克隆策略

RSI买入卖出¶

In [22]:

# 本代码由可视化策略环境自动生成 2021年11月15日 02:00
# 本代码单元只能在可视化模式下编辑。您也可以拷贝代码，粘贴到新建的代码单元或者策略，然后修改。


# 回测引擎：初始化函数，只执行一次
def m19_initialize_bigquant_run(context):
    # 加载预测数据
    context.ranker_prediction = context.options['data'].read_df()

    # 系统已经设置了默认的交易手续费和滑点，要修改手续费可使用如下函数
    context.set_commission(PerOrder(buy_cost=0.0003, sell_cost=0.0013, min_cost=5))
    # 预测数据，通过options传入进来，使用 read_df 函数，加载到内存 (DataFrame)
    # 设置买入的股票数量，这里买入预测股票列表排名靠前的5只
    stock_count = 50
    # 每只的股票的权重，如下的权重分配会使得靠前的股票分配多一点的资金，[0.339160, 0.213986, 0.169580, ..]
    context.stock_weights = T.norm([1 / math.log(i + 2) for i in range(0, stock_count)])
    # 设置每只股票占用的最大资金比例
    context.max_cash_per_instrument = 0.055
    context.options['hold_days'] = 2
    context.datecont = 0
  
    context.short_ma = 5 # 移动平均线指标参数
    context.long_ma = 30 
    context.short_macd = 2 # macd指标参数
    context.long_macd = 4
    context.smoothperiod = 4
    context.rsiperiod = 2
    from zipline.finance.slippage import SlippageModel
    class FixedPriceSlippage(SlippageModel):
        def process_order(self, data, order, bar_volume=0, trigger_check_price=0):
            if order.limit is None:
                price_field = self._price_field_buy if order.amount > 0 else self._price_field_sell
                price = data.current(order.asset, price_field)
            else:
                price = data.current(order.asset, self._price_field_buy)
            # 返回希望成交的价格和数量
            return (price, order.amount)
    # 设置price_field,默认是开盘买入，收盘卖出
    context.fix_slippage = FixedPriceSlippage(price_field_buy='open', price_field_sell='close')
    context.set_slippage(us_equities=context.fix_slippage)

def m19_handle_data_bigquant_run(context, data):
        
    import talib
        
    #获取当日日期
    today = data.current_dt.strftime('%Y-%m-%d')
    stock_hold_now = [equity.symbol for equity in context.portfolio.positions ]
    #大盘风控模块，读取风控数据    
    benckmark_risk=context.benckmark_risk[today]
    context.symbol
    #当risk为1时，市场有风险，全部平仓，不再执行其它操作
    if benckmark_risk > 0:
        for instrument in stock_hold_now:
            context.order_target(symbol(instrument), 0)
        print(today,'大盘风控止损触发,全仓卖出')
        return

    #------------------------------------------止损模块START--------------------------------------------
    equities = {e.symbol: p for e, p in context.portfolio.positions.items() if p.amount>0}
    
    # 新建当日止损股票列表是为了handle_data 策略逻辑部分不再对该股票进行判断
    stoploss_stock = [] 
    if len(equities) > 0:
        for i in equities.keys():
            stock_market_price = data.current(context.symbol(i), 'price')  # 最新市场价格
            last_sale_date = equities[i].last_sale_date   # 上次交易日期
            delta_days = data.current_dt - last_sale_date  
            hold_days = delta_days.days # 持仓天数
            # 建仓以来的最高价
            highest_price_since_buy = data.history(context.symbol(i), 'high', hold_days, '1d').max()
            # 确定止损位置
            stoploss_line = highest_price_since_buy - highest_price_since_buy * 0.025
            #record('止损位置', stoploss_line)
            # 如果价格下穿止损位置
            if stock_market_price < stoploss_line:
                context.order_target_percent(context.symbol(i), 0)     
                stoploss_stock.append(i)
        if len(stoploss_stock)>0:
            print('日期：', today, '股票：', stoploss_stock, '出现跟踪止损状况')
    #-------------------------------------------止损模块END---------------------------------------------    
   
    # 按日期过滤得到今日的预测数据
    ranker_prediction = context.ranker_prediction[
        context.ranker_prediction.date == data.current_dt.strftime('%Y-%m-%d')]
    
    intervals = []
    temp = 0
    rsi_predict = pd.read_csv("/home/bigquant/work/userlib/repeated.csv",encoding="utf-8",names=["M1","M2","M3","M4","M5","M6","M7","M8","M9","M10","M11","M12","M13"])    
    rsi_instruments=list(ranker_prediction.instrument[:len(context.stock_weights)])
    
    
    for i,instrument in enumerate(rsi_instruments):
        volume_since_buy = data.history(context.symbol(instrument), 'volume', 3, '1d')
        close_price = data.current(context.symbol(instrument), 'close')  #当收盘价
        high_price = data.current(context.symbol(instrument), 'high')  #当天最高价
        last_close_price = data.history(context.symbol(instrument), 'close', 1, '1d')#前一天收盘价
        last_open_price = data.history(context.symbol(instrument), 'open', 1, '1d')#前一天收盘价
        last_high_price = data.history(context.symbol(instrument), 'high', 1, '1d')#前一天收盘价
        last_low_price = data.history(context.symbol(instrument), 'low', 1, '1d')#前一天收盘价
        
