Skip to content

Releases: Ronit26Mehta/YieldYatra-An-Autonomous-AI-Agent-for-DeFi-Trading-with-Aptos-Integration

1.0.0

07 Mar 14:53
@Ronit26Mehta Ronit26Mehta
1.0.0
3a5fce1
Compare
Choose a tag to compare
1.0.0 Latest
Latest

📦 YieldYatra Release Notes

🚀 Version 1.0.0 - Initial Release (March 7, 2025)

YieldYatra is an Autonomous AI-powered agent designed for Decentralized Finance (DeFi) trading, integrated with Aptos blockchain. This initial release introduces robust AI strategies, a fully functional backend server, interactive frontend for backtesting, and comprehensive documentation of implemented methodologies.

🌟 New Features

  • Autonomous AI Trading Algorithms:
    • Implemented advanced AI trading algorithms, including:
      • Kage no Suiri (Shadow Logic): GARCH and Hidden Markov Model (HMM) for volatility-based trading signals.
      • Kitsune no Kōsen (Fox’s Beam): CNN and Dynamic Time Warping (DTW) for predictive modeling.
      • Ryu no Riron (Dragon’s Theory): Fractal analysis using Lyapunov exponents and chaotic modeling.
      • Sakura no Kagami (Cherry Blossom Mirror): Mirror regression for trend reversal prediction.
      • Hikari no Suishin (Momentum Strategy): Principal Component Analysis (PCA)-driven momentum detection.
      • Tenshi no Shikaku (Angel’s Geometry): Topological data analysis for identifying support/resistance levels.
      • Zen no Ritsu (Zen Rhythm): Wavelet transforms for rhythm-based market analysis.

✨ Key Features

  • Autonomous Trading: Fully automated execution of trades based on AI-generated signals.
  • Advanced Metrics: Implements proprietary Yield Score and Risk Index metrics for optimal risk-adjusted portfolio management.
  • Aptos Blockchain Integration: On-chain transaction handling via the Aptos SDK.
  • Visualization Dashboard: Interactive charts and detailed performance analytics powered by Plotly and Streamlit.
  • Strategic Backtesting: Robust backtesting engine supporting both real-time (via CCXT) and historical (CSV) data.
  • Comprehensive Logging: Detailed trade logs for auditability and performance tracking.

🛠 Technical Stack

  • Backend: Flask, CCXT, Pandas, NumPy, SciPy, Plotly, Aptos SDK
  • Frontend: Streamlit, Plotly, Requests
  • Blockchain Integration: Aptos Python SDK
  • Visualization: Plotly, Matplotlib, Streamlit

📊 Interactive Dashboard

  • Easy-to-use frontend built with Streamlit, providing:
    • Interactive strategy configuration
    • Real-time performance visualization
    • Downloadable trade summaries and historical data

📖 Documentation

Included detailed theoretical and practical documentation:

🖥 Frontend Application

The frontend interface is built using Streamlit and includes:

  • Professional and intuitive UI for configuring and running trading backtests.
  • Advanced, interactive visualizations powered by Plotly.

📈 Strategies Implemented

The following strategies are integrated and ready for use:

Strategy Type Primary Technique
RSI Relative Strength Index-based signals  
MA Moving Average crossover signals  
RSI_MA Combined RSI and Moving Average strategy  
KAGE Volatility and Hidden Markov Models  
KITSUNE Neural Networks with DTW  
RYU Fractal dimension and Lyapunov exponent  
SAKURA Mirror regression modeling  
HIKARI Momentum-based principal components  
TENSHI Topological persistence analysis  
ZEN Wavelet-based rhythm analysis  
defAI AI-driven rebalancing and yield optimization  

🔗 Aptos Blockchain Integration

  • Secure, automated transaction execution using Aptos official Python SDK.
  • Real-time portfolio rebalancing triggered by AI-generated signals.

📍 Installation & Usage

To set up and run YieldYatra locally:

git clone https://github.com/yourusername/YieldYatra.git
cd YieldYatra
pip install -r requirements.txt

Start Backend


python backend/aptos_backend.py


Start Frontend


streamlit run frontend/aptos_frontend.py

🙌 Contribution Guidelines

We welcome community contributions! To contribute:

  1. Fork the repository.
  2. Create your feature branch (git checkout -b feature/amazing-feature).
  3. Commit changes and create a Pull Request.

