Merge pull request 'vcpkg' (#1) from vcpkg into main

Reviewed-on: #1

.gitmodules

@@ -1,13 +0,0 @@
-
-[submodule "lib/json"]
-	path = lib/json
-	url = https://github.com/nlohmann/json.git
-[submodule "lib/catch2"]
-	path = tests/lib/catch2
-	url = https://github.com/catchorg/Catch2.git
-[submodule "lib/mdlp"]
-	path = tests/lib/mdlp
-	url = https://github.com/rmontanana/mdlp
-[submodule "tests/lib/Files"]
-	path = tests/lib/Files
-	url = https://github.com/rmontanana/ArffFiles

CMakeLists.txt

@@ -45,6 +45,8 @@ endif()
 find_package(Python3 3.11 COMPONENTS Interpreter Development REQUIRED)
 message("Python3_LIBRARIES=${Python3_LIBRARIES}")
 
+find_package(nlohmann_json CONFIG REQUIRED)
+
 # CMakes modules
 # --------------
 set(CMAKE_MODULE_PATH ${CMAKE_CURRENT_SOURCE_DIR}/cmake/modules ${CMAKE_MODULE_PATH})
@@ -64,9 +66,10 @@ endif (ENABLE_CLANG_TIDY)
 
 # External libraries - dependencies of PyClassifiers
 # --------------------------------------------------
-find_library(BayesNet NAMES libBayesNet BayesNet libBayesNet.a PATHS ${PyClassifiers_SOURCE_DIR}/../lib/lib REQUIRED)
+find_library(bayesnet NAMES libbayesnet bayesnet libbayesnet.a PATHS ${PyClassifiers_SOURCE_DIR}/../lib/lib REQUIRED)
-find_path(Bayesnet_INCLUDE_DIRS REQUIRED NAMES bayesnet PATHS ${PyClassifiers_SOURCE_DIR}/../lib/include)
+find_path(Bayesnet_INCLUDE_DIRS REQUIRED NAMES bayesnet PATHS ../lib/include)
-message(STATUS "BayesNet=${BayesNet}")
+
+message(STATUS "BayesNet=${bayesnet}")
 message(STATUS "Bayesnet_INCLUDE_DIRS=${Bayesnet_INCLUDE_DIRS}")
 
 
@@ -78,9 +81,8 @@ add_subdirectory(pyclfs)
 # -------
 if (ENABLE_TESTING)
   MESSAGE("Testing enabled")
-  add_git_submodule(tests/lib/catch2)
+  find_package(Catch2 CONFIG REQUIRED)
-  add_git_submodule(tests/lib/mdlp)
+  find_package(arff-files CONFIG REQUIRED)
-  add_subdirectory(tests/lib/Files)
   include(CTest)
   add_subdirectory(tests)
 endif (ENABLE_TESTING)

Makefile

@@ -52,14 +52,14 @@ 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
 	@echo ">>> Building Release PyClassifiers...";
 	@if [ -d ./$(f_release) ]; then rm -rf ./$(f_release); fi
 	@mkdir $(f_release); 
-	@cmake -S . -B $(f_release) -D CMAKE_BUILD_TYPE=Release
+	@cmake -S . -B $(f_release) -D CMAKE_BUILD_TYPE=Release -DCMAKE_TOOLCHAIN_FILE=${VCPKG_ROOT}/scripts/buildsystems/vcpkg.cmake
 	@echo ">>> Done";	
 
 opt = ""

pyclfs/AdaBoostPy.cc

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

pyclfs/AdaBoostPy.h

@@ -0,0 +1,15 @@
+#ifndef ADABOOSTPY_H
+#define ADABOOSTPY_H
+#include "PyClassifier.h"
+
+namespace pywrap {
+    class AdaBoostPy : public PyClassifier {
+    public:
+        AdaBoostPy();
+        ~AdaBoostPy() = 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 PBC4cip.cc)
+add_library(PyClassifiers ODTE.cc STree.cc SVC.cc RandomForest.cc XGBoost.cc AdaBoostPy.cc PyClassifier.cc PyWrap.cc)
-target_link_libraries(PyClassifiers ${Python3_LIBRARIES} "${TORCH_LIBRARIES}" ${LIBTORCH_PYTHON} Boost::boost Boost::python Boost::numpy)
+target_link_libraries(PyClassifiers nlohmann_json::nlohmann_json ${Python3_LIBRARIES} "${TORCH_LIBRARIES}" ${LIBTORCH_PYTHON} Boost::boost Boost::python Boost::numpy)