        MA2 = data.history(context.symbol(instrument), 'price',2, '1d').mean() # 短期均线值
        MA5 = data.history(context.symbol(instrument), 'price',5, '1d').mean() # 长期均线值
        MA10 = data.history(context.symbol(instrument), 'price',10, '1d').mean() # 短期均线值
        MA21 = data.history(context.symbol(instrument), 'price',21, '1d').mean() # 长期均线值
        MA60 = data.history(context.symbol(instrument), 'price',60, '1d').mean() # 短期均线值
        MA120 = data.history(context.symbol(instrument), 'price',120, '1d').mean() # 长期均线值        
        MA250 = data.history(context.symbol(instrument), 'price',250, '1d').mean() # 短期均线值

        MAX21 = data.history(context.symbol(instrument), 'high',21, '1d').max() 
        MIN21 = data.history(context.symbol(instrument), 'low',21, '1d').min() 
        
        MAX10 = data.history(context.symbol(instrument), 'high',10, '1d').max() 
        MIN10 = data.history(context.symbol(instrument), 'low',10, '1d').min() 
     
        
        prices = data.history(context.symbol(instrument), 'price', context.long_ma, '1d') # 读取历史数据
        close_price_1 = data.history(context.symbol(instrument), 'close',30,'1d')  #当收盘价        
        rsi2 = talib.RSI(np.array(prices), timeperiod=2)
        CLOSEOPEN=last_close_price/last_open_price
        HIGHCLOSE=last_high_price/last_close_price
        MAX21LOW=MAX21/last_low_price
        MIN21HIGH=MIN21/last_high_price
        
        MAX10LOW=MAX21/last_low_price
        MIN10HIGH=MIN21/last_high_price

 
        data3 = pd.DataFrame()
        data3["M1"] = CLOSEOPEN
        data3["M2"] = HIGHCLOSE
        data3["M3"] = MAX21LOW
        data3["M4"] = MIN21HIGH    
        data3["M5"] = MAX10LOW
        data3["M6"] = MIN10HIGH   
        
        data3["M1"] = data3["M1"].apply(convert_number) 
        data3["M2"] = data3["M2"].apply(convert_number)
        data3["M3"] = data3["M3"].apply(convert_number)
        data3["M4"] = data3["M4"].apply(convert_number)
        data3["M5"] = data3["M5"].apply(convert_number)
        data3["M6"] = data3["M6"].apply(convert_number)   


    # 1. 资金分配
    # 平均持仓时间是hold_days，每日都将买入股票，每日预期使用 1/hold_days 的资金
    # 实际操作中，会存在一定的买入误差，所以在前hold_days天，等量使用资金；之后，尽量使用剩余资金（这里设置最多用等量的1.5倍）
    is_staging = context.trading_day_index < context.options['hold_days'] # 是否在建仓期间（前 hold_days 天）
    cash_avg = context.portfolio.portfolio_value / context.options['hold_days']
    cash_for_buy = min(context.portfolio.cash, (1 if is_staging else 1.5) * cash_avg)
    cash_for_sell = cash_avg - (context.portfolio.cash - cash_for_buy)
    positions = {e.symbol: p.amount * p.last_sale_price
                 for e, p in context.perf_tracker.position_tracker.positions.items()}

    # 2. 生成卖出订单：hold_days天之后才开始卖出；对持仓的股票，按机器学习算法预测的排序末位淘汰
    if not is_staging and cash_for_sell > 0:
        equities = {e.symbol: e for e, p in context.perf_tracker.position_tracker.positions.items()}
        instruments = list(reversed(list(ranker_prediction.instrument[ranker_prediction.instrument.apply(
                lambda x: x in equities and not context.has_unfinished_sell_order(equities[x]))])))
        # print('rank order for sell %s' % instruments)
        for instrument in instruments:
            context.order_target(context.symbol(instrument), 0)
            cash_for_sell -= positions[instrument]
            if cash_for_sell <= 0:
                break

    # 3. 生成买入订单：按机器学习算法预测的排序，买入前面的stock_count只股票
    buy_cash_weights = context.stock_weights
    buy_instruments = list(ranker_prediction.instrument[:len(buy_cash_weights)])
    max_cash_per_instrument = context.portfolio.portfolio_value * context.max_cash_per_instrument
    for i, instrument in enumerate(buy_instruments):
        cash = cash_for_buy * buy_cash_weights[i]
        if cash > max_cash_per_instrument - positions.get(instrument, 0):
            # 确保股票持仓量不会超过每次股票最大的占用资金量
            cash = max_cash_per_instrument - positions.get(instrument, 0)
        if cash > 0:
            context.order_value(context.symbol(instrument), cash)
            
def convert_number(x,intervals):
    #区间 0.025
    for i in range(len(intervals)):
        if x == intervals[i]:
            return (intervals[i], intervals[i+1])
        if x > intervals[i] and x< intervals[i+1]:
            return (round(intervals[i],3), round(intervals[i+1],3))

