复制链接

克隆策略

In [1]:

# 本代码由可视化策略环境自动生成 2022年2月6日 10:41
# 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 m1_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=0.00016, sell_cost=0.00116, 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 = 0.8
    context.options['hold_days'] = 1
# 回测引擎：每日数据处理函数，每天执行一次
def m1_handle_data_bigquant_run(context, data):
    

    
    #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    
    benchmark_risk=context.benchmark_risk[today]

    #当risk为1时，市场有风险，全部平仓，不再执行其它操作
    if benchmark_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
    
    # 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')]

    # 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:
            price = data.current(context.symbol(instrument), 'price')  # 最新价格
            stock_num = np.floor(cash/price/100)*100  # 向下取整
            context.order(context.symbol(instrument), stock_num) # 整百下单
# 回测引擎：准备数据，只执行一次
def m1_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'])

    #这里以上证指数000001.HIX为例
    benchmark_data=df[df.instrument=='399303.ZIX']
    #计算上证指数5日涨幅
    benchmark_data['ret5']=benchmark_data['close']/benchmark_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
    benchmark_data['risk'] = np.where(benchmark_data['ret5']<-0.04,1,0)
    #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)
    benchmark_data['date']=benchmark_data['date'].apply(lambda x:x.strftime('%Y-%m-%d'))
    #设置日期为索引
    benchmark_data.set_index('date',inplace=True)
    #把风控序列输出给全局变量context.benchmark_risk
    context.benchmark_risk=benchmark_data['risk']

m8 = M.input_features.v1(
    features="""return_0
turn_0"""
)

m34 = M.input_features.v1(
    features_ds=m8.data,
    features=''
)

m24 = M.instruments.v2(
    start_date='2020-12-24',
    end_date='2021-05-31',
    market='CN_STOCK_A',
    instrument_list='',
    max_count=0
)

m21 = M.advanced_auto_labeler.v2(
    instruments=m24.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：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, 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=False,
    user_functions={}
)

m22 = M.general_feature_extractor.v7(
    instruments=m24.data,
    features=m34.data,
    start_date='',
    end_date='',
    before_start_days=365
)

m23 = M.derived_feature_extractor.v3(
    input_data=m22.data,
    features=m34.data,
    date_col='date',
    instrument_col='instrument',
    drop_na=True,
    remove_extra_columns=False,
    user_functions={}
)

m2 = M.chinaa_stock_filter.v1(
    input_data=m23.data,
    index_constituent_cond=['全部'],
    board_cond=['上证主板', '深证主板', '创业板'],
    industry_cond=['全部'],
    st_cond=['正常'],
    delist_cond=['非退市'],
    output_left_data=False
)

m17 = M.join.v3(
    data1=m21.data,
    data2=m2.data,
    on='date,instrument',
    how='inner',
    sort=True
)

m3 = M.dropnan.v2(
    input_data=m17.data
)

m28 = M.instruments.v2(
    start_date='2021-06-01',
    end_date='2021-12-31',
    market='CN_STOCK_A',
    instrument_list='',
    max_count=0
)

m16 = M.general_feature_extractor.v7(
    instruments=m28.data,
    features=m34.data,
    start_date='',
    end_date='',
    before_start_days=365
)

m26 = M.derived_feature_extractor.v3(
    input_data=m16.data,
    features=m34.data,
    date_col='date',
    instrument_col='instrument',
    drop_na=True,
    remove_extra_columns=False,
    user_functions={}
)

m25 = M.chinaa_stock_filter.v1(
    input_data=m26.data,
    index_constituent_cond=['全部'],
    board_cond=['上证主板', '深证主板', '创业板'],
    industry_cond=['全部'],
    st_cond=['正常'],
    delist_cond=['非退市'],
    output_left_data=False
)

m4 = M.dropnan.v2(
    input_data=m25.data,
    features=m8.data
)

m6 = M.stock_ranker.v2(
    training_ds=m3.data,
    features=m8.data,
    test_ds=m3.data,
    predict_ds=m4.data,
    learning_algorithm='排序',
    number_of_leaves=31,
    minimum_docs_per_leaf=631,
    number_of_trees=15,
    learning_rate=0.1,
    max_bins=1023,
    feature_fraction=1,
    data_row_fraction=1,
    ndcg_discount_base=1,
    slim_data=True
)

m1 = M.trade.v4(
    instruments=m28.data,
    options_data=m6.predictions,
    start_date='',
    end_date='',
    initialize=m1_initialize_bigquant_run,
    handle_data=m1_handle_data_bigquant_run,
    prepare=m1_prepare_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'
)

[2022-02-06 10:19:38.891192] INFO: moduleinvoker: input_features.v1 开始运行..

[2022-02-06 10:19:38.934119] INFO: moduleinvoker: input_features.v1 运行完成[0.042975s].

[2022-02-06 10:19:38.938421] INFO: moduleinvoker: input_features.v1 开始运行..

[2022-02-06 10:19:38.965661] INFO: moduleinvoker: input_features.v1 运行完成[0.027224s].

[2022-02-06 10:19:38.974159] INFO: moduleinvoker: instruments.v2 开始运行..

