复制链接

克隆策略

In [ ]:

# 本代码由可视化策略环境自动生成 2022年10月28日 11:43
# 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 m20_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 = 1
    # 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 = [1]
    # 设置每只股票占用的最大资金比例
    context.max_cash_per_instrument = 1
    context.options['hold_days'] = 1

# 回测引擎：每日数据处理函数，每天执行一次
def m20_handle_data_bigquant_run(context, data):
    # 获取当前持仓
    positions = {e.symbol: p.amount * p.last_sale_price
                 for e, p in context.portfolio.positions.items()}
    
    today = data.current_dt.strftime('%Y-%m-%d')
    # 按日期过滤得到今日的预测数据
    ranker_prediction = context.ranker_prediction[
        context.ranker_prediction.date == today]
    try:
    #大盘风控模块，读取风控数据    
        benckmark_risk=ranker_prediction['bm_0'].values[0]
        if benckmark_risk > 0:
            for instrument in positions.keys():
                context.order_target(context.symbol(instrument), 0)
                print(today,'大盘风控止损触发,全仓卖出')
                return
    except:
        print('--！')
        
    #当risk为1时，市场有风险，全部平仓，不再执行其它操作    
    # 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')]

    # 通过positions对象，使用列表生成式的方法获取目前持仓的股票列表
    
    #------------------------获取 持仓信息 环节--------------------------
    # 先定义 我们要用来买卖股票的资金
    cash_for_buy = context.portfolio.cash
    #获取 我们模型今天预测的股票池
    buy_instruments = list(ranker_prediction.instrument)
    #找到我们当前的股票持仓
    current_hold_stock = [equity.symbol for equity in context.portfolio.positions ]
    #定义 一个 列表 用来储存我们今天要卖出的股票
    sell_instruments = [instrument.symbol for instrument in context.portfolio.positions.keys()]
    
    

    #----逻辑上 先卖 后买，防止资金不足---产生空单
    # 今天需要卖出的股票  存在于我们  当前的股票持仓中  
    totay_to_sell = [i for i in sell_instruments[:1] ]#这里 因为我们只有1只股票 所以可以直接卖掉 
    
    #使用一个for循环  将持仓的股票全部卖出
    for instrument in totay_to_sell:
        context.order_target(context.symbol(instrument), 0)
    # 今天需要买入的股票  存在于我们  模型当天预测的股票池 buy_instruments 中  
   
    
    totay_to_buy = [i for i in buy_instruments[:1] ]#这里 我们只买 排名最靠前的第一名   
    # 如果想买入多只股票怎么操作呢？------
    #totay_to_sell = [i for i in sell_instruments[:N] ] N=你想要买入的股票数量，比如我想买2只 我就把N改成2
    #使用一个for循环  将预测的股票前  N名  买入
    #为了方便统计，我们直接用所有的钱下单，all in 当天买入的股票
    for instrument in totay_to_buy:

        context.order_value(context.symbol(instrument),cash_for_buy)
    
    

def m20_prepare_bigquant_run(context):


     # 获取st状态和涨跌停状态
    
    context.status_df = D.features(instruments =context.instruments,start_date = context.start_date, end_date = context.end_date, 
                           fields=['st_status_0','price_limit_status_0','price_limit_status_1'])

def m20_before_trading_start_bigquant_run(context, data):
    pass     
#     # 获取涨跌停状态数据
#     df_price_limit_status=context.status_df.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)
#             try:
#                 #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_0.loc[today]>2 and _order.amount<0:
#                     cancel_order(_order)
#                     print(today,'尾盘涨停取消卖单',ins) 
#             except:
#                 continue
  