# 回测引擎：准备数据，只执行一次
def m19_prepare_bigquant_run(context):
    #在数据准备函数中一次性计算每日的大盘风控条件相比于在handle中每日计算风控条件可以提高回测速度
    # 多取50天的数据便于计算均值(保证回测的第一天均值不为Nan值)，其中context.start_date和context.end_date是回测指定的起始时间和终止时间
    start_date= (pd.to_datetime(context.start_date) - datetime.timedelta(days=50)).strftime('%Y-%m-%d') 
    df=DataSource('bar1d_index_CN_STOCK_A').read(start_date=start_date,end_date=context.end_date,fields=['close'])
    benckmark_data=df[df.instrument=='000001.HIX']
    #计算上证指数5日涨幅
    benckmark_data['ret5']=benckmark_data['close']/benckmark_data['close'].shift(5)-1
    #计算大盘风控条件，如果5日涨幅小于-4%则设置风险状态risk为1，否则为0
    benckmark_data['risk'] = np.where(benckmark_data['ret5']<-0.05,1,0)
    #修改日期格式为字符串(便于在handle中使用字符串日期索引来查看每日的风险状态)
    benckmark_data['date']=benckmark_data['date'].apply(lambda x:x.strftime('%Y-%m-%d'))
    #设置日期为索引
    benckmark_data.set_index('date',inplace=True)
    #把风控序列输出给全局变量context.benckmark_risk
    context.benckmark_risk=benckmark_data['risk']
    
    industry_start_date = context.start_date
    end_date = context.end_date
    # 获取目前的行业列表
    industry = list(set(D.history_data(D.instruments(), end_date, end_date, ['industry_sw_level1']).industry_sw_level1))
    # 获取行业指数的行情数据
    industry = ['SW'+str(j)+'.SHA' for j in industry]
    data = D.history_data(industry, industry_start_date, end_date, ['close','name'])

    # 获取股票名称 用于过滤st和退市股
    context.name_df = DataSource('instruments_CN_STOCK_A').read()
    # 获取涨跌停状态
    context.price_limit_status = DataSource('stock_status_CN_STOCK_A').read(fields=['price_limit_status'])

    
    # 计算此处每日动量较高的行业
    ret_data = data.groupby('instrument').apply(calcu_ret)
    ret_data.reset_index(inplace=True, drop=True)
    ret_data['date'] = ret_data['date'].map(lambda x:x.strftime('%Y-%m-%d'))
    context.daily_buy_industry = pd.Series({dt:seek_head_industry(ret_data.set_index('date').loc[dt]) for dt in list(set(ret_data.date))})

    # 每个交易日 每个行业的优质股 
    # 优质股的确定依据是：净资产收益率 (TTM)、营业收入同比增长率、归属母公司股东的净利润同比增长率
    features_data = D.features(D.instruments(industry_start_date, end_date), industry_start_date, end_date, ['fs_roe_ttm_0', 'fs_operating_revenue_yoy_0', 'fs_net_profit_yoy_0', 'industry_sw_level1_0'])
    # 整理出每个行业的优质股票
    context.daily_buy_stock = features_data.groupby(['date', 'industry_sw_level1_0']).apply(seek_head_stock)

# 计算不同周期的动量
def calcu_ret(df):
    df = df.sort_values('date')
    for i in [3, 9, 21, 42, 84, 126]: # 分别代表2月、4月、半年的动量
        df['ret_%s'%i] = df['close']/df['close'].shift(i)-1 
    return df

# 计算出得分
def seek_head_industry(df):
    for j in ['ret_3','ret_9','ret_21','ret_42','ret_84','ret_126']:
        df['%s'%j] = df['%s'%j].rank(ascending=True) 
    df['score'] =0.5*df['ret_3']+ 0.3*df['ret_9']+0.5*df['ret_21']+0.5*df['ret_42']+0.5*df['ret_84']+0.5*df['ret_126']  # 得分的权重分别为0.4、0.3、0.3
    result = df.sort_values('score', ascending=False)
    return list(result.instrument)[:3]  # 前3个行业

# 选出特定行业优质股票
def seek_head_stock(df):
    result = df.sort_values(['fs_roe_ttm_0', 'fs_net_profit_yoy_0', 'fs_operating_revenue_yoy_0'], ascending=False)
    return list(result.instrument[:10]) # 每个行业选10只股票
def m19_before_trading_start_bigquant_run(context, data):
    df_price_limit_status=context.price_limit_status.set_index('date')
    today=data.current_dt.strftime('%Y-%m-%d')
    # 得到当前未完成订单
    for orders in get_open_orders().values():
        # 循环，撤销订单
        for _order in orders:
            ins=str(_order.sid.symbol)
            if data.can_trade(_order.sid):
                #判断一下如果当日涨停，则取消卖单
                if  df_price_limit_status[df_price_limit_status.instrument==ins].price_limit_status.loc[today]>2 and _order.amount<0:
                    #cancel_order(_order)
                    print(today,'尾盘涨停',ins) 

m1 = M.instruments.v2(
    start_date=T.live_run_param('trading_date', '2010-01-01'),
    end_date=T.live_run_param('trading_date', '2019-01-01'),
    market='CN_STOCK_A',
    instrument_list='',
    max_count=0
)

m2 = M.advanced_auto_labeler.v2(
    instruments=m1.data,
    label_expr="""# #号开始的表示注释
# 0. 每行一个，顺序执行，从第二个开始，可以使用label字段
# 1. 可用数据字段见 https://bigquant.com/docs/data_history_data.html
#   添加benchmark_前缀，可使用对应的benchmark数据
# 2. 可用操作符和函数见 `表达式引擎 <https://bigquant.com/docs/big_expr.html>`_

