<bool>() ? "True" : "False";
+        }
+        catch (nlohmann::json::type_error) {
+            stratified = data["stratified"].get<int>() == 1 ? "True" : "False";
+        }
+        std::string discretized;
+        try {
+            discretized = data["discretized"].get<bool>() ? "True" : "False";
+        }
+        catch (nlohmann::json::type_error) {
+            discretized = data["discretized"].get<int>() == 1 ? "True" : "False";
+        }
         sheader << headerLine(
-            "Random seeds: " + fromVector("seeds") + " Discretized: " + (data["discretized"].get<bool>() ? "True" : "False")
-            + " Stratified: " + (data["stratified"].get<bool>() ? "True" : "False")
+            "Random seeds: " + fromVector("seeds") + " Discretized: " + discretized + " " + algorithm
+            + " Stratified: " + stratified + " Smooth Strategy: " + smooth
         );
-        oss << "Execution took  " << std::setprecision(2) << std::fixed << data["duration"].get<float>()
-            << " seconds,   " << data["duration"].get<float>() / 3600 << " hours, on " << data["platform"].get<std::string>();
+        Timer timer;
+        oss << "Execution took  " << timer.translate2String(data["duration"].get<float>())
+            << " on " << data["platform"].get<std::string>() << " Language: " << data["language"].get<std::string>();
         sheader << headerLine(oss.str());
         sheader << headerLine("Score is " + data["score_name"].get<std::string>());
         sheader << std::string(MAXL, '*') << std::endl;
@@ -121,12 +140,35 @@ namespace platform {
                     }
                 }
             }
+            line.str("");
+            if (lastResult.find("score_train") == lastResult.end()) {
+                line << headerLine("Train  score: -");
+            } else {
+                line << headerLine("Train  score: " + std::to_string(lastResult["score_train"].get<double>()));
+            }
+            vbody.push_back(line.str()); sbody << line.str();
             line.str(""); line << headerLine(fVector("Train scores: ", lastResult["scores_train"], 14, 12));
             vbody.push_back(line.str()); sbody << line.str();
+            line.str(""); line << headerLine("Test   score: " + std::to_string(lastResult["score"].get<double>()));
+            vbody.push_back(line.str()); sbody << line.str();
             line.str(""); line << headerLine(fVector("Test  scores: ", lastResult["scores_test"], 14, 12));
             vbody.push_back(line.str()); sbody << line.str();
+            line.str("");
+            if (lastResult.find("train_time") == lastResult.end()) {
+                line << headerLine("Train  time: -");
+            } else {
+                line << headerLine("Train  time: " + std::to_string(lastResult["train_time"].get<double>()));
+            }
+            vbody.push_back(line.str()); sbody << line.str();
             line.str(""); line << headerLine(fVector("Train  times: ", lastResult["times_train"], 10, 3));
             vbody.push_back(line.str()); sbody << line.str();
+            line.str("");
+            if (lastResult.find("test_time") == lastResult.end()) {
+                line << headerLine("Test  time: -");
+            } else {
+                line << headerLine("Test  time: " + std::to_string(lastResult["test_time"].get<double>()));
+            }
+            vbody.push_back(line.str()); sbody << line.str();
             line.str(""); line << headerLine(fVector("Test   times: ", lastResult["times_test"], 10, 3));
             vbody.push_back(line.str()); sbody << line.str();
 

src/reports/ReportExcel.cpp

@@ -17,26 +17,7 @@ namespace platform {
             worksheet_set_column(worksheet, i, i, columns_sizes.at(i), NULL);
         }
     }
-    void ReportExcel::createWorksheet()
-    {
-        const std::string name = data["model"].get<std::string>();
-        std::string suffix = "";
-        std::string efectiveName;
-        int num = 1;
-        // Create a sheet with the name of the model
-        while (true) {
-            efectiveName = name + suffix;
-            if (workbook_get_worksheet_by_name(workbook, efectiveName.c_str())) {
-                suffix = std::to_string(++num);
-            } else {
-                worksheet = workbook_add_worksheet(workbook, efectiveName.c_str());
-                break;
-            }
-            if (num > 100) {
-                throw std::invalid_argument("Couldn't create sheet " + efectiveName);
-            }
-        }
-    }
+
 
