Trade Bot

Overview

This Trade Bot is a FastAPI-based application combined with an AI-powered trading bot that interacts with a PostgreSQL database to store and retrieve trading data. It also provides system logs for monitoring and makes automated trade decisions based on an advanced trading strategy incorporating AI-based filtering, EMA crossover strategy, ADX, RSI, ATR, and volume analysis.

Features

• AI-Powered Trade Filtering: Uses a trained XGBoost model to validate trades before execution.
• Dynamic Market Analysis: Incorporates trend confirmation, volatility checks, and volume filters.
• Fetch Trading Data from PostgreSQL: Stores and retrieves past trades efficiently.
• Retrieve System Logs: Tracks all trading actions in ./logs/system_log.logs.
• Automated Trading Strategy: Executes trades using AI-enhanced decision-making.
• API Endpoints with FastAPI: Provides real-time trading data access.
• 🚀 Dynamic Bot Settings Update via RabbitMQ.
• Docker Support for Easy Deployment.

Requirements

• Python 3.11+
• PostgreSQL
• RabbitMQ (for dynamic settings updates)
• Docker (optional for containerized deployment)

Installation

1. Clone the repository:

   git clone https://github.com/Frogleim/Trade-Bot-Model.git
   cd Trade-Bot-Model

2. Create a virtual environment and activate it:

   python -m venv venv
   source venv/bin/activate  # On Windows use `venv\Scripts\activate`

3. Install dependencies:

   pip install -r requirements.txt

4. Set up the environment variables in .env file (if applicable):

   DATABASE_URL=postgresql://postgres:admin@localhost:5433/tb
   ATR_PERIOD=14

5. Train and Save the AI Model (First-Time Setup Only):

   python train_xgboost.py  # Trains and saves `xgboost_model.model` and `scaler.pkl`

Running the Application

Locally

Start the API:

uvicorn main:app --host 0.0.0.0 --port 8000

Start the Bot:

python bot/main.py

With Docker

1. Build and start the services using Docker Compose:

   docker-compose up --build

API Endpoints

Get All Trades

Endpoint: GET /trades

Response:

[
    {
        "id": 1,
        "symbol": "BTC/USD",
        "entry_price": 45000.0,
        "exit_price": 46000.0,
        "pnl": 1000.0,
        "long_ema": 45500.0,
        "short_ema": 45800.0,
        "adx": 25.5,
        "atr": 300.0,
        "rsi": 60.2,
        "volume": 5000.0,
        "side": "buy"
    }
]

Get System Logs

Endpoint: GET /system_logs

Response:

{
    "logs": [
        "2025-02-05 12:00:00 - Trade executed: BTC/USD Buy 1.5 BTC",
        "2025-02-05 12:05:00 - Trade closed: BTC/USD Sell 1.5 BTC"
    ]
}

AI-Powered Trading Bot

The bot fetches market data using Binance API and applies the following AI-enhanced trading strategy:

• AI Trade Validation: Uses XGBoost to confirm high-probability trades.
• EMA Crossover: Identifies bullish and bearish crossovers.
• ADX Confirmation: Ensures strong trends before making a trade.
• RSI Filter: Confirms if the asset is in a good buying or selling range.
• ATR for Volatility: Ensures market has sufficient volatility before entry.
• Volume Confirmation: Uses historical volume to confirm trend strength.

The bot determines whether to go long, short, or hold based on the AI's decision-making process.

Docker Compose Configuration

The project uses the following docker-compose.yml setup:

version: '3.9'

services:
  bot:
    env_file:
      - ./bot/tools/.env
    build:
      context: ./bot
    volumes:
      - tools:/logs
      - tools:/bot/tools  # Shared tools directory
    restart: always

  api:
    env_file:
      - ./bot/tools/.env
    build:
      context: .
    volumes:
      - tools:/logs
      - tools:/bot/tools
    ports:
      - "8000:8000"
    restart: always

  db:
    image: "postgres:16"
    hostname: 'pgdb'
    ports:
      - 5432:5432
    environment:
      - POSTGRES_DB=tb
      - POSTGRES_USER=postgres
      - POSTGRES_PASSWORD=admin
    container_name: bot_db
    volumes:
      - pgdbdata:/var/lib/postgresql/data/
    restart: always

  rabbitmq:
    image: "rabbitmq:3-management"
    container_name: rabbitmq
    hostname: 'rabbitmq'
    ports:
      - "5672:5672"
      - "15672:15672"
    environment:
      - RABBITMQ_DEFAULT_USER=guest
      - RABBITMQ_DEFAULT_PASS=guest
    restart: always

  receiver:
    build:
      context: ./bot/rabbit
    command: ["python3", "receiver.py"]
    depends_on:
      - rabbitmq
    environment:
      - PYTHONUNBUFFERED=1
    volumes:
      - tools:/logs
      - tools:/bot/tools  # Shared tools directory
    restart: always

volumes:
  coins_trade:
  pgdbdata:
  tools:  # This ensures tools is a named volume

Contributing

Feel free to submit issues and pull requests for improvements.