# 计算收益：5日收盘价(作为卖出价格)除以明日开盘价(作为买入价格)
1*((shift(close, -3) / shift(open, -1))-1)

# 极值处理：用1%和99%分位的值做clip
clip(label, all_quantile(label, 0.01), all_quantile(label, 0.99))

# 将分数映射到分类，这里使用20个分类
all_wbins(label, 20)

# 过滤掉一字涨停的情况 (设置label为NaN，在后续处理和训练中会忽略NaN的label)
where(shift(high, -1) == shift(low, -1), NaN, label)
""",
    start_date='',
    end_date='',
    benchmark='000300.SHA',
    drop_na_label=True,
    cast_label_int=True
)

m3 = M.input_features.v1(
    features="""where((ta_rsi(close_1, timeperiod=2) > ta_rsi(close_0, timeperiod=2)), (correlation(max(close_0,20)/low_0,min(close_0,20)/high_0),21), (-1* 1))
"""
)

m15 = M.general_feature_extractor.v7(
    instruments=m1.data,
    features=m3.data,
    start_date='',
    end_date='',
    before_start_days=400
)

m16 = M.derived_feature_extractor.v3(
    input_data=m15.data,
    features=m3.data,
    date_col='date',
    instrument_col='instrument',
    drop_na=False,
    remove_extra_columns=False
)

m7 = M.join.v3(
    data1=m2.data,
    data2=m16.data,
    on='date,instrument',
    how='inner',
    sort=False
)

m5 = M.dropnan.v2(
    input_data=m7.data
)

m4 = M.stock_ranker_train.v6(
    training_ds=m5.data,
    features=m3.data,
    learning_algorithm='排序',
    number_of_leaves=30,
    minimum_docs_per_leaf=1000,
    number_of_trees=20,
    learning_rate=0.1,
    max_bins=1023,
    feature_fraction=1,
    data_row_fraction=1,
    ndcg_discount_base=1,
    m_lazy_run=False
)

m9 = M.instruments.v2(
    start_date=T.live_run_param('trading_date', '2019-01-01'),
    end_date=T.live_run_param('trading_date', '2021-11-11'),
    market='CN_STOCK_A',
    instrument_list='',
    max_count=0
)

m17 = M.general_feature_extractor.v7(
    instruments=m9.data,
    features=m3.data,
    start_date='',
    end_date='',
    before_start_days=400
)

m18 = M.derived_feature_extractor.v3(
    input_data=m17.data,
    features=m3.data,
    date_col='date',
    instrument_col='instrument',
    drop_na=False,
    remove_extra_columns=False
)

m11 = M.chinaa_stock_filter.v1(
    input_data=m18.data,
    index_constituent_cond=['全部'],
    board_cond=['上证主板', '深证主板', '创业板'],
    industry_cond=['传媒/信息服务', '公用事业', '农林牧渔', '化工', '医药生物', '商业贸易', '国防军工', '家用电器', '建筑材料/建筑建材', '有色金属', '机械设备', '汽车/交运设备', '电子', '电气设备', '计算机', '轻工制造', '通信', '采掘', '钢铁', '非银金融', '食品饮料'],
    st_cond=['正常'],
    delist_cond=['非退市'],
    output_left_data=False
)

m10 = M.dropnan.v2(
    input_data=m11.data
)

m8 = M.stock_ranker_predict.v5(
    model=m4.model,
    data=m10.data,
    m_lazy_run=False
)

m19 = M.trade.v4(
    instruments=m9.data,
    options_data=m8.predictions,
    start_date='',
    end_date='',
    initialize=m19_initialize_bigquant_run,
    handle_data=m19_handle_data_bigquant_run,
    prepare=m19_prepare_bigquant_run,
    before_trading_start=m19_before_trading_start_bigquant_run,
    volume_limit=0.025,
    order_price_field_buy='open',
    order_price_field_sell='close',
    capital_base=1000000,
    auto_cancel_non_tradable_orders=True,
    data_frequency='daily',
    price_type='真实价格',
    product_type='股票',
    plot_charts=True,
    backtest_only=False,
    benchmark=''
)

[2021-11-15 01:44:27.913563] INFO: moduleinvoker: instruments.v2 开始运行..

[2021-11-15 01:44:28.083414] INFO: moduleinvoker: 命中缓存

[2021-11-15 01:44:28.085190] INFO: moduleinvoker: instruments.v2 运行完成[0.171637s].

[2021-11-15 01:44:28.094072] INFO: moduleinvoker: advanced_auto_labeler.v2 开始运行..

[2021-11-15 01:44:32.843031] INFO: 自动标注(股票): 加载历史数据: 5414277 行

[2021-11-15 01:44:32.844840] INFO: 自动标注(股票): 开始标注 ..

[2021-11-15 01:44:39.578126] INFO: moduleinvoker: advanced_auto_labeler.v2 运行完成[11.48404s].

[2021-11-15 01:44:39.585948] INFO: moduleinvoker: input_features.v1 开始运行..

[2021-11-15 01:44:39.597810] INFO: moduleinvoker: 命中缓存

[2021-11-15 01:44:39.599204] INFO: moduleinvoker: input_features.v1 运行完成[0.013274s].

[2021-11-15 01:44:39.616109] INFO: moduleinvoker: general_feature_extractor.v7 开始运行..

