Añade AdaBoost and tests

CMakeLists.txt

@@ -68,8 +68,8 @@ endif (ENABLE_CLANG_TIDY)
 # --------------------------------------------------
 # find_library(BayesNet NAMES libBayesNet BayesNet libBayesNet.a PATHS ${PyClassifiers_SOURCE_DIR}/../lib/lib REQUIRED)
 # find_path(Bayesnet_INCLUDE_DIRS REQUIRED NAMES bayesnet PATHS ${PyClassifiers_SOURCE_DIR}/../lib/include)
-find_library(bayesnet NAMES libbayesnet bayesnet libbayesnet.a PATHS ${Platform_SOURCE_DIR}/../lib/lib REQUIRED)
-find_path(Bayesnet_INCLUDE_DIRS REQUIRED NAMES bayesnet PATHS ${Platform_SOURCE_DIR}/../lib/include)
+find_library(bayesnet NAMES libbayesnet bayesnet libbayesnet.a PATHS ../lib REQUIRED)
+find_path(Bayesnet_INCLUDE_DIRS REQUIRED NAMES bayesnet PATHS ../lib/include)
 message(STATUS "BayesNet=${bayesnet}")
 message(STATUS "Bayesnet_INCLUDE_DIRS=${Bayesnet_INCLUDE_DIRS}")
 
@@ -82,9 +82,8 @@ add_subdirectory(pyclfs)
 # -------
 if (ENABLE_TESTING)
   MESSAGE("Testing enabled")
-  add_git_submodule(lib/catch2)
-  add_git_submodule(lib/mdlp)
-  add_subdirectory(lib/Files)
+  find_package(Catch2 CONFIG REQUIRED)
+  find_package(arff-files CONFIG REQUIRED)
   include(CTest)
   add_subdirectory(tests)
 endif (ENABLE_TESTING)
@@ -96,4 +95,4 @@ install(TARGETS PyClassifiers
         LIBRARY DESTINATION lib
         CONFIGURATIONS Release)
 install(DIRECTORY pyclfs/ DESTINATION include/pyclassifiers FILES_MATCHING CONFIGURATIONS Release PATTERN "*.h" PATTERN "*.hpp")
-install(FILES ${Bayesnet_INCLUDE_DIRS}/bayesnet/config.h DESTINATION include/pyclassifiers CONFIGURATIONS Release)
+install(FILES ${Bayesnet_INCLUDE_DIRS}/bayesnet/config.h DESTINATION include/pyclassifiers CONFIGURATIONS Release)

Makefile

@@ -52,7 +52,7 @@ debug: ## Build a debug version of the project
 	@echo ">>> Building Debug PyClassifiers...";
 	@if [ -d ./$(f_debug) ]; then rm -rf ./$(f_debug); fi
 	@mkdir $(f_debug); 
-	@cmake -S . -B $(f_debug) -D CMAKE_BUILD_TYPE=Debug -D ENABLE_TESTING=ON -D CODE_COVERAGE=ON
+	@cmake -S . -B $(f_debug) -D CMAKE_BUILD_TYPE=Debug -D ENABLE_TESTING=ON -D CODE_COVERAGE=ON -DCMAKE_TOOLCHAIN_FILE=${VCPKG_ROOT}/scripts/buildsystems/vcpkg.cmake
 	@echo ">>> Done";
 
 release: ## Build a Release version of the project

pyclfs/AdaBoost.cc

@@ -0,0 +1,20 @@
+#include "AdaBoost.h"
+
+namespace pywrap {
+    AdaBoost::AdaBoost() : PyClassifier("sklearn.ensemble", "AdaBoostClassifier", true)
+    {
+        validHyperparameters = { "n_estimators", "n_jobs", "random_state" };
+    }
+    int AdaBoost::getNumberOfEdges() const
+    {
+        return callMethodSumOfItems("get_n_leaves");
+    }
+    int AdaBoost::getNumberOfStates() const
+    {
+        return callMethodSumOfItems("get_depth");
+    }
+    int AdaBoost::getNumberOfNodes() const
+    {
+        return callMethodSumOfItems("node_count");
+    }
+} /* namespace pywrap */

pyclfs/AdaBoost.h

@@ -0,0 +1,15 @@
+#ifndef ADABOOST_H
+#define ADABOOST_H
+#include "PyClassifier.h"
+
+namespace pywrap {
+    class AdaBoost : public PyClassifier {
+    public:
+        AdaBoost();
+        ~AdaBoost() = default;
+        int getNumberOfEdges() const override;
+        int getNumberOfStates() const override;
+        int getNumberOfNodes() const override;
+    };
+} /* namespace pywrap */
+#endif /* ADABOOST_H */

pyclfs/CMakeLists.txt

@@ -4,5 +4,5 @@ include_directories(
     ${PyClassifiers_SOURCE_DIR}/lib/json/include
     ${Bayesnet_INCLUDE_DIRS}
 )
-add_library(PyClassifiers ODTE.cc STree.cc SVC.cc RandomForest.cc XGBoost.cc PyClassifier.cc PyWrap.cc)
+add_library(PyClassifiers ODTE.cc STree.cc SVC.cc RandomForest.cc XGBoost.cc AdaBoost.cc PyClassifier.cc PyWrap.cc)
 target_link_libraries(PyClassifiers nlohmann_json::nlohmann_json ${Python3_LIBRARIES} "${TORCH_LIBRARIES}" ${LIBTORCH_PYTHON} Boost::boost Boost::python Boost::numpy)