     void ReportExcel::createFile()
     {
@@ -44,7 +25,8 @@ namespace platform {
             workbook = workbook_new((Paths::excel() + Paths::excelResults()).c_str());
         }
         if (worksheet == NULL) {
-            createWorksheet();
+            const std::string name = data["model"].get<std::string>();
+            worksheet = createWorksheet(name);
         }
         setProperties(data["title"].get<std::string>());
         formatColumns();
@@ -67,7 +49,10 @@ namespace platform {
         worksheet_merge_range(worksheet, 0, 0, 0, 12, message.c_str(), styles["headerFirst"]);
         worksheet_merge_range(worksheet, 1, 0, 1, 12, data["title"].get<std::string>().c_str(), styles["headerRest"]);
         worksheet_merge_range(worksheet, 2, 0, 3, 0, ("Score is " + data["score_name"].get<std::string>()).c_str(), styles["headerRest"]);
-        worksheet_merge_range(worksheet, 2, 1, 3, 3, "Execution time", styles["headerRest"]);
+        writeString(2, 1, "Smooth", "headerRest");
+        std::string smooth = data.find("smooth_strategy") != data.end() ? data["smooth_strategy"].get<std::string>() : "ORIGINAL";
+        writeString(3, 1, smooth, "headerSmall");
+        worksheet_merge_range(worksheet, 2, 2, 3, 3, "Execution time", styles["headerRest"]);
         oss << std::setprecision(2) << std::fixed << data["duration"].get<float>() << " s";
         worksheet_merge_range(worksheet, 2, 4, 2, 5, oss.str().c_str(), styles["headerRest"]);
         oss.str("");
@@ -83,7 +68,9 @@ namespace platform {
         worksheet_merge_range(worksheet, 3, 10, 3, 11, oss.str().c_str(), styles["headerSmall"]);
         oss.str("");
         oss.clear();
-        oss << "Discretized: " << (data["discretized"].get<bool>() ? "True" : "False");
+        std::string discretize_algo = data.find("discretization_algorithm") != data.end() ? data["discretization_algorithm"].get<std::string>() : "mdlp";
+        std::string algorithm = data["discretized"].get<bool>() ? " (" + discretize_algo + ")" : "";
+        oss << "Discretized: " << (data["discretized"].get<bool>() ? "True" : "False") << algorithm;
         worksheet_write_string(worksheet, 3, 12, oss.str().c_str(), styles["headerSmall"]);
     }
     void ReportExcel::header_notes(int row)
@@ -195,15 +182,7 @@ namespace platform {
             }
             // Classificacion report
             if (lastResult.find("confusion_matrices") != lastResult.end()) {
-                // auto score = platform2::Scores::create_aggregate(lastResult, "confusion_matrices");
-                // row++;
-                // writeString(row, 1, "Classification Report", "bodyHeader");
-                // row++;
-                // auto output = platform2::Scores::classification_report("", "test");
-                // for (const auto& item : output) {
-                //     writeString(row, 1, item, "text");
-                //     row++;
-                // }
+                create_classification_report(lastResult);
             }
             // Set with of columns to show those totals completely
             worksheet_set_column(worksheet, 1, 1, 12, NULL);
@@ -215,7 +194,129 @@ namespace platform {
             footer(totalScore, row);
         }
     }
+    void ReportExcel::create_classification_report(const json& result)
+    {
 