    
    

m1 = M.instruments.v2(
    start_date='2008-10-20',
    end_date='2021-03-15',
    market='CN_STOCK_A',
    instrument_list='',
    max_count=0
)

m2 = M.advanced_auto_labeler.v2(
    instruments=m1.data,
    label_expr="""# 计算收益：2日收盘价(作为卖出价格)除以明日开盘价(作为买入价格)
shift(close, -2) / 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30个分类
all_wbins(label, 30)""",
    start_date='',
    end_date='',
    benchmark='000300.HIX',
    drop_na_label=True,
    cast_label_int=True
)

m3 = M.input_features.v1(
    features="""std(turn_0,10)
avg_turn_5
rank_swing_volatility_60_0
fe1=sum(max(0,high_0-delay((high_0+low_0+close_0)/3,1)),26)/sum(max(0,delay((high_0+low_0+close_0)/3,1)-1),26)*100
mean(turn_0*(return_0-1),30)
std(return_0-1,60)
std(return_0-1,15)
fe2=((close_0-sum(min(low_0,delay(close_0,1)),6))/sum(max(high_0,delay(close_0,1))-min(low_0,delay(close_0,1)),6)*12*24+(close_0-sum(min(low_0,delay(close_0,1)),12))/sum(max(high_0,delay(close_0,1))-min(low_0,delay(close_0,1)),12)*6*24+(close_0-sum(min(low_0,delay(close_0,1)),24))/sum(max(high_0,delay(close_0,1))-min(low_0,delay(close_0,1)),24)*6*24)*100/(6*12+6*24+12*24)
#----   全选小市值的股票？"""
)

m4 = M.input_features.v1(
    features_ds=m3.data,
    features='cond28=sum(price_limit_status_0==3,30)>1'
)

m15 = M.general_feature_extractor.v7(
    instruments=m1.data,
    features=m4.data,
    start_date='',
    end_date='',
    before_start_days=120
)

m16 = M.derived_feature_extractor.v3(
    input_data=m15.data,
    features=m4.data,
    date_col='date',
    instrument_col='instrument',
    drop_na=True,
    remove_extra_columns=False
)

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

m5 = M.filtet_st_stock.v7(
    input_1=m7.data
)

m10 = M.filter.v3(
    input_data=m5.data_1,
    expr='cond28',
    output_left_data=True
)

m13 = M.dropnan.v1(
    input_data=m10.data
)

m9 = M.instruments.v2(
    start_date=T.live_run_param('trading_date', '2021-01-15'),
    end_date=T.live_run_param('trading_date', '2022-10-20'),
    market='CN_STOCK_A',
    instrument_list='',
    max_count=0
)

m17 = M.general_feature_extractor.v7(
    instruments=m9.data,
    features=m4.data,
    start_date='',
    end_date='',
    before_start_days=120
)

m18 = M.derived_feature_extractor.v3(
    input_data=m17.data,
    features=m4.data,
    date_col='date',
    instrument_col='instrument',
    drop_na=False,
    remove_extra_columns=False
)

m11 = M.filtet_st_stock.v7(
    input_1=m18.data
)

m12 = M.filter.v3(
    input_data=m11.data_1,
    expr='cond28',
    output_left_data=True
)

m14 = M.dropnan.v1(
    input_data=m12.data
)

m24 = M.input_features.v1(
    features="""# #号开始的表示注释
# 多个特征，每行一个，可以包含基础特征和衍生特征
bm_0=where(ta_macd_dif(close,2,4,4)-ta_macd_dea(close,2,4,4)<0,1,0)
#bm_0=where((ta_macd_dif(close,2,4,4)-ta_macd_dea(close,2,4,4))/shift(ta_macd_dif(close,2,4,4)-ta_macd_dea(close,2,4,4),1)<0,1,0)"""
)

m23 = M.index_feature_extract.v3(
    input_1=m9.data,
    input_2=m24.data,
    before_days=100,
    index='000044.HIX'
)

m27 = M.input_features.v1(
    features="""std(turn_0,10)
avg_turn_5
rank_swing_volatility_60_0
fe1
mean(turn_0*(return_0-1),30)
std(return_0-1,60)
std(return_0-1,15)
fe2"""
)

