cda.ai-modular-framework 0.1.0

A modular AI framework

cda.AI-Modular-Framework

cda.ai-modular-framework is a highly modular and extensible AI development framework designed to facilitate seamless integration and management of various AI components and services. This framework is ideal for developers looking to build robust, adaptable AI solutions with ease. By leveraging a structured, plug-and-play architecture, this framework ensures maintainability, scalability, and flexibility.

Key Features

Modular Architecture

• Unified Control Engine: Manages core infrastructure tasks, including provisioning, logging, monitoring, and policy enforcement.
• Feature Service: Allows for the addition and management of various AI features and custom components.
• Operational Layer: Handles essential operational tasks such as input/output management, logging, and temporary storage.

Seamless Integration

• OpenAI Service: Integrates with OpenAI's GPT models and other AI services, providing functionalities for generating and summarizing content.
• Custom Components: Supports the addition of user-defined components to extend the framework’s capabilities.
• Frontend Chatbot Interface: Includes a WebSocket-based interface for real-time user interactions and AI-driven responses.

Flexibility and Extensibility

• Custom Functions: Easily add and manage custom functions for specific AI tasks.
• Nested Modularity: Each part of the system is modular, allowing for components to be extended or replaced without disrupting the overall system.

Discussing the Application/Repo: cda.ai-modular-framework

Purpose: The cda.ai-modular-framework is designed to provide a flexible, modular system for building AI-powered applications. It incorporates various components such as OpenAI services, LangChain, and other libraries to facilitate a wide range of functionalities from natural language processing to data retrieval and processing.

Core Features:

1. Unified Control Engine: Manages the deployment, scaling, and monitoring of various components.
2. Feature Services: Houses different AI services and tools such as OpenAI and LangChain functionalities.
3. Operational Layer: Manages input/output operations, logging, and temporary storage.
4. API Endpoints: Provides RESTful API interfaces using FastAPI for interacting with the system.

Directory Structure

The repository follows a clear and consistent directory structure:

cda.ai-modular-framework/
│
├── app/
│   ├── unified_control_engine/
│   │   ├── __init__.py
│   │   ├── unified_control_engine.py
│   │   └── frontend_chatbot_interface.py
│   ├── feature_service/
│   │   ├── __init__.py
│   │   ├── feature_service.py
│   │   ├── openai_service/
│   │   │   ├── __init__.py
│   │   │   ├── openai_service.py
│   │   │   └── functions/
│   │   │       ├── __init__.py
│   │   │       ├── function1.py
│   │   │       └── function2.py
│   │   └── custom_components/
│   │       ├── __init__.py
│   │       ├── component1.py
│   │       ├── component2.py
│   ├── operational_layer/
│   │   ├── __init__.py
│   │   ├── operational_layer.py
│   │   ├── message_handler/
│   │   │   ├── __init__.py
│   │   │   └── message_handler.py
│   │   └── tool_manager/
│   │       ├── __init__.py
│   │       └── tool_manager.py
│   ├── request_handler/
│   │   ├── __init__.py
│   │   └── request_handler.py
│   ├── api/
│   │   ├── __init__.py
│   │   └── api.py
│   ├── main.py
│   └── .env
│
├── .github/
│   └── workflows/
│       └── docker-build-test-push-and-package.yml
│
├── docker-compose.yaml
├── Dockerfile
├── setup.py
├── README.md
├── requirements.txt
└── LICENSE.md

Here is the optimized and formatted README.md content:

Getting Started

Prerequisites

• Python 3.8+
• FastAPI
• Uvicorn
• OpenAI API key
• Kubernetes (for infrastructure management)

Installation

1. Clone the repository:

   git clone https://github.com/Cdaprod/cda.ai-modular-framework.git
   cd cda.ai-modular-framework
   

2. Install the required packages:

   pip install -r requirements.txt
   

3. Set up environment variables by creating a .env file in the root directory:

   OPENAI_API_KEY=your-openai-api-key
   

4. Run the application:

   uvicorn app.main:app --reload
   

Running and Interacting with the Application

To ensure the application remains up and allows you to docker exec into it for live work, you should run the Docker container in detached mode. Here’s how you can achieve this:

Docker Compose

docker-compose.yml:

version: '3.8'

services:
  app:
    build:
      context: .
      dockerfile: Dockerfile
    image: cdaprod/cda.ai-modular-framework:latest
    ports:
      - "8000:8000"
    environment:
      - OPENAI_API_KEY=${OPENAI_API_KEY}
    networks:
      - app-network

networks:
  app-network:
    driver: bridge

Running the Container:

docker-compose up -d

Docker Run

Running the Container in Detached Mode:

docker run -d --name cda_modular_framework -p 8000:8000 cdaprod/cda.ai-modular-framework:latest

Interacting with the Running Container

Once the container is running, you can exec into it to perform live work:

docker exec -it cda_modular_framework /bin/bash

Development and Debugging Workflow

1. Running in Detached Mode: This keeps the container running in the background, allowing you to attach to it as needed.
2. Persistent Data and Configuration:

   • Ensure any configuration or data changes are persistent across container restarts by using Docker volumes or bind mounts.

   • For example, to mount a local directory to the container:

     docker run -d --name cda_modular_framework -p 8000:8000 -v /local/path/to/config:/container/path/to/config cdaprod/cda.ai-modular-framework:latest
     

CI/CD Integration

The GitHub Actions workflow ensures that every push or pull request is built, tested, and pushed to Docker Hub. Additionally, it can handle different Python versions, ensuring compatibility and robustness.

Summary

• Keep Container Running: Use docker-compose up -d or docker run -d to keep the container running in the background.
• Live Interaction: Use docker exec -it to interact with the running container.
• Persistent Changes: Use volumes or bind mounts to ensure changes are persistent.
• CI/CD Pipeline: Automated workflows ensure that your application is continuously integrated and deployed, supporting multiple Python versions and architectures.

This setup allows you to develop, debug, and deploy your modular AI framework efficiently, ensuring it remains flexible and extensible for future components and functionalities.

Usage

• Adding Features: Extend the FeatureService by adding new features in the app/feature_service directory.
• Custom Functions: Add custom functions in the app/feature_service/openai_service/functions directory.
• Real-time Interaction: Use the WebSocket-based frontend chatbot interface for real-time user interactions.

Contribution

We welcome contributions to enhance the capabilities and features of the cda.ai-modular-framework. Please follow the guidelines outlined in CONTRIBUTING.md to submit issues and pull requests.

License

This project is licensed under Private License. See the LICENSE file for details.

Contact

For further information, questions, or suggestions, please contact David Cannan at [e-mail].