+        auto matrix_sheet = createWorksheet("clf_report");
+        lxw_worksheet* tmp = worksheet;
+        worksheet = matrix_sheet;
+        if (matrix_sheet == NULL) {
+            throw std::invalid_argument("Couldn't create sheet classif_report");
+        }
+        row = 1;
+        int col = 0;
+        if (result.find("confusion_matrices_train") != result.end()) {
+            // Train classification report
+            auto score = Scores::create_aggregate(result, "confusion_matrices_train");
+            auto train = score.classification_report_json("Train");
+            std::tie(row, col) = write_classification_report(train, row, 0);
+            int new_row = 0;
+            int new_col = col + 1;
+            for (int i = 0; i < result["confusion_matrices_train"].size(); ++i) {
+                auto item = result["confusion_matrices_train"][i];
+                auto score_item = Scores(item);
+                auto title = "Train Fold " + std::to_string(i);
+                std::tie(new_row, new_col) = write_classification_report(score_item.classification_report_json(title), 1, new_col);
+                new_col++;
+            }
+            col = new_col;
+            worksheet_merge_range(matrix_sheet, 0, 0, 0, col - 1, "Train Classification Report", efectiveStyle("headerRest"));
+        }
+        // Test classification report
+        worksheet_merge_range(matrix_sheet, row, 0, row, col - 1, "Test Classification Report", efectiveStyle("headerRest"));
+        auto score = Scores::create_aggregate(result, "confusion_matrices");
+        auto test = score.classification_report_json("Test");
+        int init_row = ++row;
+        std::tie(row, col) = write_classification_report(test, init_row, 0);
+        int new_row = 0;
+        int new_col = col + 1;
+        for (int i = 0; i < result["confusion_matrices"].size(); ++i) {
+            auto item = result["confusion_matrices"][i];
+            auto score_item = Scores(item);
+            auto title = "Test Fold " + std::to_string(i);
+            std::tie(new_row, new_col) = write_classification_report(score_item.classification_report_json(title), init_row, new_col);
+            new_col++;
+        }
+        // Format columns (change size to fit the content)
+        for (int i = 0; i < new_col; ++i) {
+            // doesn't work with from col to col, so...
+            worksheet_set_column(worksheet, i, i, 12, NULL);
+        }
+        worksheet = tmp;
+    }
+    std::pair<int, int> ReportExcel::write_classification_report(const json& result, int init_row, int init_col)
+    {
+        row = init_row;
+        auto text = result["title"].get<std::string>();
+        worksheet_merge_range(worksheet, row, init_col, row + 1, init_col + 5, text.c_str(), efectiveStyle("bodyHeader"));
+        row += 2;
+        int col = init_col + 2;
+        // Headers
+        bool first_item = true;
+        for (const auto& item : result["headers"]) {
+            auto text = item.get<std::string>();
+            if (first_item) {
+                first_item = false;
+                worksheet_merge_range(worksheet, row, init_col, row, init_col + 1, text.c_str(), efectiveStyle("bodyHeader"));
+            } else {
+                writeString(row, col++, text, "bodyHeader");
+            }
+        }
+        row++;
+        // Classes f1-score
+        for (const auto& item : result["body"]) {
+            col = init_col + 2;
+            for (const auto& value : item) {
+                if (value.is_string()) {
+                    worksheet_merge_range(worksheet, row, init_col, row, init_col + 1, value.get<std::string>().c_str(), efectiveStyle("text"));
+                } else {
+                    if (value.is_number_integer()) {
+                        writeInt(row, col++, value.get<int>(), "ints");
+                    } else {
+                        writeDouble(row, col++, value.get<double>(), "result");
+                    }
+                }
+            }
+            row++;
+        }
+        // Accuracy and average f1-score
+        for (const auto& item : { "accuracy", "averages", "weighted" }) {
+            col = init_col + 2;
+            for (const auto& value : result[item]) {
+                if (value.is_string()) {
+                    worksheet_merge_range(worksheet, row, init_col, row, init_col + 1, value.get<std::string>().c_str(), efectiveStyle("text"));
+                } else {
+                    if (value.is_number_integer()) {
+                        writeInt(row, col++, value.get<int>(), "ints");
+                    } else {
+                        writeDouble(row, col++, value.get<double>(), "result");
+                    }
+                }
+            }
+            row++;
+        }
+        // Confusion matrix
+        auto n_items = result["confusion_matrix"].size();
+        worksheet_merge_range(worksheet, row, init_col, row, init_col + n_items + 1, "Confusion Matrix", efectiveStyle("bodyHeader"));
+        row++;
+        boldGreen();
+        for (int i = 0; i < n_items; ++i) {
+            col = init_col + 2;
+            auto label = result["body"][i][0].get<std::string>();
+            worksheet_merge_range(worksheet, row, init_col, row, init_col + 1, label.c_str(), efectiveStyle("text"));
+            for (int j = 0; j < result["confusion_matrix"][i].size(); ++j) {
+                auto value = result["confusion_matrix"][i][j];
+                if (i == j) {
+                    writeInt(row, col++, value.get<int>(), "ints_bold");
+                } else {
+                    writeInt(row, col++, value.get<int>(), "ints");
+                }
+            }
+            row++;
+        }
+        int maxcol = std::max(init_col + 5, int(init_col + n_items + 1));
+        return { row, maxcol };
+    }
     void ReportExcel::showSummary()
     {
         for (const auto& item : summary) {
@@ -225,7 +326,6 @@ namespace platform {
             row += 1;
         }
     }
-
     void ReportExcel::footer(double totalScore, int row)
     {
         showSummary();