[2021-11-15 01:44:40.441520] INFO: 基础特征抽取: 年份 2008, 特征行数=37738

[2021-11-15 01:44:41.774242] INFO: 基础特征抽取: 年份 2009, 特征行数=374574

[2021-11-15 01:44:43.363259] INFO: 基础特征抽取: 年份 2010, 特征行数=431567

[2021-11-15 01:44:45.137400] INFO: 基础特征抽取: 年份 2011, 特征行数=511455

[2021-11-15 01:44:46.813593] INFO: 基础特征抽取: 年份 2012, 特征行数=565675

[2021-11-15 01:44:48.597260] INFO: 基础特征抽取: 年份 2013, 特征行数=564168

[2021-11-15 01:44:50.434825] INFO: 基础特征抽取: 年份 2014, 特征行数=569948

[2021-11-15 01:44:52.125057] INFO: 基础特征抽取: 年份 2015, 特征行数=569698

[2021-11-15 01:44:54.678275] INFO: 基础特征抽取: 年份 2016, 特征行数=641546

[2021-11-15 01:44:57.458777] INFO: 基础特征抽取: 年份 2017, 特征行数=743233

[2021-11-15 01:45:00.636618] INFO: 基础特征抽取: 年份 2018, 特征行数=816987

[2021-11-15 01:45:01.744733] INFO: 基础特征抽取: 年份 2019, 特征行数=0

[2021-11-15 01:45:01.956174] INFO: 基础特征抽取: 总行数: 5826589

[2021-11-15 01:45:01.960796] INFO: moduleinvoker: general_feature_extractor.v7 运行完成[22.344677s].

[2021-11-15 01:45:01.967844] INFO: moduleinvoker: derived_feature_extractor.v3 开始运行..

[2021-11-15 01:45:16.055955] INFO: derived_feature_extractor: 提取完成 rsi_0 = ta_rsi(close_0, timeperiod=2), 5.290s

[2021-11-15 01:45:21.447282] INFO: derived_feature_extractor: 提取完成 rsi_1 = ta_rsi(close_1, timeperiod=2), 5.390s

[2021-11-15 01:45:21.456458] INFO: derived_feature_extractor: 提取完成 CLOSEOPEN=close_0/open_0, 0.008s

[2021-11-15 01:45:21.464879] INFO: derived_feature_extractor: 提取完成 HIGHCLOSE=high_0/close_0, 0.007s

[2021-11-15 01:45:21.978765] INFO: derived_feature_extractor: 提取完成 MAX21=max(close_0,20)/low_0, 0.513s

[2021-11-15 01:45:22.491840] INFO: derived_feature_extractor: 提取完成 MIN21=min(close_0,20)/high_0, 0.512s

[2021-11-15 01:45:23.003127] INFO: derived_feature_extractor: 提取完成 MAX10=max(close_0,10)/low_0, 0.510s

[2021-11-15 01:45:23.514556] INFO: derived_feature_extractor: 提取完成 MIN10=min(close_0,10)/high_0, 0.510s

[2021-11-15 01:45:27.311669] INFO: derived_feature_extractor: 提取完成 MA2 = mean(close_0,2), 3.796s

[2021-11-15 01:45:31.210108] INFO: derived_feature_extractor: 提取完成 MA5 = mean(close_0,5), 3.897s

[2021-11-15 01:45:35.142493] INFO: derived_feature_extractor: 提取完成 MA10 = mean(close_0,10), 3.931s

[2021-11-15 01:45:39.561922] INFO: derived_feature_extractor: 提取完成 MA21 = mean(close_0,21), 4.418s

[2021-11-15 01:45:43.701860] INFO: derived_feature_extractor: 提取完成 MA60 = mean(close_0,60), 4.138s

[2021-11-15 01:45:47.848810] INFO: derived_feature_extractor: 提取完成 MA120 = mean(close_0,120), 4.145s

[2021-11-15 01:45:52.136897] INFO: derived_feature_extractor: 提取完成 MA250 = mean(close_0,250), 4.286s

[2021-11-15 01:45:53.349757] INFO: derived_feature_extractor: /y_2008, 37738

[2021-11-15 01:45:53.923016] INFO: derived_feature_extractor: /y_2009, 374574

[2021-11-15 01:45:54.752296] INFO: derived_feature_extractor: /y_2010, 431567

[2021-11-15 01:45:55.707179] INFO: derived_feature_extractor: /y_2011, 511455

[2021-11-15 01:45:56.797075] INFO: derived_feature_extractor: /y_2012, 565675

[2021-11-15 01:45:58.004836] INFO: derived_feature_extractor: /y_2013, 564168

[2021-11-15 01:45:59.238957] INFO: derived_feature_extractor: /y_2014, 569948

[2021-11-15 01:46:00.444155] INFO: derived_feature_extractor: /y_2015, 569698

[2021-11-15 01:46:01.635465] INFO: derived_feature_extractor: /y_2016, 641546

[2021-11-15 01:46:03.063086] INFO: derived_feature_extractor: /y_2017, 743233

[2021-11-15 01:46:04.640715] INFO: derived_feature_extractor: /y_2018, 816987

[2021-11-15 01:46:05.320152] INFO: moduleinvoker: derived_feature_extractor.v3 运行完成[63.352288s].