pyclfs/PBC4cip.cc

@@ -1,8 +0,0 @@
-#include "PBC4cip.h"
-
-namespace pywrap {
-    PBC4cip::PBC4cip() : PyClassifier("core.PBC4cip", "PBC4cip", true)
-    {
-        validHyperparameters = { "random_state" };
-    }
-} /* namespace pywrap */

pyclfs/PBC4cip.h

@@ -1,13 +0,0 @@
-#ifndef PBC4CIP_H
-#define PBC4CIP_H
-#include "PyClassifier.h"
-
-namespace pywrap {
-    class PBC4cip : public PyClassifier {
-    public:
-        PBC4cip();
-        ~PBC4cip() = default;
-    };
-
-} /* namespace pywrap */
-#endif /* PBC4CIP_H */

pyclfs/PyClassifier.cc

@@ -93,11 +93,19 @@ namespace pywrap {
             PyErr_Print();
             throw std::runtime_error("Error creating object for predict in " + module + " and class " + className);
         }
-        int* data = reinterpret_cast<int*>(prediction.get_data());
+        if (xgboost) {
-        std::vector<int> vPrediction(data, data + prediction.shape(0));
+            long* data = reinterpret_cast<long*>(prediction.get_data());
-        auto resultTensor = torch::tensor(vPrediction, torch::kInt32);
+            std::vector<int> vPrediction(data, data + prediction.shape(0));
-        Py_XDECREF(incoming);
+            auto resultTensor = torch::tensor(vPrediction, torch::kInt32);
-        return resultTensor;
+            Py_XDECREF(incoming);
+            return resultTensor;
+        } else {
+            int* data = reinterpret_cast<int*>(prediction.get_data());
+            std::vector<int> vPrediction(data, data + prediction.shape(0));
+            auto resultTensor = torch::tensor(vPrediction, torch::kInt32);
+            Py_XDECREF(incoming);
+            return resultTensor;
+        }
     }
     torch::Tensor PyClassifier::predict_proba(torch::Tensor& X)
     {
@@ -118,11 +126,19 @@ namespace pywrap {
             PyErr_Print();
             throw std::runtime_error("Error creating object for predict_proba in " + module + " and class " + className);
         }
-        double* data = reinterpret_cast<double*>(prediction.get_data());
+        if (xgboost) {
-        std::vector<double> vPrediction(data, data + prediction.shape(0) * prediction.shape(1));
+            float* data = reinterpret_cast<float*>(prediction.get_data());
-        auto resultTensor = torch::tensor(vPrediction, torch::kFloat64).reshape({ prediction.shape(0), prediction.shape(1) });
+            std::vector<float> vPrediction(data, data + prediction.shape(0) * prediction.shape(1));
-        Py_XDECREF(incoming);
+            auto resultTensor = torch::tensor(vPrediction, torch::kFloat64).reshape({ prediction.shape(0), prediction.shape(1) });
-        return resultTensor;
+            Py_XDECREF(incoming);
+            return resultTensor;
+        } else {
+            double* data = reinterpret_cast<double*>(prediction.get_data());
+            std::vector<double> vPrediction(data, data + prediction.shape(0) * prediction.shape(1));
+            auto resultTensor = torch::tensor(vPrediction, torch::kFloat64).reshape({ prediction.shape(0), prediction.shape(1) });
+            Py_XDECREF(incoming);
+            return resultTensor;
+        }
     }
     float PyClassifier::score(torch::Tensor& X, torch::Tensor& y)
     {