src/reports/ReportExcel.h

@@ -1,5 +1,7 @@
 #ifndef REPORT_EXCEL_H
 #define REPORT_EXCEL_H
+#include <algorithm>
+#include "main/Scores.h"
 #include "common/Colors.h"
 #include "ReportBase.h"
 #include "ExcelFile.h"
@@ -13,12 +15,13 @@ namespace platform {
     private:
         void formatColumns();
         void createFile();
-        void createWorksheet();
         void header() override;
         void body() override;
         void showSummary() override;
         void footer(double totalScore, int row);
         void append_notes(const json& r, int row);
+        void create_classification_report(const json& result);
+        std::pair<int, int> write_classification_report(const json& result, int init_row, int init_col);
         void header_notes(int row);
     };
 };

src/results/Result.cpp

@@ -6,6 +6,7 @@
 #include "common/DotEnv.h"
 #include "common/CLocale.h"
 #include "common/Paths.h"
+#include "common/Symbols.h"
 #include "Result.h"
 
 namespace platform {
@@ -64,27 +65,46 @@ namespace platform {
 
     void Result::save()
     {
-        std::ofstream file(Paths::results() + "/" + getFilename());
+        std::ofstream file(Paths::results() + getFilename());
         file << data;
         file.close();
     }
     std::string Result::getFilename() const
     {
         std::ostringstream oss;
-        oss << "results_" << data.at("score_name").get<std::string>() << "_" << data.at("model").get<std::string>() << "_"
-            << data.at("platform").get<std::string>() << "_" << data["date"].get<std::string>() << "_"
-            << data["time"].get<std::string>() << "_" << (data["stratified"] ? "1" : "0") << ".json";
+        std::string stratified;
+        try {
+            stratified = data["stratified"].get<bool>() ? "1" : "0";
+        }
+        catch (nlohmann::json_abi_v3_11_3::detail::type_error) {
+            stratified = data["stratified"].get<int>() == 1 ? "1" : "0";
+        }
+        oss << "results_"
+            << data.at("score_name").get<std::string>() << "_"
+            << data.at("model").get<std::string>() << "_"
+            << data.at("platform").get<std::string>() << "_"
+            << data["date"].get<std::string>() << "_"
+            << data["time"].get<std::string>() << "_"
+            << stratified << ".json";
         return oss.str();
     }
-
-
-    std::string Result::to_string(int maxModel) const
+    std::string Result::to_string(int maxModel, int maxTitle) const
     {
         auto tmp = ConfigLocale();
         std::stringstream oss;
-        std::string s = data["stratified"].get<bool>() ? "S" : "";
-        std::string d = data["discretized"].get<bool>() ? "D" : "";
-        std::string sd = s + d;
+        std::string s, d;
+        try {
+            s = data["stratified"].get<bool>() ? "S" : " ";
+        }
+        catch (nlohmann::json_abi_v3_11_3::detail::type_error) {
+            s = data["stratified"].get<int>() == 1 ? "S" : " ";
+        }
+        try {
+            d = data["discretized"].get<bool>() ? "D" : " ";
+        }
+        catch (nlohmann::json_abi_v3_11_3::detail::type_error) {
+            d = data["discretized"].get<int>() == 1 ? "D" : " ";
+        }
         auto duration = data["duration"].get<double>();
         double durationShow = duration > 3600 ? duration / 3600 : duration > 60 ? duration / 60 : duration;
         std::string durationUnit = duration > 3600 ? "h" : duration > 60 ? "m" : "s";
@@ -93,11 +113,15 @@ namespace platform {
         oss << std::setw(11) << std::left << data["score_name"].get<std::string>() << " ";
         oss << std::right << std::setw(10) << std::setprecision(7) << std::fixed << score << " ";
         oss << std::left << std::setw(12) << data["platform"].get<std::string>() << " ";
-        oss << std::left << std::setw(2) << sd << " ";
+        oss << s << d << " ";
         auto completeString = isComplete() ? "C" : "P";
         oss << std::setw(1) << " " << completeString << "  ";
         oss << std::setw(5) << std::right << std::setprecision(2) << std::fixed << durationShow << " " << durationUnit << " ";
-        oss << std::setw(50) << std::left << data["title"].get<std::string>() << " ";
+        auto title = data["title"].get<std::string>();
+        if (title.size() > maxTitle) {
+            title = title.substr(0, maxTitle - 1) + Symbols::ellipsis;
+        }
+        oss << std::setw(maxTitle) << std::left << title;
         return  oss.str();
     }
 }