[2021-11-15 01:46:05.329669] INFO: moduleinvoker: join.v3 开始运行..

[2021-11-15 01:46:14.412224] INFO: join: /y_2008, 行数=0/37738, 耗时=1.446808s

[2021-11-15 01:46:16.863996] INFO: join: /y_2009, 行数=0/374574, 耗时=2.449762s

[2021-11-15 01:46:19.517745] INFO: join: /y_2010, 行数=431036/431567, 耗时=2.650492s

[2021-11-15 01:46:22.359243] INFO: join: /y_2011, 行数=510928/511455, 耗时=2.837539s

[2021-11-15 01:46:25.664802] INFO: join: /y_2012, 行数=564588/565675, 耗时=3.301356s

[2021-11-15 01:46:28.910187] INFO: join: /y_2013, 行数=563145/564168, 耗时=3.240762s

[2021-11-15 01:46:32.242735] INFO: join: /y_2014, 行数=567880/569948, 耗时=3.327823s

[2021-11-15 01:46:35.523628] INFO: join: /y_2015, 行数=560435/569698, 耗时=3.276256s

[2021-11-15 01:46:38.902054] INFO: join: /y_2016, 行数=637477/641546, 耗时=3.373851s

[2021-11-15 01:46:42.705429] INFO: join: /y_2017, 行数=738265/743233, 耗时=3.797633s

[2021-11-15 01:46:46.697894] INFO: join: /y_2018, 行数=802867/816987, 耗时=3.986743s

[2021-11-15 01:46:46.810353] INFO: join: 最终行数: 5376621

[2021-11-15 01:46:46.836857] INFO: moduleinvoker: join.v3 运行完成[41.507182s].

[2021-11-15 01:46:46.847807] INFO: moduleinvoker: dropnan.v2 开始运行..

[2021-11-15 01:46:46.992828] INFO: dropnan: /y_2008, 0/0

[2021-11-15 01:46:47.018836] INFO: dropnan: /y_2009, 0/0

[2021-11-15 01:46:47.720556] INFO: dropnan: /y_2010, 363553/431036

[2021-11-15 01:46:48.553852] INFO: dropnan: /y_2011, 426883/510928

[2021-11-15 01:46:49.462686] INFO: dropnan: /y_2012, 508115/564588

[2021-11-15 01:46:50.397311] INFO: dropnan: /y_2013, 545486/563145

[2021-11-15 01:46:51.458258] INFO: dropnan: /y_2014, 552613/567880

[2021-11-15 01:46:52.463548] INFO: dropnan: /y_2015, 515849/560435

[2021-11-15 01:46:53.509292] INFO: dropnan: /y_2016, 597461/637477

[2021-11-15 01:46:54.737679] INFO: dropnan: /y_2017, 646319/738265

[2021-11-15 01:46:56.116822] INFO: dropnan: /y_2018, 738089/802867

[2021-11-15 01:46:56.313372] INFO: dropnan: 行数: 4894368/5376621

[2021-11-15 01:46:56.322745] INFO: moduleinvoker: dropnan.v2 运行完成[9.474926s].

[2021-11-15 01:46:56.332299] INFO: moduleinvoker: stock_ranker_train.v6 开始运行..

[2021-11-15 01:47:03.025767] INFO: StockRanker: 特征预处理 ..

[2021-11-15 01:47:11.205947] INFO: StockRanker: prepare data: training ..

[2021-11-15 01:47:19.349902] INFO: StockRanker: sort ..

[2021-11-15 01:48:20.235478] INFO: StockRanker训练: dc22c7a2 准备训练: 4894368 行数

[2021-11-15 01:48:20.237156] INFO: StockRanker训练: AI模型训练，将在4894368*15=7341.55万数据上对模型训练进行20轮迭代训练。预计将需要22～44分钟。请耐心等待。

[2021-11-15 01:48:20.471062] INFO: StockRanker训练: 正在训练 ..

[2021-11-15 01:48:20.520667] INFO: StockRanker训练: 任务状态: Pending

[2021-11-15 01:48:30.562063] INFO: StockRanker训练: 任务状态: Running

[2021-11-15 01:49:00.700051] INFO: StockRanker训练: 00:00:28.8195781, finished iteration 1

[2021-11-15 01:49:20.776379] INFO: StockRanker训练: 00:00:49.9125706, finished iteration 2

[2021-11-15 01:49:40.867098] INFO: StockRanker训练: 00:01:11.4219957, finished iteration 3

[2021-11-15 01:50:00.954496] INFO: StockRanker训练: 00:01:36.7384061, finished iteration 4

[2021-11-15 01:50:31.078748] INFO: StockRanker训练: 00:02:04.2477695, finished iteration 5

[2021-11-15 01:51:01.206201] INFO: StockRanker训练: 00:02:32.3117463, finished iteration 6

[2021-11-15 01:51:31.344090] INFO: StockRanker训练: 00:03:01.9379880, finished iteration 7

[2021-11-15 01:52:01.472908] INFO: StockRanker训练: 00:03:31.7846060, finished iteration 8

[2021-11-15 01:52:31.629686] INFO: StockRanker训练: 00:04:03.8006611, finished iteration 9

[2021-11-15 01:53:01.754473] INFO: StockRanker训练: 00:04:37.6559063, finished iteration 10

