Machine Learning Implement Algorithm in C++ with free codes

Intro – Machine Learning Implement

Machine Learning Implement is a field of AI (Artificial Intelligence) that focuses on enabling machines to learn from data and improve their performance overtime. Implementing machine learning Algorithms from scratch can provide a deep understanding of how these algorithms work. Here We Discuss the Implementation of Linear Regression Machine Learning Algorithm.

Linear Regression

Linear Regression is a supervised learning algorithm used for predicting a continuous target variable based on one or more input features. The goal is to find a linear relationship between the input features and the target variable.

Formula and Algorithm

The formula for simple linear regression (with one input feature) is:

y = b0 +b1 ⋅ x

• y is the predicted value.
• x is the input feature.
• b0​ is the y-intercept.
• b1​ is the slope of the line.

The algorithm involves the following steps:

1. Initialize the parameters b0​ and b1​.
2. Compute the cost function (mean squared error).
3. Update the parameters using gradient descent.
4. Repeat steps 2 and 3 until convergence.

C++ Codes of the Project

Here is the Complete Codes of the Project,

#include <iostream>
#include <vector>
#include <cmath>

// Function to compute the mean squared error
double computeCost(const std::vector<double>& x, const std::vector<double>& y, double b0, double b1) {
    double cost = 0;
    int n = x.size();
    for (int i = 0; i < n; ++i) {
        double prediction = b0 + b1 * x[i];
        cost += std::pow(prediction - y[i], 2);
    }
    return cost / (2 * n);
}

// Function to perform gradient descent
void gradientDescent(std::vector<double>& x, std::vector<double>& y, double& b0, double& b1, double learningRate, int iterations) {
    int n = x.size();
    for (int iter = 0; iter < iterations; ++iter) {
        double sum0 = 0, sum1 = 0;
        for (int i = 0; i < n; ++i) {
            double prediction = b0 + b1 * x[i];
            sum0 += prediction - y[i];
            sum1 += (prediction - y[i]) * x[i];
        }
        b0 -= learningRate * sum0 / n;
        b1 -= learningRate * sum1 / n;

        // Optional: print cost for each iteration to monitor convergence
        std::cout << "Iteration " << iter + 1 << ": Cost = " << computeCost(x, y, b0, b1) << "\n";
    }
}

int main() {
    // Example data
    std::vector<double> x = {1, 2, 3, 4, 5};
    std::vector<double> y = {3, 4, 2, 5, 6};

    double b0 = 0; // Initial intercept
    double b1 = 0; // Initial slope
    double learningRate = 0.01;
    int iterations = 1000;

    gradientDescent(x, y, b0, b1, learningRate, iterations);

    std::cout << "Final parameters: b0 = " << b0 << ", b1 = " << b1 << "\n";
    return 0;
}