Skip to content

lightyfr/BitcoinPricePredictor-LTSM

Repository files navigation

Bitcoin Price Predictor

An advanced deep learning model for predicting Bitcoin prices using LSTM neural networks and technical analysis.

Python Version TensorFlow License

🚀 Features

  • Real-time Bitcoin price data fetching via yfinance
  • Advanced technical indicators (MA, RSI, MACD)
  • LSTM neural network with dropout and batch normalization
  • Comprehensive visualization suite
  • Multiple performance metrics

📊 Latest Model Performance

Metric Value
MSE 51,790,828
RMSE 7,334
MAE 4,434
R² Score 0.863
Accuracy 91.67%

🛠️ Installation

Clone repository

git clone https://github.com/yourusername/bitcoin-price-predictor.git
cd bitcoin-price-predictor

Install dependencies

pip install -e .

Basic Usage

python -m bitcoin_predictor.main

Advanced Configuration

python -m bitcoin_predictor.main \
    --symbol BTC-USD \
    --period max \
    --sequence-length 60 \
    --batch-size 32 \
    --epochs 100

Running tests:

python -m unittest discover bitcoin_predictor/tests

Visualizations

The model generates several visualizations in the visualizations directory:

  • Training history plots
  • Prediction vs Actual comparisons
  • Feature correlation heatmaps
  • Error distribution analysis
  • Rolling metrics plots

Dependencies

  • numpy
  • pandas
  • scikit-learn
  • tensorflow
  • yfinance
  • matplotlib
  • seaborn
  • python-graphviz

Logging

The project uses Python's logging framework. Logs are stored in the logs directory with timestamped filenames.

Contributing

  1. Fork the repository
  2. Create a feature branch
  3. Commit your changes
  4. Push to the branch
  5. Create a Pull Request

License

This project is licensed under AGPL-3.0

About

No description, website, or topics provided.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages