Machine Learning Projects Collection

Overview

This repository contains three comprehensive machine learning projects, each demonstrating different aspects of machine learning implementation, including deep learning architectures. These projects showcase both from-scratch implementations and usage of popular libraries like scikit-learn and PyTorch.

Details

Click on each folder for detailed readme files.

ML_Projects

Each folder contains:

• code files for two different code files corresponding to two different projects
• Report for a detailed overview of each code file
• Readme for a brief overview of each code file

Repository Structure

ML_Projects/
├── first/                   # Machine Learning Fundamentals
│   ├── ass1q2.ipynb        # Logistic Regression from scratch
│   ├── ass1q3.ipynb        # Regression analysis using libraries
│   ├── htd1.csv            # Heart Disease dataset
│   ├── htd2.csv            # Electricity Bill dataset
│   ├── readme.md           # Project-specific documentation
│   └── report.pdf          # Detailed analysis and results
│
├── second/                  # Neural Network Implementation
│   ├── ass3q2.ipynb        # Neural Network from scratch
│   ├── ass3q3.ipynb        # sklearn Neural Network implementation
│   ├── NN_models.zip       # Trained neural network models
│   ├── readme.md           # Project-specific documentation
│   └── report.pdf          # Detailed analysis and results
│
├── third/                   # Advanced ML and Deep Learning
│   ├── ass4q2.ipynb        # KMeans clustering implementation
│   ├── ass4q3.ipynb        # CNN & MLP with PyTorch
│   ├── cnn_model.pth       # Trained CNN model weights
│   ├── mlp_model.pth       # Trained MLP model weights
│   ├── readme.md           # Project-specific documentation
│   └── report.pdf          # Detailed analysis and results
│
└── README.md               # This file

Projects Summary

1. Machine Learning Fundamentals (first/)

• Logistic Regression implementation from scratch
• Advanced regression techniques using libraries
• Focus on feature engineering and model evaluation
• Datasets: Heart Disease and Electricity Bill

2. Neural Network Implementation (second/)

• Custom Neural Network implementation from scratch
• sklearn-based neural network models
• MNIST and Fashion-MNIST dataset analysis
• Multiple trained models with different configurations

3. Advanced ML and Deep Learning (third/)

• KMeans clustering implementation from scratch
• CNN and MLP implementation using PyTorch
• CIFAR-10 dataset analysis
• Trained model weights and performance comparisons

Getting Started

Prerequisites

• Python 3.8+
• Jupyter Notebook
• Required libraries:
  • NumPy
  • Pandas
  • Matplotlib
  • scikit-learn
  • PyTorch
  • torchvision

Installation

1. Clone the repository

git clone <repository-url>
cd ML_Projects

2. Install dependencies

pip install numpy pandas matplotlib scikit-learn torch torchvision

Usage

Each project directory contains its own:

• Detailed README with specific instructions
• Jupyter notebooks with implementation
• Dataset files or loading instructions
• Comprehensive report with analysis

Project Status

• All projects are complete and documented
• Models are trained and saved
• Results are analyzed and reported

Additional Information

For detailed information about each project:

• First Project: See first/readme.md
• Second Project: See second/readme.md
• Third Project: See third/readme.md

For technical details, implementation specifics, and comprehensive analysis, refer to the respective report.pdf files in each project directory.