m6 = M.stock_ranker_train.v5(
    training_ds=m13.data,
    features=m27.data,
    learning_algorithm='排序',
    number_of_leaves=40,
    minimum_docs_per_leaf=250,
    number_of_trees=80,
    learning_rate=0.1,
    max_bins=1023,
    feature_fraction=1,
    m_lazy_run=False
)

m8 = M.stock_ranker_predict.v5(
    model=m6.model,
    data=m14.data,
    m_lazy_run=False
)

m22 = M.join.v3(
    data1=m8.predictions,
    data2=m23.data_1,
    on='date',
    how='left',
    sort=False
)

m21 = M.sort.v4(
    input_ds=m22.data,
    sort_by='date',
    group_by='--',
    keep_columns='--',
    ascending=True
)

m20 = M.trade.v4(
    instruments=m9.data,
    options_data=m21.sorted_data,
    start_date='',
    end_date='',
    initialize=m20_initialize_bigquant_run,
    handle_data=m20_handle_data_bigquant_run,
    prepare=m20_prepare_bigquant_run,
    before_trading_start=m20_before_trading_start_bigquant_run,
    volume_limit=0,
    order_price_field_buy='open',
    order_price_field_sell='close',
    capital_base=100000,
    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='000300.SHA'
)

[2022-10-28 11:41:51.196569] INFO: moduleinvoker: instruments.v2 开始运行..

[2022-10-28 11:41:51.205182] INFO: moduleinvoker: 命中缓存

[2022-10-28 11:41:51.209657] INFO: moduleinvoker: instruments.v2 运行完成[0.013126s].

[2022-10-28 11:41:51.225228] INFO: moduleinvoker: advanced_auto_labeler.v2 开始运行..

[2022-10-28 11:41:51.234859] INFO: moduleinvoker: 命中缓存

[2022-10-28 11:41:51.238252] INFO: moduleinvoker: advanced_auto_labeler.v2 运行完成[0.013011s].

[2022-10-28 11:41:51.244500] INFO: moduleinvoker: input_features.v1 开始运行..

[2022-10-28 11:41:51.252126] INFO: moduleinvoker: 命中缓存

[2022-10-28 11:41:51.254434] INFO: moduleinvoker: input_features.v1 运行完成[0.009939s].

[2022-10-28 11:41:51.261020] INFO: moduleinvoker: input_features.v1 开始运行..

[2022-10-28 11:41:51.269128] INFO: moduleinvoker: 命中缓存

[2022-10-28 11:41:51.272423] INFO: moduleinvoker: input_features.v1 运行完成[0.011409s].

[2022-10-28 11:41:51.317436] INFO: moduleinvoker: general_feature_extractor.v7 开始运行..

[2022-10-28 11:41:51.325771] INFO: moduleinvoker: 命中缓存

[2022-10-28 11:41:51.328121] INFO: moduleinvoker: general_feature_extractor.v7 运行完成[0.010718s].

[2022-10-28 11:41:51.345936] INFO: moduleinvoker: derived_feature_extractor.v3 开始运行..

[2022-10-28 11:41:51.354194] INFO: moduleinvoker: 命中缓存

[2022-10-28 11:41:51.357838] INFO: moduleinvoker: derived_feature_extractor.v3 运行完成[0.011915s].

[2022-10-28 11:41:51.372989] INFO: moduleinvoker: join.v3 开始运行..

[2022-10-28 11:41:51.381102] INFO: moduleinvoker: 命中缓存

[2022-10-28 11:41:51.384832] INFO: moduleinvoker: join.v3 运行完成[0.011856s].

[2022-10-28 11:41:51.408398] INFO: moduleinvoker: filtet_st_stock.v7 开始运行..

[2022-10-28 11:41:51.417743] INFO: moduleinvoker: 命中缓存

[2022-10-28 11:41:51.421012] INFO: moduleinvoker: filtet_st_stock.v7 运行完成[0.012616s].