src/results/Result.h

@@ -18,7 +18,7 @@ namespace platform {
         void save();
         // Getters
         json getJson();
-        std::string to_string(int maxModel) const;
+        std::string to_string(int maxModel, int maxTitle) const;
         std::string getFilename() const;
         std::string getDate() const { return data["date"].get<std::string>(); };
         std::string getTime() const { return data["time"].get<std::string>(); };
@@ -28,10 +28,13 @@ namespace platform {
         std::string getModel() const { return data["model"].get<std::string>(); };
         std::string getPlatform() const { return data["platform"].get<std::string>(); };
         std::string getScoreName() const { return data["score_name"].get<std::string>(); };
+
         bool isComplete() const { return complete; };
         json getData() const { return data; }
         // Setters
         void setTitle(const std::string& title) { data["title"] = title; };
+        void setSmoothStrategy(const std::string& smooth_strategy) { data["smooth_strategy"] = smooth_strategy; };
+        void setDiscretizationAlgorithm(const std::string& discretization_algo) { data["discretization_algorithm"] = discretization_algo; };
         void setLanguage(const std::string& language) { data["language"] = language; };
         void setLanguageVersion(const std::string& language_version) { data["language_version"] = language_version; };
         void setDuration(double duration) { data["duration"] = duration; };
@@ -44,7 +47,6 @@ namespace platform {
         void setStratified(bool stratified) { data["stratified"] = stratified; };
         void setNFolds(int nfolds) { data["folds"] = nfolds; };
         void setPlatform(const std::string& platform_name) { data["platform"] = platform_name; };
-
     private:
         json data;
         bool complete;

src/results/ResultsDatasetExcel.cpp

@@ -20,7 +20,7 @@ namespace platform {
         // Body header
         row = 2;
         int col = 0;
-        for (const auto& name : { "Nº", "Model", "Date", "Time", "Score", "Hyperparameters" }) {
+        for (const auto& name : { "#", "Model", "Date", "Time", "Score", "Hyperparameters" }) {
             writeString(row, col++, name, "bodyHeader");
         }
         // Body

tests/CMakeLists.txt

@@ -3,7 +3,7 @@ if(ENABLE_TESTING)
     include_directories(
         ${Platform_SOURCE_DIR}/src
         ${Platform_SOURCE_DIR}/lib/argparse/include
-        ${Platform_SOURCE_DIR}/lib/mdlp
+        ${Platform_SOURCE_DIR}/lib/mdlp/src
         ${Platform_SOURCE_DIR}/lib/Files
         ${Platform_SOURCE_DIR}/lib/json/include
         ${Platform_SOURCE_DIR}/lib/folding
@@ -13,10 +13,10 @@ if(ENABLE_TESTING)
     )
     set(TEST_SOURCES_PLATFORM 
         TestUtils.cpp TestPlatform.cpp TestResult.cpp TestScores.cpp 
-        ${Platform_SOURCE_DIR}/src/common/Datasets.cpp ${Platform_SOURCE_DIR}/src/common/Dataset.cpp
+        ${Platform_SOURCE_DIR}/src/common/Datasets.cpp ${Platform_SOURCE_DIR}/src/common/Dataset.cpp ${Platform_SOURCE_DIR}/src/common/Discretization.cpp
         ${Platform_SOURCE_DIR}/src/main/Scores.cpp
     )
     add_executable(${TEST_PLATFORM} ${TEST_SOURCES_PLATFORM})
-    target_link_libraries(${TEST_PLATFORM} PUBLIC "${TORCH_LIBRARIES}" ArffFiles mdlp Catch2::Catch2WithMain BayesNet)
+    target_link_libraries(${TEST_PLATFORM} PUBLIC "${TORCH_LIBRARIES}" mdlp Catch2::Catch2WithMain BayesNet)
     add_test(NAME ${TEST_PLATFORM} COMMAND ${TEST_PLATFORM})
 endif(ENABLE_TESTING)

tests/TestPlatform.cpp

@@ -7,14 +7,15 @@
 #include <string>
 #include "TestUtils.h"
 #include "folding.hpp"
+#include <ArffFiles.hpp>
 #include <bayesnet/classifiers/TAN.h>
-#include "config.h"
+#include "config_platform.h"
 
