bbfamily/abu

Project overview

The project explicitly targets users who want to move beyond complex code-based quantitative workflows by combining machine learning strategy optimization, backtesting with execution-price deviation and fee accounting, and a non-programming UI in a single library.

Project type

Infrastructure

Deployment

Refer to project documentation

License

GPL-3.0

Best for

• Users who need quantitative trading capabilities across stocks, options, futures, and Bitcoin but want to avoid a complex code-heavy workflow
• Users seeking machine learning-based strategy optimization combined with backtesting that accounts for execution-price deviations and trading fees

Key capabilities

• Uses multiple machine learning techniques to optimize strategies and guide their use in live trading.
• Provides automatic support and resistance line drawing, gap analysis, and traditional technical-indicator analysis.
• Backtests market-timing strategies against historical data and accounts for execution-price deviations and trading fees.
• Includes an abupy_ui directory for operating the quantitative system through a non-programming interface.
• Supports switching data modes, storage, and sources and connecting external sources for stocks, futures, Bitcoin, and Litecoin.

Getting started

• Deploy a Python environment using Anaconda, then import abupy in Python to verify the installation. The complete installation sequence is not documented in the supplied section.

Evidence and sources

Open original GitHub project