R&D Template

This is a base template focused on Research and Development of the projects.

✨ Features

• Base structure
• Custom configuration
• Common recommended files and folders
• Best practices

🚀 Quick Start

# Clone the template

git clone https://github.com/humblebeeintel/rnd-template.git my_project
cd my_project

# Setup environment

python -m venv .venv
source .venv/bin/activate  

# Windows: 
# .venv\Scripts\activate

# Install dependencies (choose one)

pip install -r requirements/requirements.cpu.txt  # For CPU
pip install -r requirements/requirements.cuda.txt  # For GPU

# Initialize project structure

chmod +x ./scripts/setup.sh && ./scripts/setup.sh

# Create your .env file for credentials

cp .env.example .env

📂 Project Structure

├── configs/              # Your configuration files
├── data/                 # Data files (gitignored)
├── experiments/          # Notebooks and experiment code
├── models/               # Saved models
├── modules/              # Reusable code
├── scripts/              # Utility scripts
├── .env                  # Environment variables
└── config.py             # Configuration management

💡 How to use this template

📦 Use modules for external source code

GitHub repositories or open-source packages that aren't on PyPI but are useful for your workflow

Integrate an external GitHub repo to the template

cd modules
git clone https://github.com/some-user/some-cv-library.git

Then you can import it in your code like:

from modules.some_cv_library import some_module

Tip: Keep external modules in their own subfolders to avoid namespace issues and make version control easier.

⛰️ Environment Variables (.env)

Define API keys and sensitive information:

API_KEY=your_secret_key
DATASET_PATH=/path/to/data

Important: Make sure you don't commit it to GitHub and you have it in the .gitignore file

📝 Documentation

• Check docs/ for detailed documentation
• Example notebooks are in experiments/notebooks
• Common modules are in modules/ directory

✨ Other considerations

• Add your data to data/
• Create preprocessing scripts in scripts/
• Build your model or move pre-trained models to models/
• Track experiments with logging