title: VNPY TUTORIAL 0x05 first strategy description: Vnpy Quant Tutorial, First Strategy keywords: Vnpy, Quant Tutorial, First Strategy date: 2024-09-05 category: "quant"
Warning: This page is translated by MACHINE, which may lead to POOR QUALITY or INCORRECT INFORMATION, please read with CAUTION!
"Learn Python the Hard Way" is also the first time I have learned Python's stepping stalls from a complete 0 -based Xiaobai many years ago. Yes, I won't come to me.
Strategic code in
- mac:/users/
/strategies/hackstrategy.py - Windows: Google
from vnpy_ctastrategy import (
CtaTemplate,
StopOrder,
TickData,
BarData,
TradeData,
OrderData,
BarGenerator,
ArrayManager,
)
class HackStrategy(CtaTemplate):
"" "演示用的简单双均线" ""
# 策略作者
author = "Smart Trader"
# 定义参数
fast_window = 10
slow_window = 20
# 定义变量
fast_ma0 = 0.0
fast_ma1 = 0.0
slow_ma0 = 0.0
slow_ma1 = 0.0
# 添加参数和变量名到对应的列表
parameters = ["fast_window", "slow_window"]
variables = ["fast_ma0", "fast_ma1", "slow_ma0", "slow_ma1"]
def __init__(self, cta_engine, strategy_name, vt_symbol, setting):
"" "" ""
super().__init__(cta_engine, strategy_name, vt_symbol, setting)
# K线合成器：从Tick合成分钟K线用
self.bg = BarGenerator(self.on_bar)
# 时间序列容器：计算技术指标用
self.am = ArrayManager()
def on_init(self):
"""
当策略被初始化时调用该函数。
"""
# 输出个日志信息，下同
self.write_log("策略初始化")
# 加载10天的历史数据用于初始化回放
self.load_bar(10)
def on_start(self):
"""
当策略被启动时调用该函数。
"""
self.write_log("策略启动")
# 通知图形界面更新（策略最新状态）
# 不调用该函数则界面不会变化
self.put_event()
def on_stop(self):
"""
当策略被停止时调用该函数。
"""
self.write_log("策略停止")
self.put_event()
def on_tick(self, tick: TickData):
"""
通过该函数收到Tick推送。
"""
self.bg.update_tick(tick)
def on_bar(self, bar: BarData):
"""
通过该函数收到新的1分钟K线推送。
"""
am = self.am
# 更新K线到时间序列容器中
am.update_bar(bar)
# 若缓存的K线数量尚不够计算技术指标，则直接返回
if not am.inited:
return
# 计算快速均线
fast_ma = am.sma(self.fast_window, array=True)
self.fast_ma0 = fast_ma[-1] # T时刻数值
self.fast_ma1 = fast_ma[-2] # T-1时刻数值
# 计算慢速均线
slow_ma = am.sma(self.slow_window, array=True)
self.slow_ma0 = slow_ma[-1]
self.slow_ma1 = slow_ma[-2]
# 判断是否金叉
cross_over = (self.fast_ma0 > self.slow_ma0 and
self.fast_ma1 < self.slow_ma1)
# 判断是否死叉
cross_below = (self.fast_ma0 < self.slow_ma0 and
self.fast_ma1 > self.slow_ma1)
# 如果发生了金叉
if cross_over:
# 为了保证成交，在K线收盘价上加5发出限价单
price = bar.close_price + 5
# 当前无仓位，则直接开多
if self.pos == 0:
self.buy(price, 1)
# 当前持有空头仓位，则先平空，再开多
elif self.pos < 0:
self.cover(price, 1)
self.buy(price, 1)
# 如果发生了死叉
elif cross_below:
price = bar.close_price - 5
# 当前无仓位，则直接开空
if self.pos == 0:
self.short(price, 1)
# 当前持有空头仓位，则先平多，再开空
elif self.pos > 0:
self.sell(price, 1)
self.short(price, 1)
self.put_event()
def on_order(self, order: OrderData):
"""
通过该函数收到委托状态更新推送。
"""
pass
def on_trade(self, trade: TradeData):
"""
通过该函数收到成交推送。
"""
# 成交后策略逻辑仓位发生变化，需要通知界面更新。
self.put_event()
def on_stop_order(self, stop_order: StopOrder):
"""
通过该函数收到本地停止单推送。
"""
pass
Click: Function-> Ctastrategy, you can find the strategy just now in the upper left corner.
Vnpy's strategic file storage location modification
Vnpy's strategic file storage position: /USERS/xucongyong/miniconda3/envs/vnpy_env/python3.10/site-packages/vnpy/engine.py
def load_strategy_class(self) -> None:
path1: Path = Path(__file__).parent.joinpath("strategies")
self.load_strategy_class_from_folder(path1, "vnpy_ctastrategy.strategies")
path2: Path = Path.cwd().joinpath("strategies")
self.load_strategy_class_from_folder(path2, "strategies")
path3: Path = '/Users/xucongyong/newstrategies/'
self.load_strategy_class_from_folder(path3, "strategies")