C++ Machine Learning Library

cpp-ml-library is a C++ library implementing foundational machine learning algorithms inspired by the Udemy course "Machine Learning A-Z: AI, Python & R" taught by Kirill Eremenko and Hadelin de Ponteves. This project aims to reimplement key algorithms from the course in C++ to provide reusable implementations for educational purposes, experimentation, and practical applications in larger C++ projects.

Project Structure

cpp-ml-library/
├── ml_library_include/     # Header files
├── tests/                  # Unit tests
├── examples/               # Example usage files
├── docs/                   # Documentation files
└── CMakeLists.txt          # CMake configuration file

Getting Started

Requirements

• C++20
• CMake 3.18+ (if using CMake for building)
• Any C++ compiler that supports C++20 (e.g., GCC 10+, Clang 10+, MSVC)

Building the Library

To build the library with CMake:

mkdir build
cd build
cmake ..
make

Installing the Library

After building, you can install the library system-wide:

make install

Usage

To use this library in your C++ project, include the master header file:

#include "ml_library_include/ml.h"

Implemented Algorithms

The following machine learning algorithms are planned, inspired by concepts and techniques taught in the Udemy course:

1. Regression

   • Polynomial Regression (PolynomialRegression)
   • Multi-Linear Regression (MultilinearRegression)
   • Logistic Regression (LogisticRegression)
   • Decision Tree Regression (DecisionTreeRegressor)
   • Random Forest Regression (RandomForestRegressor)
   • K-Nearest Neighbors (KNNRegressor)

2. Classification

   • Decision Tree Classifier (DecisionTreeClassifier)
   • Random Forest Classifier (RandomForestClassifier)
   • K-Nearest Neighbors (KNNClassifier)

3. Clustering

   • K-Means Clustering (KMeans)
   • Hierarchical clustering (HierarchicalClustering)

4. Neural Networks

   • Neural Network (NeuralNetwork)
   • Artificial Neural Network (ANN)
   • Convolutional Neural Network (CNN)

5. Association Rule Learning

   • Apriori
   • Eclat

6. Support Vector Machine

   • Support Vector Regression (SupportVectorRegression)

Algorithm Implementation Progress

Algorithm Category	Algorithm	Implemented	Tests	Examples
Regression	Polynomial Regression (PolynomialRegression)	[x]	[ ]	[x]
	Logistic Regression (LogisticRegression)	[x]	[ ]	[x]
	Multi-Linear Regression (MultilinearRegression)	[x]	[ ]	[x]
	Decision Tree Regression (DecisionTreeRegressor)	[ ]	[ ]	[ ]
	Random Forest Regression (RandomForestRegressor)	[ ]	[ ]	[ ]
	K-Nearest Neighbors (KNNRegressor)	[x]	[ ]	[ ]
Classification	Decision Tree Classifier (DecisionTreeClassifier)	[ ]	[ ]	[ ]
	Random Forest Classifier (RandomForestClassifier)	[ ]	[ ]	[ ]
	K-Nearest Neighbors (KNNClassifier)	[x]	[ ]	[ ]
Clustering	K-Means Clustering (KMeans)	[x]	[ ]	[ ]
	Hierarchical clustering (HierarchicalClustering)	[x]	[ ]	[ ]
Neural Networks	Neural Network (NeuralNetwork)	[x]	[x]	[x]
	Artificial Neural Network	[ ]	[ ]	[ ]
	Convolutional Neural Network	[ ]	[ ]	[ ]
Association Rule Learning	Apriori	[x]	[x]	[x]
	Eclat	[x]	[x]	[x]
Support Vector Machine	Support Vector Regression (SupportVectorRegression)	[ ]	[ ]	[ ]

Examples

Each algorithm will include example usage. Examples are located in the examples/ directory.

Example: Decision Tree Classifier

#include "ml/ml.h"
#include <vector>
#include <iostream>

int main() {
    ml::tree::DecisionTreeClassifier dtc;
    std::vector<std::vector<double>> data = { {5.1, 3.5, 1.4, 0.2}, {4.9, 3.0, 1.4, 0.2} };
    std::vector<int> labels = { 0, 1 };
    dtc.fit(data, labels);

    int prediction = dtc.predict({5.1, 3.5, 1.4, 0.2});
    std::cout << "Predicted class: " << prediction << std::endl;

    return 0;
}

Documentation

The documentation for this project is generated using Doxygen and is available online at GitHub Pages. The documentation provides details on each class, function, and algorithm.

Contributing

Contributions are welcome! Please submit a pull request or open an issue if you'd like to discuss ideas or report bugs.

License

This project is licensed under the MIT License. See the LICENSE file for details.