tests/CMakeLists.txt

@@ -2,9 +2,6 @@ if(ENABLE_TESTING)
     set(TEST_PYCLASSIFIERS "unit_tests_pyclassifiers")
     include_directories(
         ${PyClassifiers_SOURCE_DIR}
-        ${PyClassifiers_SOURCE_DIR}/lib/Files
-        ${PyClassifiers_SOURCE_DIR}/lib/mdlp/src
-        ${PyClassifiers_SOURCE_DIR}/lib/json/include
         ${Python3_INCLUDE_DIRS}
         ${TORCH_INCLUDE_DIRS}
         ${CMAKE_BINARY_DIR}/configured_files/include
@@ -13,5 +10,5 @@ if(ENABLE_TESTING)
     file(GLOB_RECURSE PyClassifiers_SOURCES "${PyClassifiers_SOURCE_DIR}/pyclfs/*.cc")
     set(TEST_SOURCES_PYCLASSIFIERS TestPythonClassifiers.cc TestUtils.cc ${PyClassifiers_SOURCES})
     add_executable(${TEST_PYCLASSIFIERS} ${TEST_SOURCES_PYCLASSIFIERS})
-    target_link_libraries(${TEST_PYCLASSIFIERS} PUBLIC "${TORCH_LIBRARIES}" ${Python3_LIBRARIES} ${LIBTORCH_PYTHON} Boost::boost Boost::python Boost::numpy ArffFiles fimdlp Catch2::Catch2WithMain)
+    target_link_libraries(${TEST_PYCLASSIFIERS} PUBLIC "${TORCH_LIBRARIES}" ${Python3_LIBRARIES} ${LIBTORCH_PYTHON} Boost::boost Boost::python Boost::numpy fimdlp Catch2::Catch2WithMain)
 endif(ENABLE_TESTING)

tests/TestPythonClassifiers.cc

@@ -10,14 +10,16 @@
 #include "pyclfs/SVC.h"
 #include "pyclfs/RandomForest.h"
 #include "pyclfs/XGBoost.h"
+#include "pyclfs/AdaBoost.h"
 #include "pyclfs/ODTE.h"
 #include "TestUtils.h"
+#include <iostream>
 
 TEST_CASE("Test Python Classifiers score", "[PyClassifiers]")
 {
     map <pair<std::string, std::string>, float> scores = {
         // Diabetes
-        {{"diabetes", "STree"}, 0.81641}, {{"diabetes", "ODTE"}, 0.854166687}, {{"diabetes", "SVC"}, 0.76823}, {{"diabetes", "RandomForest"}, 1.0},
+        {{"diabetes", "STree"}, 0.81641}, {{"diabetes", "ODTE"}, 0.856770813f}, {{"diabetes", "SVC"}, 0.76823}, {{"diabetes", "RandomForest"}, 1.0},
         // Ecoli
         {{"ecoli", "STree"}, 0.8125}, {{"ecoli", "ODTE"}, 0.875}, {{"ecoli", "SVC"}, 0.89583}, {{"ecoli", "RandomForest"}, 1.0},
         // Glass
@@ -35,8 +37,8 @@ TEST_CASE("Test Python Classifiers score", "[PyClassifiers]")
     map<std::string, std::string> versions = {
         {"ODTE", "1.0.0-1"},
         {"STree", "1.4.0"},
-        {"SVC", "1.5.1"},
-        {"RandomForest", "1.5.1"}
+        {"SVC", "1.5.2"},
+        {"RandomForest", "1.5.2"}
     };
     auto clf = models[name];
 
@@ -58,6 +60,15 @@ TEST_CASE("Test Python Classifiers score", "[PyClassifiers]")
         REQUIRE(clf->getVersion() == versions[name]);
     }
 }
+TEST_CASE("AdaBoostClassifier", "[PyClassifiers]")
+{
+    auto raw = RawDatasets("iris", false);
+    auto clf = pywrap::AdaBoost();
+    clf.fit(raw.Xt, raw.yt, raw.featurest, raw.classNamet, raw.statest);
+    clf.setHyperparameters(nlohmann::json::parse("{ \"n_estimators\": 100 }"));
+    auto score = clf.score(raw.Xt, raw.yt);
+    REQUIRE(score == Catch::Approx(0.9599999f).epsilon(raw.epsilon));
+}
 TEST_CASE("Classifiers features", "[PyClassifiers]")
 {
     auto raw = RawDatasets("iris", false);

tests/TestUtils.h

@@ -5,8 +5,8 @@
 #include <vector>
 #include <map>
 #include <tuple>
-#include "ArffFiles.h"
-#include "CPPFImdlp.h"
+#include "ArffFiles/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);

vcpkg.json

@@ -8,7 +8,7 @@
       "libtorch-bin",
       "folding",
       "argparse",
-      "libxlsxwriter"
+      "catch2"
     ],
     "overrides": [
       {
@@ -36,8 +36,8 @@
         "version": "3.2"
       },
       {
-        "name": "libxlsxwriter",
-        "version": "1.2.2"
+        "name": "catch2",
+        "version": "3.8.1"
       },
       {
         "name": "nlohmann-json",