Skip to content

README.md

Latest commit

 

History

History
197 lines (145 loc) · 3.91 KB

README.md

File metadata and controls

197 lines (145 loc) · 3.91 KB
Raw

Agent LLM RAG App Backend

Table of Contents

  1. Backend Structure
  2. Installation and Setup
  3. Environment Variables Setup
  4. Running the Application
  5. Using Docker
  6. Troubleshooting

1. Backend Structure

.
├── __pycache__
├── .theflow
├── api
├── data
├── logs
├── src
├── venv
├── .env
├── .env.example
├── .gitignore
├── app_fastapi.py
├── docker-compose.yaml
├── README.md
├── requirements.txt

Explanation

  • api/: Contains API logic for FastAPI.
  • data/: Stores data, including Qdrant storage.
  • logs/: Stores log files.
  • src/: Additional source code.
  • venv/: Virtual environment containing dependencies.
  • .env: Environment variables file.
  • .gitignore: Lists files and directories to ignore in Git.
  • app_fastapi.py: Main FastAPI application.
  • docker-compose.yaml: Docker Compose configuration.
  • requirements.txt: Backend dependencies.

2. Installation and Setup

2.1. Prerequisites

  • Install Docker.
  • Install Git.
  • Use Python 3.9+.

2.2. Setup Instructions

  1. Clone the repository

    git clone https://github.com/buithanhdam/rag-app-agent-llm.git
cd rag-app-agent-llm/backend

  2. Set up the virtual environment

    • Unix/macOS: 
      python3 -m venv venv
source venv/bin/activate
    • Windows: 
      python -m venv venv
.\venv\Scripts\activate

  3. Install dependencies

    • Backend (FastAPI): 
      pip install -r requirements.txt

3. Environment Variables Setup

  1. Copy .env.example to .env

    cp .env.example .env

  2. Configure the environment variables

    GOOGLE_API_KEY=<your_google_api_key>
OPENAI_API_KEY=<your_openai_api_key>
ANTHROPIC_API_KEY=<your_anthropic_api_key>
BACKEND_API_URL=http://localhost:8000
QDRANT_URL=http://localhost:6333

MYSQL_USER=your_mysql_user
MYSQL_PASSWORD=your_mysql_password
MYSQL_HOST=your_mysql_host
MYSQL_PORT=your_mysql_port
MYSQL_DB=your_mysql_db
MYSQL_ROOT_PASSWORD=root_password

AWS_ACCESS_KEY_ID=
AWS_SECRET_ACCESS_KEY=
AWS_REGION_NAME=
AWS_STORAGE_TYPE=
AWS_ENDPOINT_URL=

4. Running the Application

4.1. Run FastAPI Backend

uvicorn app_fastapi:app --host 0.0.0.0 --port 8000 --reload
  • Access API at: http://127.0.0.1:8000

5. Using Docker

5.1. Build and Run Services

  1. Ensure Docker is running.

  2. Build and start containers:

    docker-compose build
docker-compose up

5.2. Included Services

  • fastapi: Exposes port 8000.
  • qdrant: Exposes ports 6333, 6334.
  • mysql: Exposes port 3306.

5.3. Set Up MySQL Database

docker exec -it your-container-name bash
mysql -u root -p

  • Enter root password (configured in .env or docker-compose.yml).

  • Run the following SQL queries:

    CREATE USER 'user'@'%' IDENTIFIED BY '1';
GRANT ALL PRIVILEGES ON ragagent.* TO 'user'@'%';
FLUSH PRIVILEGES;
    CREATE DATABASE ragagent;

5.4. Accessing Services

  • FastAPI: http://localhost:8000

5.5. Stopping Services

docker-compose down

5.6. Additional Docker Information

  • Docker network: rag-app-network (bridge).
  • FastAPI service uses ./Dockerfile.backend.

6. Troubleshooting

  • Ensure .env is correctly configured.
  • If there are changes in Docker, run: 
    docker-compose up --build
  • Verify that required ports (8000, 6333) are not in use.

This documentation provides all necessary steps to set up, run, and manage the project efficiently.