Skip to main content

High-performance quantitative trading framework based on Rust and Python

Project description

AKQuant

PyPI Version Python Versions License AKShare Downloads

AKQuant 是一款专为量化投研设计的下一代高性能混合框架。核心引擎采用 Rust 编写以确保极致的执行效率,同时提供优雅的 Python 接口以维持灵活的策略开发体验。

🚀 核心亮点:

  • 极致性能:得益于 Rust 的零开销抽象与 Zero-Copy 数据架构,回测速度较传统纯 Python 框架(如 Backtrader)提升 X倍+
  • 原生 ML 支持:内置 Walk-forward Validation(滚动训练)框架,无缝集成 PyTorch/Scikit-learn,让 AI 策略开发从实验到回测一气呵成。
  • 因子表达式引擎:内置 Polars 驱动的高性能因子计算引擎,支持 Rank(Ts_Mean(Close, 5)) 等 Alpha101 风格公式,自动处理并行计算与数据对齐。
  • 参数优化:内置多进程网格搜索(Grid Search)框架,支持策略参数的高效并行优化。
  • 专业级风控:内置完善的订单流管理与即时风控模块,支持多资产组合回测。

👉 阅读完整文档 | English Documentation

安装说明

AKQuant 已发布至 PyPI,无需安装 Rust 环境即可直接使用。

pip install akquant

快速开始

以下是一个简单的策略示例:

import akquant as aq
import akshare as ak
from akquant import Strategy

# 1. 准备数据
# 使用 akshare 获取 A 股历史数据 (需安装: pip install akshare)
df = ak.stock_zh_a_daily(symbol="sh600000", start_date="20250212", end_date="20260212")


class MyStrategy(Strategy):
    def on_bar(self, bar):
        # 简单策略示例:
        # 当收盘价 > 开盘价 (阳线) -> 买入
        # 当收盘价 < 开盘价 (阴线) -> 卖出

        # 获取当前持仓
        current_pos = self.get_position(bar.symbol)

        if current_pos == 0 and bar.close > bar.open:
            self.buy(symbol=bar.symbol, quantity=100)
            print(f"[{bar.timestamp_str}] Buy 100 at {bar.close:.2f}")

        elif current_pos > 0 and bar.close < bar.open:
            self.close_position(symbol=bar.symbol)
            print(f"[{bar.timestamp_str}] Sell 100 at {bar.close:.2f}")


# 运行回测
result = aq.run_backtest(
    data=df,
    strategy=MyStrategy,
    initial_cash=100000.0,
    symbol="sh600000"
)

# 打印回测结果
print("\n=== Backtest Result ===")
print(result)

运行结果示例:

=== Backtest Result ===
BacktestResult:
                                            Value
start_time              2025-02-12 00:00:00+08:00
end_time                2026-02-12 00:00:00+08:00
duration                        365 days, 0:00:00
total_bars                                    249
trade_count                                  62.0
initial_market_value                     100000.0
end_market_value                          99804.0
total_pnl                                  -196.0
unrealized_pnl                                0.0
total_return_pct                           -0.196
annualized_return                        -0.00196
volatility                               0.002402
total_profit                                548.0
total_loss                                 -744.0
total_commission                              0.0
max_drawdown                                345.0
max_drawdown_pct                         0.344487
win_rate                                22.580645
loss_rate                               77.419355
winning_trades                               14.0
losing_trades                                48.0
avg_pnl                                  -3.16129
avg_return_pct                          -0.199577
avg_trade_bars                           1.967742
avg_profit                              39.142857
avg_profit_pct                           3.371156
avg_winning_trade_bars                        4.5
avg_loss                                    -15.5
avg_loss_pct                            -1.241041
avg_losing_trade_bars                    1.229167
largest_win                                 120.0
largest_win_pct                         10.178117
largest_win_bars                              7.0
largest_loss                                -70.0
largest_loss_pct                        -5.380477
largest_loss_bars                             1.0
max_wins                                      2.0
max_losses                                    9.0
sharpe_ratio                            -0.816142
sortino_ratio                           -1.066016
profit_factor                            0.736559
ulcer_index                              0.001761
upi                                     -1.113153
equity_r2                                0.399577
std_error                                68.64863
calmar_ratio                            -0.568962
exposure_time_pct                       48.995984
var_95                                   -0.00023
var_99                                   -0.00062
cvar_95                                 -0.000405
cvar_99                                  -0.00069
sqn                                     -0.743693
kelly_criterion                         -0.080763
max_leverage                              0.01458
min_margin_level                        68.587671