[2022-10-28 11:41:51.437974] INFO: moduleinvoker: filter.v3 开始运行..

[2022-10-28 11:41:51.449388] INFO: moduleinvoker: 命中缓存

[2022-10-28 11:41:51.452444] INFO: moduleinvoker: filter.v3 运行完成[0.014483s].

[2022-10-28 11:41:51.462807] INFO: moduleinvoker: dropnan.v1 开始运行..

[2022-10-28 11:41:51.470179] INFO: moduleinvoker: 命中缓存

[2022-10-28 11:41:51.473157] INFO: moduleinvoker: dropnan.v1 运行完成[0.010338s].

[2022-10-28 11:41:51.483771] INFO: moduleinvoker: instruments.v2 开始运行..

[2022-10-28 11:41:51.491875] INFO: moduleinvoker: 命中缓存

[2022-10-28 11:41:51.494829] INFO: moduleinvoker: instruments.v2 运行完成[0.011065s].

[2022-10-28 11:41:51.515043] INFO: moduleinvoker: general_feature_extractor.v7 开始运行..

[2022-10-28 11:41:51.524728] INFO: moduleinvoker: 命中缓存

[2022-10-28 11:41:51.527728] INFO: moduleinvoker: general_feature_extractor.v7 运行完成[0.012692s].

[2022-10-28 11:41:51.539871] INFO: moduleinvoker: derived_feature_extractor.v3 开始运行..

[2022-10-28 11:41:58.893337] INFO: derived_feature_extractor: 提取完成 cond28=sum(price_limit_status_0==3,30)>1, 2.072s

[2022-10-28 11:42:00.997223] INFO: derived_feature_extractor: 提取完成 std(turn_0,10), 2.101s

[2022-10-28 11:42:06.678025] INFO: derived_feature_extractor: 提取完成 fe1=sum(max(0,high_0-delay((high_0+low_0+close_0)/3,1)),26)/sum(max(0,delay((high_0+low_0+close_0)/3,1)-1),26)*100, 5.678s

[2022-10-28 11:42:08.829250] INFO: derived_feature_extractor: 提取完成 mean(turn_0*(return_0-1),30), 2.148s

[2022-10-28 11:42:10.889593] INFO: derived_feature_extractor: 提取完成 std(return_0-1,60), 2.058s

[2022-10-28 11:42:13.162929] INFO: derived_feature_extractor: 提取完成 std(return_0-1,15), 2.271s

[2022-10-28 11:42:29.497395] INFO: derived_feature_extractor: 提取完成 fe2=((close_0-sum(min(low_0,delay(close_0,1)),6))/sum(max(high_0,delay(close_0,1))-min(low_0,delay(close_0,1)),6)*12*24+(close_0-sum(min(low_0,delay(close_0,1)),12))/sum(max(high_0,delay(close_0,1))-min(low_0,delay(close_0,1)),12)*6*24+(close_0-sum(min(low_0,delay(close_0,1)),24))/sum(max(high_0,delay(close_0,1))-min(low_0,delay(close_0,1)),24)*6*24)*100/(6*12+6*24+12*24), 16.333s

[2022-10-28 11:42:30.724762] INFO: derived_feature_extractor: /y_2020, 283706

[2022-10-28 11:42:33.075964] INFO: derived_feature_extractor: /y_2021, 1061527

[2022-10-28 11:42:35.765468] INFO: derived_feature_extractor: /y_2022, 916020

[2022-10-28 11:42:36.708803] INFO: moduleinvoker: derived_feature_extractor.v3 运行完成[45.168858s].

[2022-10-28 11:42:36.731180] INFO: moduleinvoker: filtet_st_stock.v7 开始运行..

[2022-10-28 11:42:50.352635] INFO: moduleinvoker: filtet_st_stock.v7 运行完成[13.621442s].

[2022-10-28 11:42:50.370301] INFO: moduleinvoker: filter.v3 开始运行..

