Machine Learning Algorithms (MATLAB Implementation)

Implementation of machine learning algorithms from Andrew Ng's Coursera course, originally developed in MATLAB/OCTAVE as part of the programming assignments.

📌 Project Overview

This repository contains:

• MATLAB implementations of core ML algorithms
• Solutions to Coursera programming exercises
• Helper functions used in the course

🧠 Algorithms Implemented

Core Algorithms

Algorithm	MATLAB Files	Course Week
Linear Regression	linearReg.m, gradientDescent.m	Week 2
Logistic Regression	logisticReg.m, costFunction.m	Week 3
Neural Networks	nnCostFunction.m, predict.m	Week 4
Support Vector Machines	svmTrain.m, gaussianKernel.m	Week 7
K-Means & PCA	kMeans.m, pca.m	Week 8

🛠️ Setup Instructions

MATLAB Installation

1. Ensure you have MATLAB installed (R2020b or later recommended)
2. Clone this repository:

   git clone https://github.com/imbilalbutt/Machine-learning-algorithms.git

3. Add the project folder to MATLAB path:

   addpath(genpath('/path/to/repository'));

OCTAVE Alternative

For users without MATLAB:

1. Install GNU Octave
2. All files are compatible with Octave 6.0+

📊 Running the Code

Example: Linear Regression

% Load data
data = load('ex1data1.txt');
X = data(:, 1); y = data(:, 2);

% Run gradient descent
theta = gradientDescent(X, y, zeros(2,1), 0.01, 1500);

% Plot results
plot(X(:,2), y, 'rx', 'MarkerSize', 10);
hold on;
plot(X(:,2), X*theta, '-');

Example: Neural Networks

% Load pre-trained weights
load('ex3weights.mat');

% Make predictions
pred = predict(Theta1, Theta2, X);
accuracy = mean(double(pred == y)) * 100;
fprintf('Training Set Accuracy: %f\n', accuracy);

📈 Course Exercise Mapping

MATLAB File	Corresponding Exercise
ex1.m	Linear Regression with One Variable
ex2.m	Logistic Regression
ex3.m	Multi-class Classification with NNs
ex4.m	Neural Network Backpropagation
ex5.m	Bias-Variance Analysis
ex6.m	Support Vector Machines
ex7.m	K-Means & PCA

🔍 Key Features

• Original MATLAB/OCTAVE implementations as done in the course
• Vectorized operations for efficient computation
• Detailed comments explaining each step
• Visualization scripts for algorithm behavior

📝 Notes for Developers

1. File naming follows Coursera's original structure
2. All functions are self-contained (no toolboxes required)
3. Test data included in /data directory

🤝 Contribution Guidelines

While this is primarily a course repository, improvements are welcome:

• Bug fixes in implementations
• Additional documentation
• Performance optimizations

Please maintain MATLAB/OCTAVE compatibility when submitting changes.

📜 License

MIT License - See LICENSE for details.

🙏 Acknowledgments

• Andrew Ng and Stanford University for the course materials
• Coursera for hosting the content
• MathWorks for MATLAB development environment

Key differences from the Python version:
1. **MATLAB-specific setup instructions**
2. **Original course file structure** preserved
3. **MATLAB code examples** with proper syntax
4. **Octave compatibility** noted
5. **Exercise mapping** to original Coursera problems
6. **No dependency management** (MATLAB doesn't use pip/requirements.txt)

Would you like me to:
1. Add specific MATLAB debugging tips?
2. Include screenshots of example outputs?
3. Add a comparison between MATLAB and Python implementations?