[2021-11-15 01:53:31.877748] INFO: StockRanker训练: 00:05:03.8231992, finished iteration 11

[2021-11-15 01:54:02.033886] INFO: StockRanker训练: 00:05:30.2059077, finished iteration 12

[2021-11-15 01:54:22.137765] INFO: StockRanker训练: 00:05:57.2497080, finished iteration 13

[2021-11-15 01:54:52.275963] INFO: StockRanker训练: 00:06:25.6178879, finished iteration 14

[2021-11-15 01:55:22.423269] INFO: StockRanker训练: 00:06:53.4935099, finished iteration 15

[2021-11-15 01:55:52.559177] INFO: StockRanker训练: 00:07:21.7218294, finished iteration 16

[2021-11-15 01:56:22.716325] INFO: StockRanker训练: 00:07:49.3531211, finished iteration 17

[2021-11-15 01:56:42.803226] INFO: StockRanker训练: 00:08:17.4164273, finished iteration 18

[2021-11-15 01:57:12.925857] INFO: StockRanker训练: 00:08:46.0558649, finished iteration 19

[2021-11-15 01:57:43.117618] INFO: StockRanker训练: 00:09:15.1899706, finished iteration 20

[2021-11-15 01:57:43.119138] INFO: StockRanker训练: 任务状态: Succeeded

[2021-11-15 01:57:43.296572] INFO: moduleinvoker: stock_ranker_train.v6 运行完成[646.964282s].

[2021-11-15 01:57:43.311440] INFO: moduleinvoker: instruments.v2 开始运行..

[2021-11-15 01:57:43.324925] INFO: moduleinvoker: 命中缓存

[2021-11-15 01:57:43.326499] INFO: moduleinvoker: instruments.v2 运行完成[0.015065s].

[2021-11-15 01:57:43.343062] INFO: moduleinvoker: general_feature_extractor.v7 开始运行..

[2021-11-15 01:57:44.981374] INFO: 基础特征抽取: 年份 2017, 特征行数=80874

[2021-11-15 01:57:48.100506] INFO: 基础特征抽取: 年份 2018, 特征行数=816396

[2021-11-15 01:57:51.770180] INFO: 基础特征抽取: 年份 2019, 特征行数=884867

[2021-11-15 01:57:55.603341] INFO: 基础特征抽取: 年份 2020, 特征行数=945961

[2021-11-15 01:57:59.124177] INFO: 基础特征抽取: 年份 2021, 特征行数=894681

[2021-11-15 01:57:59.300158] INFO: 基础特征抽取: 总行数: 3622779

[2021-11-15 01:57:59.307894] INFO: moduleinvoker: general_feature_extractor.v7 运行完成[15.964852s].

[2021-11-15 01:57:59.314713] INFO: moduleinvoker: derived_feature_extractor.v3 开始运行..

[2021-11-15 01:58:09.942589] INFO: derived_feature_extractor: 提取完成 rsi_0 = ta_rsi(close_0, timeperiod=2), 5.476s

[2021-11-15 01:58:15.782267] INFO: derived_feature_extractor: 提取完成 rsi_1 = ta_rsi(close_1, timeperiod=2), 5.838s

[2021-11-15 01:58:15.788937] INFO: derived_feature_extractor: 提取完成 CLOSEOPEN=close_0/open_0, 0.005s

[2021-11-15 01:58:15.795136] INFO: derived_feature_extractor: 提取完成 HIGHCLOSE=high_0/close_0, 0.005s

[2021-11-15 01:58:16.126488] INFO: derived_feature_extractor: 提取完成 MAX21=max(close_0,20)/low_0, 0.330s

[2021-11-15 01:58:16.494781] INFO: derived_feature_extractor: 提取完成 MIN21=min(close_0,20)/high_0, 0.365s

[2021-11-15 01:58:16.811860] INFO: derived_feature_extractor: 提取完成 MAX10=max(close_0,10)/low_0, 0.316s

[2021-11-15 01:58:17.133811] INFO: derived_feature_extractor: 提取完成 MIN10=min(close_0,10)/high_0, 0.320s

[2021-11-15 01:58:19.650040] INFO: derived_feature_extractor: 提取完成 MA2 = mean(close_0,2), 2.515s

[2021-11-15 01:58:22.134996] INFO: derived_feature_extractor: 提取完成 MA5 = mean(close_0,5), 2.483s

[2021-11-15 01:58:24.615014] INFO: derived_feature_extractor: 提取完成 MA10 = mean(close_0,10), 2.478s

[2021-11-15 01:58:27.067021] INFO: derived_feature_extractor: 提取完成 MA21 = mean(close_0,21), 2.451s

[2021-11-15 01:58:29.535275] INFO: derived_feature_extractor: 提取完成 MA60 = mean(close_0,60), 2.467s

[2021-11-15 01:58:32.037767] INFO: derived_feature_extractor: 提取完成 MA120 = mean(close_0,120), 2.501s

[2021-11-15 01:58:34.503954] INFO: derived_feature_extractor: 提取完成 MA250 = mean(close_0,250), 2.465s

[2021-11-15 01:58:35.370364] INFO: derived_feature_extractor: /y_2017, 80874

[2021-11-15 01:58:36.593339] INFO: derived_feature_extractor: /y_2018, 816396

[2021-11-15 01:58:38.374952] INFO: derived_feature_extractor: /y_2019, 884867

[2021-11-15 01:58:40.169887] INFO: derived_feature_extractor: /y_2020, 945961

[2021-11-15 01:58:41.965404] INFO: derived_feature_extractor: /y_2021, 894681

[2021-11-15 01:58:42.691895] INFO: moduleinvoker: derived_feature_extractor.v3 运行完成[43.377163s].