 
 TEST_CASE("Test Platform version", "[Platform]")
 {
     std::string version = { platform_project_version.begin(), platform_project_version.end() };
-    REQUIRE(version == "1.0.4");
+    REQUIRE(version == "1.1.0");
 }
 TEST_CASE("Test Folding library version", "[Folding]")
 {
@@ -24,10 +25,15 @@ TEST_CASE("Test Folding library version", "[Folding]")
 TEST_CASE("Test BayesNet version", "[BayesNet]")
 {
     std::string version = bayesnet::TAN().getVersion();
-    REQUIRE(version == "1.0.5.1");
+    REQUIRE(version == "1.0.6");
 }
 TEST_CASE("Test mdlp version", "[mdlp]")
 {
     std::string version = mdlp::CPPFImdlp::version();
-    REQUIRE(version == "1.1.2");
+    REQUIRE(version == "2.0.0");
+}
+TEST_CASE("Test Arff version", "[Arff]")
+{
+    std::string version = ArffFiles().version();
+    REQUIRE(version == "1.1.0");
 }

tests/TestResult.cpp

@@ -7,19 +7,20 @@
 #include "common/DotEnv.h"
 #include "common/Datasets.h"
 #include "common/Paths.h"
-#include "config.h"
+#include "config_platform.h"
 
 
 TEST_CASE("ZeroR comparison in reports", "[Report]")
 {
     auto dotEnv = platform::DotEnv(true);
-    auto margin = 1e-2;
-    std::string dataset = "liver-disorders";
+    auto margin = 1e-4;
+    std::string dataset_name = "liver-disorders";
     auto dt = platform::Datasets(false, platform::Paths::datasets());
-    dt.loadDataset(dataset);
-    std::vector<int> distribution = dt.getClassesCounts(dataset);
-    double nSamples = dt.getNSamples(dataset);
+    auto& dataset = dt.getDataset(dataset_name);
+    dataset.load();
+    std::vector<int> distribution = dataset.getClassesCounts();
+    double nSamples = dataset.getNSamples();
     std::vector<int>::iterator maxValue = max_element(distribution.begin(), distribution.end());
     double mark = *maxValue / nSamples * (1 + margin);
-    REQUIRE(mark == Catch::Approx(0.585507f).epsilon(1e-5));
+    REQUIRE(mark == Catch::Approx(0.57976811f).epsilon(margin));
 }

tests/TestScores.cpp

@@ -9,7 +9,7 @@
 #include "common/Paths.h"
 #include "common/Colors.h"
 #include "main/Scores.h"
-#include "config.h"
+#include "config_platform.h"
 
 using json = nlohmann::ordered_json;
 auto epsilon = 1e-4;
@@ -128,7 +128,7 @@ TEST_CASE("Confusion Matrix JSON", "[Scores]")
     REQUIRE(res_json_str["Car"][1] == 2);
     REQUIRE(res_json_str["Car"][2] == 3);
 }
-TEST_CASE("Classification Report", "[Scores]") -
+TEST_CASE("Classification Report", "[Scores]")
 {
     std::vector<int> y_test = { 0, 2, 2, 2, 2, 0, 1, 2, 0, 2 };
     std::vector<int> y_pred = { 0, 1, 2, 2, 1, 1, 1, 0, 0, 2 };

tests/TestUtils.cpp

@@ -1,5 +1,5 @@
 #include "TestUtils.h"
-#include "config.h"
+#include "config_platform.h"
 
 class Paths {
 public:

tests/TestUtils.h

@@ -5,8 +5,8 @@
 #include <vector>
 #include <map>
 #include <tuple>
-#include "ArffFiles.h"
-#include "CPPFImdlp.h"
+#include <ArffFiles.hpp>
+#include <fimdlp/CPPFImdlp.h>
 
 bool file_exists(const std::string& name);
 std::pair<vector<mdlp::labels_t>, map<std::string, int>> discretize(std::vector<mdlp::samples_t>& X, mdlp::labels_t& y, std::vector<string> features);

tests/data/all.txt

@@ -1,8 +1,8 @@
-diabetes,class, all
-ecoli,class, all
-glass,Type, all
-iris,class, all
-kdd_JapaneseVowels,speaker, [2,3,4,5,6,7,8,9,10,11,12,13]
-letter,class, all
-liver-disorders,selector, all
-mfeat-factors,class, all
+diabetes;class;all
+ecoli;class;all
+glass;Type,all
+iris;class;all
+kdd_JapaneseVowels;speaker;[2,3,4,5,6,7,8,9,10,11,12,13]
+letter;class;all
+liver-disorders;selector;all
+mfeat-factors;class;all