MD YOLO

A project for developing and experimenting with YOLO (You Only Look Once) models and their variations for object detection.

Project Overview

This repository contains experiments, implementations, and comparisons of various YOLO architectures including:

• YOLOv3
• YOLOv4
• YOLOv5
• YOLOv6
• YOLOv7
• YOLOv8
• YOLO-NAS

Project Structure

md_yolo/
│
├── data/               # Dataset storage and processing scripts
│   ├── raw/            # Raw datasets
│   ├── processed/      # Processed datasets
│   └── augmented/      # Augmented datasets for training
│
├── models/             # Saved model checkpoints and configurations
│   ├── yolov3/
│   ├── yolov5/
│   ├── yolov8/
│   └── ...
│
├── notebooks/          # Jupyter notebooks for exploration and visualization
│
├── src/                # Source code
│   ├── models/         # Model implementations
│   ├── utils/          # Utility functions
│   ├── train.py        # Training script
│   ├── eval.py         # Evaluation script
│   └── infer.py        # Inference script
│
└── tests/              # Unit tests

Setup

# Create virtual environment using uv
uv venv

# Activate virtual environment
source .venv/bin/activate  # On Linux/Mac
# or
.venv\Scripts\activate     # On Windows

# Install dependencies
uv pip install -r requirements.txt

Usage

Details on how to use the project will be added as the project evolves.

License

MIT