📜 License

YieldYatra is licensed under MIT License. Feel free to modify, distribute, and contribute.

📌 Notes

  • Ensure backend server and Aptos blockchain credentials are configured.
  • Contact maintainers for any deployment or setup queries.

🌟 Happy Trading! 🌟

Here's a complete, professional, and structured GitHub release notes template (`RELEASE.md`) suitable for the YieldYatra project:

📦 YieldYatra Release Notes

🚀 Version 1.0.0 - Initial Release (March 7, 2025)

YieldYatra is an Autonomous AI-powered agent designed for Decentralized Finance (DeFi) trading, integrated with Aptos blockchain. This initial release introduces robust AI strategies, a fully functional backend server, interactive frontend for backtesting, and comprehensive documentation of implemented methodologies.

🌟 New Features

  • Autonomous AI Trading Algorithms:
    • Implemented advanced AI trading algorithms, including:
      • Kage no Suiri (Shadow Logic): GARCH and Hidden Markov Model (HMM) for volatility-based trading signals.
      • Kitsune no Kōsen (Fox’s Beam): CNN and Dynamic Time Warping (DTW) for predictive modeling.
      • Ryu no Riron (Dragon’s Theory): Fractal analysis using Lyapunov exponents and chaotic modeling.
      • Sakura no Kagami (Cherry Blossom Mirror): Mirror regression for trend reversal prediction.
      • Hikari no Suishin (Momentum Strategy): Principal Component Analysis (PCA)-driven momentum detection.
      • Tenshi no Shikaku (Angel’s Geometry): Topological data analysis for identifying support/resistance levels.
      • Zen no Ritsu (Zen Rhythm): Wavelet transforms for rhythm-based market analysis.

✨ Key Features

  • Autonomous Trading: Fully automated execution of trades based on AI-generated signals.
  • Advanced Metrics: Implements proprietary Yield Score and Risk Index metrics for optimal risk-adjusted portfolio management.
  • Aptos Blockchain Integration: On-chain transaction handling via the Aptos SDK.
  • Visualization Dashboard: Interactive charts and detailed performance analytics powered by Plotly and Streamlit.
  • Strategic Backtesting: Robust backtesting engine supporting both real-time (via CCXT) and historical (CSV) data.
  • Comprehensive Logging: Detailed trade logs for auditability and performance tracking.

🛠 Technical Stack

  • Backend: Flask, CCXT, Pandas, NumPy, SciPy, Plotly, Aptos SDK
  • Frontend: Streamlit, Plotly, Requests
  • Blockchain Integration: Aptos Python SDK
  • Visualization: Plotly, Matplotlib, Streamlit

📊 Interactive Dashboard

  • Easy-to-use frontend built with Streamlit, providing:
    • Interactive strategy configuration
    • Real-time performance visualization
    • Downloadable trade summaries and historical data

📖 Documentation

Included detailed theoretical and practical documentation:

  • [Aptos AI Agent](docs/Aptos%20AI%20Agent.pdf): Explanation of AI metrics (Yield Score, Risk Index), pseudo-code, and Aptos integration.
  • [Aptos Trader Implementation](docs/aptos%20trader.pdf): Detailed backend implementation strategies and pseudo-code.
  • [Japanese-Inspired Trading Strategies](docs/japanses-inspired-trading-strategy.pdf): Comprehensive theoretical and mathematical frameworks for each trading model.

🖥 Frontend Application

The frontend interface is built using Streamlit and includes:

  • Professional and intuitive UI for configuring and running trading backtests.
  • Advanced, interactive visualizations powered by Plotly.

📈 Strategies Implemented

The following strategies are integrated and ready for use:

Strategy Type Primary Technique
RSI Relative Strength Index-based signals
MA Moving Average crossover signals
RSI_MA Combined RSI and Moving Average strategy
KAGE Volatility and Hidden Markov Models
KITSUNE Neural Networks with DTW
RYU Fractal dimension and Lyapunov exponent
SAKURA Mirror regression modeling
HIKARI Momentum-based principal components
TENSHI Topological persistence analysis
ZEN Wavelet-based rhythm analysis
defAI AI...
Read more
Assets 2
Loading