Image Captions Api

Description

This Flask-based application generates captions for images using machine learning. It leverages Hypercorn and asyncio for improved performance.

Setup & Installation

Local Environment

Prerequisites:

• Python 3.9
• Pip
• Virtualenv (optional)

Steps:

1. Clone the repository.
2. (Optional) Create and activate a virtual environment:

   python -m venv venv #python3 for linux
   source venv/bin/activate  # linux
   venv\Scripts\activate  # Windows

3. Install dependencies:

   pip install -r requirements.txt

Docker Environment

Prerequisites:

• Docker
• Docker Compose

Steps:

1. Run docker-compose up to start the application.

Performance Enhancement with Hypercorn and Asyncio

The application uses Hypercorn as an ASGI server, along with asyncio, to enhance performance through asynchronous processing. This setup allows for handling multiple requests efficiently.

Scaling Up with Kubernetes

For scaling the application:

1. Create Kubernetes configurations for deployment, services, and horizontal pod autoscalers.
2. Deploy to a Kubernetes cluster, adjusting replicas as needed for load.

Usage

Send a POST request to /caption with an image file. curl -X POST -F "file=@path/to/your/image.jpg" http://localhost:5000/caption

Running Tests

Execute tests using python -m unittest from the src directory.

Contributing

Contributions are welcome following the standard fork, branch, and pull request workflow.

License

MIT License