Machine Learning Projects Repository

This repository contains multiple machine learning projects ranging from image processing to demand forecasting in the cab booking industry. Each project is self-contained with its Jupyter notebooks, datasets, and documentation.

Projects Overview

1. Handwritten Character Recognition

• Description: This project focuses on recognizing handwritten characters using machine learning techniques.
• Notebook: Handwritten Character Recognition Notebook

2. Image Stitching GAN (Generative Adversarial Network)

• Description: This project demonstrates the use of GANs for stitching images to create panoramas.
• Components:
  • Notebook: Image-Stitching-GAN Notebook
  • Documentation: Project Report
  • Sample Outputs and Inputs:
    • High-Resolution Output:
    • Input 1:
    • Input 2:
    • Generated Panorama:
  • Read More: Additional Info

3. ML Forecast Cab Booking Demand

• Description: Analyze and predict the demand for cabs using historical booking data.
• Components:
  • Notebook: Forecast Cab Booking Demand Notebook
  • Documentation: Demand Forecasting Report

4. ML Model for Auto Insurance Industry

• Description: Develop a model to assist in risk assessment and premium calculation for auto insurance.
• Notebook: ML Model for Auto Insurance Industry Notebook

5. Super-Resolution and Low-Light Image Enhancement

• Description: This repository focuses on two advanced image processing techniques. The first project enhances the resolution of images through super-resolution techniques, aiming to reconstruct a high-resolution image from its low-resolution counterpart. The second project improves the visibility and quality of images taken in low-light conditions using various enhancement algorithms. Both projects are designed to demonstrate the practical applications of these techniques in improving image quality for better analysis and visualization.
• Notebooks:
  • Super-Resolution: Super-Resolution Notebook showcases the method to enhance image resolution.
  • Low-Light Enhancement: Low-Light Enhancement Notebook demonstrates techniques to enhance images captured in poorly lit environments.
• Outputs and Figures:
  • Super-Resolution Output: Demonstrated in SupRes_LowLightEnhance/output-super-res/SupRes-out.JPG.
  • Low-Light Enhancement Outputs: Available in the SupRes_LowLightEnhance/output-low-light directory with multiple enhanced examples.
  • Figures: Visual results can be found in the SupRes_LowLightEnhance/fig directory illustrating the enhancements.
• Presentation: Detailed insights and methodologies are discussed in LowLight_SuperRes.pdf, which includes a comprehensive presentation of the projects.

Getting Started

To get started with any of these projects, clone the repository, navigate to the project's directory, and open the Jupyter notebooks in a Jupyter environment:

git clone https://github.com/your-repo/ML.git
cd ML/<project_directory>
jupyter notebook <notebook_name>.ipynb

Contribution

Contributions to improve the projects or add new features are welcome. Please fork the repository and submit a pull request with your changes.