chart-sayer

Chart Sayer is a lightweight service designed to analyze chart images and extract key trading information such as entry and exit points based on user-defined rules. The service can be integrated with platforms like Discord and Telegram to automate the process of chart analysis.

Features

• Image Processing: Analyze chart images to detect entry and exit points using machine vision and OCR.
• Bot Integration: Seamlessly integrate with Discord and Telegram to receive and process chart images.
• Customizable Rules: Define custom rules for interpreting chart features, such as color codes for entry and exit points.

Installation

Method 1: Using Docker (Recommended)

1. Clone the repository:

   git clone https://github.com/lowhung/chart-sayer.git
   cd chart-sayer

2. Create an .env file based on .env.example.

3. Run with Docker Compose:

   docker-compose up -d

   This will start both the Chart Sayer service and a Redis instance.

Method 2: Manual Installation

1. Clone the repository:

   git clone https://github.com/lowhung/chart-sayer.git
   cd chart-sayer

2. Install the required dependencies:

   poetry install

3. Set up a Redis instance (optional, for position storage):

   # Install Redis on your system or use a cloud-hosted instance
   # Update the REDIS_URL in your .env file accordingly

Environment Variables

Ensure you have the following environment variables set in your .env file:

• TELEGRAM_TOKEN: Your Telegram bot token.
• DISCORD_TOKEN: Your Discord bot token.
• DISCORD_CLIENT_ID: Your Discord client ID.
• DISCORD_CLIENT_SECRET: Your Discord client secret.
• DISCORD_PUBLIC_KEY: Your Discord public key.
• OPENAI_API_KEY: Your OpenAI API key.
• REDIS_URL: URL for the Redis connection (default: redis://localhost:6379/0, or redis://redis:6379/0 when using Docker).

Usage

Using Docker Compose

The easiest way to run Chart Sayer is using Docker Compose:

# Start the services
docker-compose up -d

# View logs
docker-compose logs -f

# Stop the services
docker-compose down

# Rebuild and restart after making changes
docker-compose up -d --build

This will start the FastAPI server on port 8000 and a Redis instance on port 6379. You can access the API at http://localhost:8000.

The setup also includes Redis Commander, a web-based Redis management tool, which you can access at http://localhost:8081.

Running the FastAPI Server Manually

To run the FastAPI server, ensure you have set up your environment variables in a .env file based on the .env.example. Then, execute the following command:

poetry run uvicorn src.main:app --host 0.0.0.0 --port 8000

This will start the server, allowing you to interact with the API endpoints for Telegram and Discord bot integration.

Command-Line Interface

Chart Sayer provides a convenient command-line interface for processing chart images without running the full server. You can use Poetry to run the CLI tool:

poetry run analyze-chart <image_path> <config_path>

For example:

poetry run analyze-chart src/chart_images/aave_test_image.png src/config/chart_config.json

This will analyze the chart image and output the extracted information based on your configuration:

Analysis Result: Entry: 162.46, Stop Loss: 161.82, Take Profit: 243.01

Alternatively, you can run the Python script directly:

poetry run python src/cli.py <image_path> <config_path>

Customization

Chart Sayer allows users to customize the analysis rules and output format through configuration files.

Configuration File

You can use a configuration file to define your settings. A sample chart_config.json file is provided in the src/config directory:

{
  "clients": [
    "discord",
    "telegram"
  ],
  "entry_color": "green",
  "stop_loss_color": "red",
  "take_profit_color": "blue",
  "output_format": "Entry: {entry}, Stop Loss: {stop_loss}, Take Profit: {take_profit}",
  "indicators": [
    "moving_average",
    "parabolic_sar"
  ],
  "image_processing": {
    "model": "gpt-4o",
    "max_tokens": 500
  }
}

Configuration Options

The configuration file allows you to customize various aspects of the bot's behavior and image processing. Here are the options you can set:

• clients: List of clients to integrate with, such as discord and telegram.
• entry_color: The color used to identify entry points on the chart.
• stop_loss_color: The color used to identify stop loss points.
• take_profit_color: The color used to identify take profit points.
• output_format: The format string for the output message, using placeholders like {entry}, {stop_loss}, and {take_profit}.
• indicators: List of indicators to use, such as moving_average and parabolic_sar.
• telegram: Configuration specific to the Telegram bot, including webhook_mode and webhook_url.
• discord: Configuration specific to the Discord bot, including gateway_mode and webhook_mode.
• image_processing: Settings for image processing, including the model and max_tokens.
• redis: Configuration for Redis, including enabled flag.

These options allow you to tailor the bot's functionality to your specific needs.

Contributing

We welcome contributions from the community. Please fork the repository and submit a pull request with your changes. For major changes, please open an issue first to discuss what you would like to change.