SimpleML

• SimpleML is a C++ implementation of various machine learning algorithms.
• The code is simple and easy to understand.
• It can help you get a clear understanding of machine learning algorithms.
• A dockerized test environment is available.

1. Requirements

• Eigen 3.3.9
• g++ 9.0 or higher

2. Supported ML algorithms

• Classification

  • Decision tree
  • Naive Bayes
  • K-nearest neighbors

• Clustering

  • K-means
  • Gaussian mixture

• Dimensionality reduction

  • Principal Component Analysis (PCA)

• Regression

  • Ordinary Least Squares (OLS)

3. Examples

• Example code of all models is already written in main.cpp file.
• Please uncomment the code you want to use and check if file_name (path to dataset) is correct.
• Ex 1) Decision tree classifier

#include <iostream>
#include <string>
#include <Eigen/Dense>
#include "./headers/file_manage.h"
#include "./headers/model_evaluation.h"
#include "./headers/decision_tree.h"
using namespace std;
using namespace Eigen;

int main()
{
	string file_name = "./dataset/iris.csv";

	MatrixXf X;
	VectorXi Y;

	// read csv
	SimpleML::read_csv(file_name, X, Y);

	// run decision tree classifier
	SimpleML::DecisionTree dt;
	dt.fit(X, Y);
	dt.print_tree();

	// evaluate
	VectorXi predicted = dt.predict(X);
	float acc = SimpleML::calc_accuracy(Y, predicted);
	cout << "\nAccuracy: " << acc * 100 << "%" << endl;

	// K-fold cross validation. Uncomment below if necessary
	/*float acc = SimpleML::evaluate_classification_model(dt, X, Y, 5);
	cout << "Accuracy(k-fold): " << acc * 100 << "%" << endl;*/
}

• Ex 2) Gaussian mixture

#include <iostream>
#include <string>
#include <Eigen/Dense>
#include "./headers/file_manage.h"
#include "./headers/model_evaluation.h"
#include "./headers/gaussian_mixture.h"
using namespace std;
using namespace Eigen;

int main()
{
	string file_name = "./dataset/iris.csv";

	MatrixXf X;
	VectorXi Y;

	// read csv
	SimpleML::read_csv(file_name, X, Y);

	// run gaussian mixture model
	SimpleML::GaussianMixture gm(3);
	gm.fit(X);

	// evaluate
	vector<vector<int>> predicted = gm.predict(X);
	float sil = SimpleML::silhouette_score(X, predicted);
	cout << "Silhouette score : " << sil << endl;

	return 0;
}

4. Compile & Run

• Pull the docker image

# host shell
docker pull stnamjef/eigen_3.3.9:1.0

• Run the docker image

# pwd -> the project directory (SimpleML)
docker run -it -v $(pwd):/usr/build stnamjef/eigen_3.3.9:1.0

• Compile

# container shell at /usr/build
g++ main.cpp --std=c++17 -O2 -o simpleml

• Run

./simpleml

5. Ongoing list

• CLI options
• Logistic regression, SVM