Repository files navigation

Cerberus is a modular, containerized AI solution offering APIs for sentiment analysis, text summarization, and text generation. Built on top of fine-tuned and quantized Hugging Face models, Cerberus is designed for scalability, efficiency, and easy integration into workflows.

• Sentiment Analysis: Quickly classify text sentiment as positive, negative, or neutral.
• Summarization: Generate concise and meaningful summaries for long text.
• Text Generation: Produce creative or structured text based on input prompts.
• Scalable Architecture: Deployed using containerized services for easy scaling.
• Optimized Models: Uses fine-tuned and quantized Hugging Face models for speed and accuracy.
• OpenAPI Documentation: Built-in API documentation for developers.
• Extensible: Easily add new AI models and services.

• Programming Language: Python 3.10+
• Frameworks: FastAPI for API development
• AI Models: Hugging Face (Transformers, Optimum, Accelerate)
• Machine Learning Backend: PyTorch with MPS support for macOS (Apple Silicon)
• Data Handling: Datasets library by Hugging Face
• Deployment:
  • Docker for containerization
  • Docker Compose for service orchestration

• Cerberus
  • Features
  • Technologies Used
  • Table of Contents
  • Getting Started
    • Prerequisites
    • Installation
  • Usage
    • Endpoints
    • Example Requests
  • API Documentation
  • Development
    • Run Locally Without Docker
    • Test the Models
  • Contributing
  • License

• Python 3.10 or higher

• Poetry for dependency management:

  curl -sSL https://install.python-poetry.org| python3 -

• Docker and Docker Compose installed:Docker Installation Guide

1. Clone the Repository:

   git clone https://github.com/BonnardValentin/cerberuscd cerberus

2. Install Dependencies with Poetry:

   poetry install

3. Set Up Environment Variables: Create a.env file in the project root with the following:

   MODEL_PATH="./models/quantized"API_KEY="your-secret-key"

4. Start the Application: Using Docker Compose:

   docker-compose up --build

5. Test the Application: Access the API athttp://localhost:8000.

• Sentiment Analysis:/analyze-sentiment/
• Summarization:/summarize/
• Text Generation:/generate-text/

Sentiment Analysis

curl -X POST http://localhost:8000/analyze-sentiment/ \-H"Content-Type: application/json" \-d'{"text": "I love Cerberus!"}'

Summarization

curl -X POST http://localhost:8000/summarize/ \-H"Content-Type: application/json" \-d'{"text": "Docker simplifies containerization by providing a platform to develop, ship, and run applications in isolated environments."}'

Text Generation

curl -X POST http://localhost:8000/generate-text/ \-H"Content-Type: application/json" \-d'{"text": "Once upon a time, in a faraway land,"}'

Cerberus provides an interactive OpenAPI UI:

• Visithttp://localhost:8000/docs for Swagger UI.
• Visithttp://localhost:8000/redoc for ReDoc.

1. Activate the Poetry environment:

   poetry env activate

2. Start the FastAPI server:

   uvicorn app.main:app --reload

Run a Python script to verify that models are loaded correctly:

fromtransformersimportpipelinemodel=pipeline("sentiment-analysis",model="./models/quantized")print(model("I love open-source!"))

We welcome contributions to Cerberus! To contribute:

1. Fork the repository.

2. Create a feature branch:

   git checkout -b feature-name

3. Commit your changes:

   git commit -m"Add new feature"

4. Push to your fork:

   git push origin feature-name

5. Open a pull request.

Please read our Contributing Guide for more details.

This project is licensed under the MIT License - see theLICENSE file for details.