[2022-10-28 11:42:50.425321] INFO: filter: 使用表达式 cond28 过滤

[2022-10-28 11:42:57.469186] INFO: filter: 过滤 /data, 186439/1985268/2171707

[2022-10-28 11:42:57.544294] INFO: moduleinvoker: filter.v3 运行完成[7.173993s].

[2022-10-28 11:42:57.567340] INFO: moduleinvoker: dropnan.v1 开始运行..

[2022-10-28 11:42:58.424633] INFO: dropnan: /data, 177965/186439

[2022-10-28 11:42:58.481578] INFO: dropnan: 行数: 177965/186439

[2022-10-28 11:42:58.491179] INFO: moduleinvoker: dropnan.v1 运行完成[0.923815s].

[2022-10-28 11:42:58.498347] INFO: moduleinvoker: input_features.v1 开始运行..

[2022-10-28 11:42:58.508417] INFO: moduleinvoker: 命中缓存

[2022-10-28 11:42:58.511607] INFO: moduleinvoker: input_features.v1 运行完成[0.01331s].

[2022-10-28 11:42:58.573262] INFO: moduleinvoker: index_feature_extract.v3 开始运行..

[2022-10-28 11:42:58.584785] INFO: moduleinvoker: 命中缓存

[2022-10-28 11:42:58.588651] INFO: moduleinvoker: index_feature_extract.v3 运行完成[0.015393s].

[2022-10-28 11:42:58.598784] INFO: moduleinvoker: input_features.v1 开始运行..

[2022-10-28 11:42:58.612918] INFO: moduleinvoker: 命中缓存

[2022-10-28 11:42:58.616712] INFO: moduleinvoker: input_features.v1 运行完成[0.017927s].

[2022-10-28 11:42:58.634458] INFO: moduleinvoker: stock_ranker_train.v5 开始运行..

[2022-10-28 11:42:58.646505] INFO: moduleinvoker: 命中缓存

[2022-10-28 11:42:58.824751] INFO: moduleinvoker: stock_ranker_train.v5 运行完成[0.190297s].

[2022-10-28 11:42:58.846768] INFO: moduleinvoker: stock_ranker_predict.v5 开始运行..

[2022-10-28 11:42:59.874071] INFO: StockRanker预测: /data ..

[2022-10-28 11:43:00.720204] INFO: moduleinvoker: stock_ranker_predict.v5 运行完成[1.873431s].

[2022-10-28 11:43:00.746295] INFO: moduleinvoker: join.v3 开始运行..

[2022-10-28 11:43:01.425758] INFO: join: /data, 行数=177965/177965, 耗时=0.533761s

[2022-10-28 11:43:01.476488] INFO: join: 最终行数: 177965

[2022-10-28 11:43:01.488418] INFO: moduleinvoker: join.v3 运行完成[0.742119s].

[2022-10-28 11:43:01.510015] INFO: moduleinvoker: sort.v4 开始运行..

[2022-10-28 11:43:03.355374] INFO: moduleinvoker: sort.v4 运行完成[1.845342s].

[2022-10-28 11:43:06.227821] INFO: moduleinvoker: backtest.v8 开始运行..

[2022-10-28 11:43:06.235826] INFO: backtest: biglearning backtest:V8.6.3

[2022-10-28 11:43:10.524591] INFO: backtest: product_type:stock by specified

[2022-10-28 11:43:10.669420] INFO: moduleinvoker: cached.v2 开始运行..

[2022-10-28 11:43:10.678210] INFO: moduleinvoker: 命中缓存

[2022-10-28 11:43:10.681006] INFO: moduleinvoker: cached.v2 运行完成[0.011614s].

设置测试数据集，查看训练迭代过程的NDCG

bigcharts-data-start/{"__type":"tabs","__id":"bigchart-aa0a6f1dc20242b3a9e0cb4bbd689843"}/bigcharts-data-end