复杂订单助手 (OCO / Bracket)

AKQuant 提供了两组复杂订单助手,减少手写订单联动逻辑:

  • create_oco_order_group(first_order_id, second_order_id, group_id=None):将两个订单绑定为 OCO,任一成交后自动撤销另一单。
  • place_bracket_order(symbol, quantity, entry_price=None, stop_trigger_price=None, take_profit_price=None, ...):一次性提交 Bracket 结构;进场成交后自动挂出止损/止盈,并在双退出单场景下自动绑定 OCO。
from akquant import OrderStatus, Strategy

class BracketHelperStrategy(Strategy):
    def __init__(self):
        self.entry_order_id = ""

    def on_bar(self, bar):
        if self.get_position(bar.symbol) > 0 or self.entry_order_id:
            return

        self.entry_order_id = self.place_bracket_order(
            symbol=bar.symbol,
            quantity=100,
            stop_trigger_price=bar.close * 0.98,
            take_profit_price=bar.close * 1.04,
            entry_tag="entry",
            stop_tag="stop",
            take_profit_tag="take",
        )

    def on_order(self, order):
        if order.id == self.entry_order_id and order.status in (
            OrderStatus.Cancelled,
            OrderStatus.Rejected,
        ):
            self.entry_order_id = ""

可直接运行完整示例:

python examples/06_complex_orders.py

流式回测 (Streaming)

如果你希望在回测执行过程中实时消费事件,可直接使用 run_backtest 并传入 on_event

def on_event(event):
    if event["event_type"] == "finished":
        payload = event["payload"]
        print("status:", payload.get("status"))
        print("callback_error_count:", payload.get("callback_error_count"))

result = aq.run_backtest(
    data=df,
    strategy=MyStrategy,
    symbol="sh600000",
    on_event=on_event,
    show_progress=False,
    stream_progress_interval=10,
    stream_equity_interval=10,
    stream_batch_size=32,
    stream_max_buffer=256,
    stream_error_mode="continue",
)

run_backtest 也支持可选 on_event。如果不传,框架会使用内部 no-op 回调, 保持与传统阻塞调用一致的返回语义;如果传入 on_event,即可在保持 run_backtest(...) 调用方式不变的同时接入实时事件。

关键参数:

  • stream_progress_interval / stream_equity_interval: 进度与权益事件采样间隔
  • stream_batch_size / stream_max_buffer: 缓冲与批量刷新控制
  • stream_error_mode: 回调异常策略,支持 "continue""fail_fast"

阶段 5 迁移 FAQ:

  • run_backtest 是否改名?不改名,调用方式保持不变。
  • run_backtest 是否还能不传 on_event?可以,不传时仍返回同样的结果对象语义。
  • 如何回滚?阶段 5 后不再支持 _engine_mode 参数级回滚,建议使用版本级回滚。
  • 文档入口在哪里?可从 中文文档首页的阶段 5 迁移入口 快速跳转到 quickstart FAQ 和 API 兼容说明。
  • 英文入口在哪里?可从 English docs Quick Links 跳转到 Phase-5 Migration FAQ 与 Compatibility Notes。

可视化 (Visualization)

AKQuant 内置了基于 Plotly 的强大可视化模块,仅需一行代码即可生成包含权益曲线、回撤分析、月度热力图等详细指标的交互式 HTML 报告。

# 生成交互式 HTML 报告,自动在浏览器中打开
result.report(
    show=True,
    compact_currency=True,  # 金额列按 K/M/B 紧凑显示(默认 True)
)

# 如果你希望金额列保留原始数值精度(不缩写),可关闭:
result.report(
    show=False,
    filename="report_raw_amount.html",
    compact_currency=False,
)