[2022-02-06 10:19:39.034980] INFO: moduleinvoker: instruments.v2 运行完成[0.060806s].

[2022-02-06 10:19:39.050647] INFO: moduleinvoker: advanced_auto_labeler.v2 开始运行..

[2022-02-06 10:19:39.815219] INFO: 自动标注(股票): 加载历史数据: 432414 行

[2022-02-06 10:19:39.816851] INFO: 自动标注(股票): 开始标注 ..

[2022-02-06 10:19:40.402490] INFO: moduleinvoker: advanced_auto_labeler.v2 运行完成[1.351839s].

[2022-02-06 10:19:40.434483] INFO: moduleinvoker: general_feature_extractor.v7 开始运行..

[2022-02-06 10:19:41.766909] INFO: 基础特征抽取: 年份 2019, 特征行数=18640

[2022-02-06 10:19:43.512931] INFO: 基础特征抽取: 年份 2020, 特征行数=944770

[2022-02-06 10:19:45.217681] INFO: 基础特征抽取: 年份 2021, 特征行数=407757

[2022-02-06 10:19:45.269043] INFO: 基础特征抽取: 总行数: 1371167

[2022-02-06 10:19:45.275974] INFO: moduleinvoker: general_feature_extractor.v7 运行完成[4.841511s].

[2022-02-06 10:19:45.287622] INFO: moduleinvoker: derived_feature_extractor.v3 开始运行..

[2022-02-06 10:19:47.271699] INFO: derived_feature_extractor: /y_2019, 18640

[2022-02-06 10:19:48.547913] INFO: derived_feature_extractor: /y_2020, 944770

[2022-02-06 10:19:49.385335] INFO: derived_feature_extractor: /y_2021, 407757

[2022-02-06 10:19:49.542430] INFO: moduleinvoker: derived_feature_extractor.v3 运行完成[4.25479s].

[2022-02-06 10:19:49.555102] INFO: moduleinvoker: chinaa_stock_filter.v1 开始运行..

[2022-02-06 10:19:49.848210] INFO: A股股票过滤: 过滤 /y_2019, 17654/0/18633

[2022-02-06 10:19:52.265540] INFO: A股股票过滤: 过滤 /y_2020, 869525/0/944374

[2022-02-06 10:19:53.413418] INFO: A股股票过滤: 过滤 /y_2021, 364571/0/407561

[2022-02-06 10:19:53.417318] INFO: A股股票过滤: 过滤完成, 1251750 + 0

[2022-02-06 10:19:53.446877] INFO: moduleinvoker: chinaa_stock_filter.v1 运行完成[3.891761s].

[2022-02-06 10:19:53.466636] INFO: moduleinvoker: join.v3 开始运行..

[2022-02-06 10:19:54.329933] INFO: join: /y_2019, 行数=0/17654, 耗时=0.20568s

[2022-02-06 10:19:56.043600] INFO: join: /y_2020, 行数=22114/869525, 耗时=1.712032s

[2022-02-06 10:19:56.902789] INFO: join: /y_2021, 行数=356168/364571, 耗时=0.855496s

[2022-02-06 10:19:56.953318] INFO: join: 最终行数: 378282

[2022-02-06 10:19:56.961600] INFO: moduleinvoker: join.v3 运行完成[3.494974s].

[2022-02-06 10:19:56.972723] INFO: moduleinvoker: dropnan.v2 开始运行..

[2022-02-06 10:19:57.040239] INFO: dropnan: /y_2019, 0/0

[2022-02-06 10:19:57.096143] INFO: dropnan: /y_2020, 22114/22114

[2022-02-06 10:19:57.326351] INFO: dropnan: /y_2021, 356168/356168

[2022-02-06 10:19:57.374522] INFO: dropnan: 行数: 378282/378282

[2022-02-06 10:19:57.384478] INFO: moduleinvoker: dropnan.v2 运行完成[0.411745s].

[2022-02-06 10:19:57.389338] INFO: moduleinvoker: instruments.v2 开始运行..

[2022-02-06 10:19:57.399892] INFO: moduleinvoker: 命中缓存

[2022-02-06 10:19:57.401195] INFO: moduleinvoker: instruments.v2 运行完成[0.011857s].

[2022-02-06 10:19:57.414344] INFO: moduleinvoker: general_feature_extractor.v7 开始运行..

[2022-02-06 10:19:59.029294] INFO: 基础特征抽取: 年份 2020, 特征行数=577725

[2022-02-06 10:20:01.527395] INFO: 基础特征抽取: 年份 2021, 特征行数=1061217

[2022-02-06 10:20:01.610609] INFO: 基础特征抽取: 总行数: 1638942

[2022-02-06 10:20:01.617563] INFO: moduleinvoker: general_feature_extractor.v7 运行完成[4.203228s].

[2022-02-06 10:20:01.623985] INFO: moduleinvoker: derived_feature_extractor.v3 开始运行..