pyclfs/PyClassifier.h

@@ -49,6 +49,7 @@ namespace pywrap {
         nlohmann::json hyperparameters;
         void trainModel(const torch::Tensor& weights, const bayesnet::Smoothing_t smoothing = bayesnet::Smoothing_t::NONE) override {};
         std::vector<std::string> notes;
+        bool xgboost = false;
     private:
         PyWrap* pyWrap;
         std::string module;

pyclfs/XGBoost.cc

@@ -5,5 +5,6 @@ namespace pywrap {
     XGBoost::XGBoost() : PyClassifier("xgboost", "XGBClassifier", true)
     {
         validHyperparameters = { "tree_method", "early_stopping_rounds", "n_jobs" };
+        xgboost = true;
     }
 } /* namespace pywrap */

tests/CMakeLists.txt

@@ -2,15 +2,13 @@ if(ENABLE_TESTING)
     set(TEST_PYCLASSIFIERS "unit_tests_pyclassifiers")
     include_directories(
         ${PyClassifiers_SOURCE_DIR}
-        ${PyClassifiers_SOURCE_DIR}/lib/Files
-        ${PyClassifiers_SOURCE_DIR}/lib/mdlp
-        ${PyClassifiers_SOURCE_DIR}/lib/json/include
         ${Python3_INCLUDE_DIRS}
         ${TORCH_INCLUDE_DIRS}
+        ${CMAKE_BINARY_DIR}/configured_files/include
         /usr/local/include
     )
     file(GLOB_RECURSE PyClassifiers_SOURCES "${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} ${LIBTORCH_PYTHON} Boost::boost Boost::python Boost::numpy ArffFiles mdlp Catch2::Catch2WithMain)
+    target_link_libraries(${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/AdaBoostPy.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
@@ -33,10 +35,10 @@ TEST_CASE("Test Python Classifiers score", "[PyClassifiers]")
         {"RandomForest", new pywrap::RandomForest()}
     };
     map<std::string, std::string> versions = {
-        {"ODTE", "1.0.0"},
+        {"ODTE", "1.0.0-1"},
-        {"STree", "1.3.2"},
+        {"STree", "1.4.0"},
-        {"SVC", "1.5.1"},
+        {"SVC", "1.5.2"},
-        {"RandomForest", "1.5.1"}
+        {"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::AdaBoostPy();
+    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);
@@ -116,33 +127,30 @@ TEST_CASE("XGBoost", "[PyClassifiers]")
     clf.setHyperparameters(hyperparameters);
     auto score = clf.score(raw.Xt, raw.yt);
     REQUIRE(score == Catch::Approx(0.98).epsilon(raw.epsilon));
+    std::cout << "XGBoost score: " << score << std::endl;
 }
-// TEST_CASE("XGBoost predict proba", "[PyClassifiers]")
+TEST_CASE("XGBoost predict proba", "[PyClassifiers]")
-// {
-//     auto raw = RawDatasets("iris", true);
-//     auto clf = pywrap::XGBoost();
-//     clf.fit(raw.Xt, raw.yt, raw.featurest, raw.classNamet, raw.statest);
-//     // nlohmann::json hyperparameters = { "n_jobs=1" };
-//     // clf.setHyperparameters(hyperparameters);
-//     auto predict = clf.predict(raw.Xt);
-//     for (int row = 0; row < predict.size(0); row++) {
-//         auto sum = 0.0;
-//         for (int col = 0; col < predict.size(1); col++) {
-//             std::cout << std::setw(12) << std::setprecision(10) << predict[row][col].item<double>() << " ";
-//             sum += predict[row][col].item<int>();
-//         }
-//         std::cout << std::endl;
-//         // REQUIRE(sum == Catch::Approx(1.0).epsilon(raw.epsilon));
-//     }
-//     std::cout << predict << std::endl;
-// }
-TEST_CASE("PBC4cip", "[PyClassifiers]")
 {
     auto raw = RawDatasets("iris", true);
-    auto clf = pywrap::PBC4cip();
+    auto clf = pywrap::XGBoost();
     clf.fit(raw.Xt, raw.yt, raw.featurest, raw.classNamet, raw.statest);
-    nlohmann::json hyperparameters = { };
+    // nlohmann::json hyperparameters = { "n_jobs=1" };
-    clf.setHyperparameters(hyperparameters);
+    // clf.setHyperparameters(hyperparameters);
-    auto score = clf.score(raw.Xt, raw.yt);
+    auto predict_proba = clf.predict_proba(raw.Xt);
-    REQUIRE(score == Catch::Approx(0.98).epsilon(raw.epsilon));
+    auto predict = clf.predict(raw.Xt);
+    // std::cout << "Predict proba: " << predict_proba << std::endl;
+    // std::cout << "Predict proba size: " << predict_proba.sizes() << std::endl;
+    // assert(predict.size(0) == predict_proba.size(0));
+    for (int row = 0; row < predict_proba.size(0); row++) {
+        // auto sum = 0.0;
+        // std::cout << "Row " << std::setw(3) << row << ": ";
+        // for (int col = 0; col < predict_proba.size(1); col++) {
+        //     std::cout << std::setw(9) << std::fixed << std::setprecision(7) << predict_proba[row][col].item<double>() << " ";
+        //     sum += predict_proba[row][col].item<double>();
+        // }
+        // std::cout << " -> " << std::setw(9) << std::fixed << std::setprecision(7) << sum << " -> " << torch::argmax(predict_proba[row]).item<int>() << " = " << predict[row].item<int>() << std::endl;
+        //     // REQUIRE(sum == Catch::Approx(1.0).epsilon(raw.epsilon));
+        REQUIRE(torch::argmax(predict_proba[row]).item<int>() == predict[row].item<int>());
+        REQUIRE(torch::sum(predict_proba[row]).item<double>() == Catch::Approx(1.0).epsilon(raw.epsilon));
+    }
 }