你也可以直接复用结构化分析结果做二次研究:

exposure = result.exposure_df()             # 暴露分解(净暴露/总暴露/杠杆)
attr_by_symbol = result.attribution_df(by="symbol")
attr_by_tag = result.attribution_df(by="tag")
capacity = result.capacity_df()             # 容量代理(成交率/换手等)
orders_by_strategy = result.orders_by_strategy()         # 按策略归属聚合订单
executions_by_strategy = result.executions_by_strategy() # 按策略归属聚合成交

Strategy Dashboard
👉 点击查看交互式报表示例 (Interactive Demo)

文档索引

🧪 测试与质量保证

AKQuant 采用严格的测试流程以确保回测引擎的准确性:

  • 单元测试: 覆盖核心 Rust 组件与 Python 接口。
  • 黄金测试 (Golden Tests): 使用合成数据验证关键业务逻辑(如 T+1、涨跌停、保证金、期权希腊值),并与锁定的基线结果进行比对,防止算法回退。

运行测试:

# 1. 安装开发依赖
pip install -e ".[dev]"

# 2. 运行所有测试
pytest

# 3. 运行 Rust 核心测试(自动处理 macOS + conda 动态库路径)
./scripts/cargo-test.sh -q

# 4. 仅运行黄金测试
pytest tests/golden/test_golden.py

贡献指南

Citation

Please use this bibtex if you want to cite this repository in your publications:

@misc{akquant,
    author = {Albert King and Yaojie Zhang and Chao Liang},
    title = {AKQuant},
    year = {2026},
    publisher = {GitHub},
    journal = {GitHub repository},
    howpublished = {\url{https://github.com/akfamily/akquant}},
}

License

MIT License

Project details


Release history Release notifications | RSS feed

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

akquant-0.1.64.tar.gz (1.1 MB view details)

Uploaded Source

Built Distributions

If you're not sure about the file name format, learn more about wheel file names.

akquant-0.1.64-pp311-pypy311_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (5.8 MB view details)

Uploaded PyPymanylinux: glibc 2.17+ x86-64

akquant-0.1.64-cp310-abi3-win_amd64.whl (5.8 MB view details)

Uploaded CPython 3.10+Windows x86-64

akquant-0.1.64-cp310-abi3-musllinux_1_2_aarch64.whl (5.6 MB view details)

Uploaded CPython 3.10+musllinux: musl 1.2+ ARM64

akquant-0.1.64-cp310-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (5.8 MB view details)

Uploaded CPython 3.10+manylinux: glibc 2.17+ x86-64

akquant-0.1.64-cp310-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl (5.4 MB view details)

Uploaded CPython 3.10+manylinux: glibc 2.17+ ARM64

akquant-0.1.64-cp310-abi3-macosx_11_0_arm64.whl (5.2 MB view details)

Uploaded CPython 3.10+macOS 11.0+ ARM64

File details

Details for the file akquant-0.1.64.tar.gz.

File metadata

  • Download URL: akquant-0.1.64.tar.gz
  • Upload date:
  • Size: 1.1 MB
  • Tags: Source
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.13.7

File hashes

Hashes for akquant-0.1.64.tar.gz
Algorithm Hash digest
SHA256 15e754f6d53e2abd6464c52c68948e41e1aa5bfe5465e8d8f2c69b1a3c3154e5
MD5 21f5fade705ff031b1cbfad2aafa3ac8
BLAKE2b-256 bcceed4ac435162fea0a746a4639add2aeab741f7522a613ed685ae4f5b81557

See more details on using hashes here.

Provenance

The following attestation bundles were made for akquant-0.1.64.tar.gz:

Publisher: release.yml on akfamily/akquant

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file akquant-0.1.64-pp311-pypy311_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.

File metadata

File hashes

Hashes for akquant-0.1.64-pp311-pypy311_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl
Algorithm Hash digest
SHA256 73d6493572d15afe3dd0c82bcd7c50f7f2f0ab2a85df4410dd3e631a37922488
MD5 9a44b562f316e2cfb3c3e191d9b100de
BLAKE2b-256 e35515923d8345cb140f7cf5e3ffef24abd2984607e82cbc878e04f032f659df

See more details on using hashes here.

