How akfamily/akquant works
AKQuant is a hybrid Rust/Python quantitative research and trading framework. Its principal workflow is to prepare market data, run a Python strategy through the backtest interface and Rust engine, apply risk and execution rules, then inspect results and reports. The repository also exposes live trading and broker integrations, streaming events, complex orders, walk-forward ML, factor expressions, TA-Lib indicators, and parameter optimization. The README documents user-facing capabilities; sampled source and the SOURCE INDEX support selected internal relationships, so uncertain runtime wiring is omitted.
Main entry points: core.rs Rust Engine · engine.py Backtest Runner · manager.rs Risk Controls · order_manager.rs Order Management
Components 24
Python Interface
- AKQuant API
python/akquant/__init__.py - Data Preparation
python/akquant/data.py - Strategy API
python/akquant/strategy.py - Backtest Runner
python/akquant/backtest/engine.py
Trading Runtime
- Rust Engine
src/engine/core.rs - Market Models
src/model/instrument.rs - Event Types
src/event.rs - Market History
src/history.rs - Risk Controls
src/risk/manager.rs - Order Management
src/order_manager.rs - Execution Models
src/execution/mod.rs - Portfolio State
src/portfolio.rs - Market Rules
src/market/manager.rs
Research Tools
- Factor Engine
python/akquant/factor/engine.py - TA-Lib Indicators
python/akquant/talib/funcs.py - Walk-forward ML
python/akquant/ml/model.py - Parameter Search
python/akquant/optimize.py
Live Connectivity
- Live Runner
python/akquant/live/_runner.py - Broker Gateway
python/akquant/gateway/factory.py
Results
- Trade Analysis
src/analysis/tracker.rs - Result Reports
python/akquant/backtest/result.py - Event Streaming
python/akquant/trade_stream.py
Other components
- Quant Developer
- Market Data Provider
Connections 28
- Quant Developer → AKQuant API: calls
- Market Data Provider → Data Preparation: provides data
- AKQuant API → Backtest Runner: runs backtest
python/akquant/__init__.py - Data Preparation → Backtest Runner: supplies input
python/akquant/backtest/engine.py - Backtest Runner → Strategy API: loads strategy
python/akquant/backtest/engine.py - Backtest Runner → Rust Engine: invokes engine
- Rust Engine → Event Types: emits events
- Rust Engine → Market History: uses history
src/engine/mod.rs - Rust Engine → Order Management: uses orders
src/engine/mod.rs - Rust Engine → Portfolio State: uses portfolio
src/engine/mod.rs - Rust Engine → Market Models: uses models
src/engine/mod.rs - Rust Engine → Risk Controls: checks orders
- Rust Engine → Execution Models: dispatches execution
- Rust Engine → Market Rules: applies rules
- Order Management → Portfolio State: updates positions
src/order_manager.rs - Order Management → Market Rules: calculates fees
src/order_manager.rs - Risk Controls → Portfolio State: checks exposure
src/risk/manager.rs - Risk Controls → Market Rules: checks market
src/risk/manager.rs - Rust Engine → Trade Analysis: tracks trades
- Backtest Runner → Result Reports: returns results
python/akquant/backtest/engine.py - Backtest Runner → Event Streaming: emits progress
- Quant Developer → Factor Engine: computes factors
- Quant Developer → TA-Lib Indicators: uses indicators
- Quant Developer → Walk-forward ML: trains models
- Quant Developer → Parameter Search: optimizes parameters
- Quant Developer → Live Runner: starts trading
- Live Runner → Broker Gateway: configures broker
- Broker Gateway → Market Data Provider: connects broker