Bitcoin Signal Prediction

A project for building, training, and evaluating machine learning models to predict Bitcoin trading signals using Binance API data.

Story

Click Here! (Thai Version)

Features

• Collects historical price data from Binance API
• Calculates technical indicators such as MA, RSI, MACD, and Bollinger Bands
• Trains multiple ML models such as LR, RF, XGB, KNN, and SVM
• Compares performances, picking the best model, and simulates trading strategies

Setup

1. Clone this repository.

2. Install:

   pip install python-dotenv python-binance pandas numpy matplotlib scikit-learn xgboost imbalanced-learn
   

3. Create a .env file with your Binance API credentials:

   API_KEY=your_api_key
   API_SECRET=your_api_secret
   

Usage

• Run the notebook main.ipynb for step-by-step instructions.
• Follow the notebook cells to collect data, train models, and simulate trading.
• To simulate trading, call the function in a code cell:

  simulate("ETHBTC")

  Or leave it blank to select a symbol interactively:

  simulate()

References

• Binance API Documentation

Disclaimer

This project is for educational purposes only. No financial advice.