Provenance

The following attestation bundles were made for akquant-0.1.64-pp311-pypy311_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl:

Publisher: release.yml on akfamily/akquant

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file akquant-0.1.64-cp310-abi3-win_amd64.whl.

File metadata

  • Download URL: akquant-0.1.64-cp310-abi3-win_amd64.whl
  • Upload date:
  • Size: 5.8 MB
  • Tags: CPython 3.10+, Windows x86-64
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.13.7

File hashes

Hashes for akquant-0.1.64-cp310-abi3-win_amd64.whl
Algorithm Hash digest
SHA256 aca1a8719abcf38453e98697cff5b3070dcf33ed356fcc3a8018324a70857ff0
MD5 91587f2ff7dbfd63a5b3537286c173ea
BLAKE2b-256 7e4fd6a573f04c264e5da5956f8dbada96c852384ab8f64bd41bc56127598bcd

See more details on using hashes here.

Provenance

The following attestation bundles were made for akquant-0.1.64-cp310-abi3-win_amd64.whl:

Publisher: release.yml on akfamily/akquant

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file akquant-0.1.64-cp310-abi3-musllinux_1_2_aarch64.whl.

File metadata

File hashes

Hashes for akquant-0.1.64-cp310-abi3-musllinux_1_2_aarch64.whl
Algorithm Hash digest
SHA256 e73613950c81033c2c44295dcaaf89590e032ed3f8e956cb8b0a233343720efe
MD5 5371b8f5e40a39decaf6053591584d8c
BLAKE2b-256 34cea60c81d692576f7a6b1dc4fd3d5b46f24b97bec9ab1cb04be9b4e5af030e

See more details on using hashes here.

Provenance

The following attestation bundles were made for akquant-0.1.64-cp310-abi3-musllinux_1_2_aarch64.whl:

Publisher: release.yml on akfamily/akquant

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file akquant-0.1.64-cp310-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.

File metadata

File hashes

Hashes for akquant-0.1.64-cp310-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl
Algorithm Hash digest
SHA256 1e972a98e0d091cd77ed434b25bba63b72a8de2a15cb338fdaa340d96733af9f
MD5 0bd4766ccc4fbc3ab3c7182faf7b1c2e
BLAKE2b-256 9d11ee8e995c288b1628dd4831d4bed30e650bfd990942a5a520decb088e1f30

See more details on using hashes here.

Provenance

The following attestation bundles were made for akquant-0.1.64-cp310-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl:

Publisher: release.yml on akfamily/akquant

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file akquant-0.1.64-cp310-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl.

File metadata

File hashes

Hashes for akquant-0.1.64-cp310-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl
Algorithm Hash digest
SHA256 2b151e4b8fd3635b99cef65aca46b9ae78e007c36709b67fcd5523885344194f
MD5 3d96ec085b6c933828485a9e401bb2a7
BLAKE2b-256 418d5c626c5331e5c0cc4596b3153d7ac3979ad012d50c5d3b64d397b0cc354e

See more details on using hashes here.

Provenance

The following attestation bundles were made for akquant-0.1.64-cp310-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl:

Publisher: release.yml on akfamily/akquant

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file akquant-0.1.64-cp310-abi3-macosx_11_0_arm64.whl.

File metadata

File hashes

Hashes for akquant-0.1.64-cp310-abi3-macosx_11_0_arm64.whl
Algorithm Hash digest
SHA256 307d11480fe34d041d1ef633fb84be313f01f563b74cceb8f1320d916a41032f
MD5 875373e5a027f0d5be6ec277c90c15b8
BLAKE2b-256 2968f5cff2a3bc7e28d9a6e172cc985bd70a4c9c9b11d2147ee800bf0c1c73e2

See more details on using hashes here.

Provenance

The following attestation bundles were made for akquant-0.1.64-cp310-abi3-macosx_11_0_arm64.whl:

Publisher: release.yml on akfamily/akquant

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page