tests/TestUtils.h

@@ -5,8 +5,8 @@
 #include <vector>
 #include <map>
 #include <tuple>
-#include "ArffFiles.h"
+#include "ArffFiles/ArffFiles.hpp"
-#include "CPPFImdlp.h"
+#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-configuration.json

@@ -0,0 +1,21 @@
+{
+  "default-registry": {
+    "kind": "git",
+    "baseline": "760bfd0c8d7c89ec640aec4df89418b7c2745605",
+    "repository": "https://github.com/microsoft/vcpkg"
+  },
+  "registries": [
+    {
+      "kind": "git",
+      "repository": "https://github.com/rmontanana/vcpkg-stash",
+      "baseline": "1ea69243c0e8b0de77c9d1dd6e1d7593ae7f3627",
+      "packages": [
+        "arff-files",
+        "bayesnet",
+        "fimdlp",
+        "folding",
+        "libtorch-bin"
+      ]
+    }
+  ]
+}

vcpkg.json

@@ -0,0 +1,47 @@
+  {
+    "name": "platform",
+    "version-string": "1.1.0",
+    "dependencies": [
+      "arff-files",
+      "nlohmann-json",
+      "fimdlp",
+      "libtorch-bin",
+      "folding",
+      "argparse",
+      "catch2"
+    ],
+    "overrides": [
+      {
+        "name": "arff-files",
+        "version": "1.1.0"
+      },
+      {
+        "name": "fimdlp",
+        "version": "2.0.1"
+      },
+      {
+        "name": "libtorch-bin",
+        "version": "2.7.0"
+      },
+      {
+        "name": "bayesnet",
+        "version": "1.1.1"
+      },
+      {
+        "name": "folding",
+        "version": "1.1.1"
+      },
+      {
+        "name": "argpase",
+        "version": "3.2"
+      },
+      {
+        "name": "catch2",
+        "version": "3.8.1"
+      },
+      {
+        "name": "nlohmann-json",
+        "version": "3.11.3"
+      }
+    ]
+  }