[2021-11-15 01:58:42.700486] INFO: moduleinvoker: chinaa_stock_filter.v1 开始运行..

[2021-11-15 01:58:43.473904] INFO: A股股票过滤: 过滤 /y_2017, 66806/0/80874

[2021-11-15 01:58:47.160754] INFO: A股股票过滤: 过滤 /y_2018, 674898/0/816396

[2021-11-15 01:58:51.152079] INFO: A股股票过滤: 过滤 /y_2019, 721770/0/884867

[2021-11-15 01:58:55.818535] INFO: A股股票过滤: 过滤 /y_2020, 739295/0/945961

[2021-11-15 01:59:00.354660] INFO: A股股票过滤: 过滤 /y_2021, 678361/0/894681

[2021-11-15 01:59:00.361374] INFO: A股股票过滤: 过滤完成, 2881130 + 0

[2021-11-15 01:59:00.400406] INFO: moduleinvoker: chinaa_stock_filter.v1 运行完成[17.699906s].

[2021-11-15 01:59:00.407940] INFO: moduleinvoker: dropnan.v2 开始运行..

[2021-11-15 01:59:00.602523] INFO: dropnan: /y_2017, 0/66806

[2021-11-15 01:59:01.006496] INFO: dropnan: /y_2018, 43242/674898

[2021-11-15 01:59:01.777438] INFO: dropnan: /y_2019, 675489/721770

[2021-11-15 01:59:02.592109] INFO: dropnan: /y_2020, 703256/739295

[2021-11-15 01:59:03.329870] INFO: dropnan: /y_2021, 622007/678361

[2021-11-15 01:59:03.449449] INFO: dropnan: 行数: 2043994/2881130

[2021-11-15 01:59:03.468495] INFO: moduleinvoker: dropnan.v2 运行完成[3.060549s].

[2021-11-15 01:59:03.476664] INFO: moduleinvoker: stock_ranker_predict.v5 开始运行..

[2021-11-15 01:59:04.066170] INFO: StockRanker预测: /y_2018 ..

[2021-11-15 01:59:05.135940] INFO: StockRanker预测: /y_2019 ..

[2021-11-15 01:59:07.384041] INFO: StockRanker预测: /y_2020 ..

[2021-11-15 01:59:09.586690] INFO: StockRanker预测: /y_2021 ..

[2021-11-15 01:59:13.900119] INFO: moduleinvoker: stock_ranker_predict.v5 运行完成[10.423437s].

[2021-11-15 01:59:13.978252] INFO: moduleinvoker: backtest.v8 开始运行..

[2021-11-15 01:59:13.984020] INFO: backtest: biglearning backtest:V8.5.0

[2021-11-15 02:00:17.920142] INFO: backtest: product_type:stock by specified

[2021-11-15 02:00:18.090322] INFO: moduleinvoker: cached.v2 开始运行..

[2021-11-15 02:00:26.616209] INFO: backtest: 读取股票行情完成:3881543

[2021-11-15 02:00:31.661502] INFO: moduleinvoker: cached.v2 运行完成[13.571214s].

[2021-11-15 02:00:34.680490] INFO: algo: TradingAlgorithm V1.8.5

[2021-11-15 02:00:36.009737] INFO: algo: trading transform...

[2021-11-15 02:00:37.783089] ERROR: moduleinvoker: module name: backtest, module version: v8, trackeback: zipline.errors.HistoryWindowStartsBeforeData: History window extends before 2018-01-02. To use this history window, start the backtest on or after 2019-01-11.

[2021-11-15 02:00:37.788131] ERROR: moduleinvoker: module name: trade, module version: v4, trackeback: zipline.errors.HistoryWindowStartsBeforeData: History window extends before 2018-01-02. To use this history window, start the backtest on or after 2019-01-11.

设置评估测试数据集，查看训练曲线
[视频教程]StockRanker训练曲线

bigcharts-data-start/{"__type":"tabs","__id":"bigchart-9bf5d570ad1c41a88db70c44901c2198"}/bigcharts-data-end

---------------------------------------------------------------------------
HistoryWindowStartsBeforeData             Traceback (most recent call last)
<ipython-input-22-5bbb6d72813f> in <module>
    404 )
    405 
--> 406 m19 = M.trade.v4(
    407     instruments=m9.data,
    408     options_data=m8.predictions,

<ipython-input-22-5bbb6d72813f> in m19_handle_data_bigquant_run(context, data)
    106         MA60 = data.history(context.symbol(instrument), 'price',60, '1d').mean() # 短期均线值
    107         MA120 = data.history(context.symbol(instrument), 'price',120, '1d').mean() # 长期均线值
--> 108         MA250 = data.history(context.symbol(instrument), 'price',250, '1d').mean() # 短期均线值
    109 
    110         MAX21 = data.history(context.symbol(instrument), 'high',21, '1d').max()

HistoryWindowStartsBeforeData: History window extends before 2018-01-02. To use this history window, start the backtest on or after 2019-01-11.