[2022-02-06 10:20:04.628760] INFO: derived_feature_extractor: /y_2020, 577725

[2022-02-06 10:20:06.154615] INFO: derived_feature_extractor: /y_2021, 1061217

[2022-02-06 10:20:06.430015] INFO: moduleinvoker: derived_feature_extractor.v3 运行完成[4.80601s].

[2022-02-06 10:20:06.437854] INFO: moduleinvoker: chinaa_stock_filter.v1 开始运行..

[2022-02-06 10:20:07.938312] INFO: A股股票过滤: 过滤 /y_2020, 525566/0/577422

[2022-02-06 10:20:10.725855] INFO: A股股票过滤: 过滤 /y_2021, 940121/0/1060655

[2022-02-06 10:20:10.728873] INFO: A股股票过滤: 过滤完成, 1465687 + 0

[2022-02-06 10:20:10.751592] INFO: moduleinvoker: chinaa_stock_filter.v1 运行完成[4.313725s].

[2022-02-06 10:20:10.759471] INFO: moduleinvoker: dropnan.v2 开始运行..

[2022-02-06 10:20:10.943417] INFO: dropnan: /y_2020, 525566/525566

[2022-02-06 10:20:11.185368] INFO: dropnan: /y_2021, 940121/940121

[2022-02-06 10:20:11.235025] INFO: dropnan: 行数: 1465687/1465687

[2022-02-06 10:20:11.244984] INFO: moduleinvoker: dropnan.v2 运行完成[0.485508s].

[2022-02-06 10:20:11.263927] INFO: moduleinvoker: stock_ranker.v2 开始运行..

[2022-02-06 10:20:11.284557] INFO: moduleinvoker: stock_ranker_train.v6 开始运行..

[2022-02-06 10:20:11.826991] INFO: StockRanker: 特征预处理 ..

[2022-02-06 10:20:11.935416] INFO: StockRanker: prepare data: training ..

[2022-02-06 10:20:15.217127] INFO: StockRanker: prepare data: test ..

[2022-02-06 10:20:18.347986] INFO: StockRanker训练: 4fa2d6c4 准备训练: 378282 行数, test: 378282 rows

[2022-02-06 10:20:18.349444] INFO: StockRanker训练: AI模型训练，将在378282*2=75.66万数据上对模型训练进行15轮迭代训练。预计将需要1～2分钟。请耐心等待。

[2022-02-06 10:20:18.535875] INFO: StockRanker训练: 正在训练 ..

[2022-02-06 10:20:18.579845] INFO: StockRanker训练: 任务状态: Pending

[2022-02-06 10:20:28.619088] INFO: StockRanker训练: 任务状态: Running

[2022-02-06 10:21:38.917601] INFO: StockRanker训练: 00:01:10.8388045, finished iteration 1

[2022-02-06 10:21:48.959453] INFO: StockRanker训练: 00:01:20.5444563, finished iteration 2

[2022-02-06 10:21:59.008699] INFO: StockRanker训练: 00:01:30.7522815, finished iteration 3

[2022-02-06 10:22:09.048129] INFO: StockRanker训练: 00:01:41.0514876, finished iteration 4

[2022-02-06 10:22:19.098253] INFO: StockRanker训练: 00:01:51.0967391, finished iteration 5

[2022-02-06 10:22:29.137158] INFO: StockRanker训练: 00:02:01.4733178, finished iteration 6

[2022-02-06 10:22:39.170010] INFO: StockRanker训练: 00:02:11.9931378, finished iteration 7

[2022-02-06 10:22:49.209565] INFO: StockRanker训练: 00:02:22.9033022, finished iteration 8

[2022-02-06 10:22:59.255300] INFO: StockRanker训练: 00:02:35.0600449, finished iteration 9

[2022-02-06 10:23:19.335874] INFO: StockRanker训练: 00:02:47.9507876, finished iteration 10

[2022-02-06 10:23:29.377741] INFO: StockRanker训练: 00:03:00.2821014, finished iteration 11

[2022-02-06 10:23:39.419141] INFO: StockRanker训练: 00:03:13.4332842, finished iteration 12

[2022-02-06 10:23:49.457842] INFO: StockRanker训练: 00:03:26.3190288, finished iteration 13

[2022-02-06 10:24:09.542298] INFO: StockRanker训练: 00:03:38.6110875, finished iteration 14

[2022-02-06 10:24:19.581351] INFO: StockRanker训练: 00:03:51.0928029, finished iteration 15

[2022-02-06 10:24:19.582805] INFO: StockRanker训练: 任务状态: Succeeded

[2022-02-06 10:24:19.789327] INFO: moduleinvoker: stock_ranker_train.v6 运行完成[248.504768s].

[2022-02-06 10:24:19.802210] INFO: moduleinvoker: stock_ranker_predict.v5 开始运行..

[2022-02-06 10:24:20.069935] INFO: StockRanker预测: /y_2020 ..

[2022-02-06 10:24:20.517943] INFO: StockRanker预测: /y_2021 ..

[2022-02-06 10:24:23.100646] INFO: moduleinvoker: stock_ranker_predict.v5 运行完成[3.298411s].

[2022-02-06 10:24:23.178701] INFO: moduleinvoker: stock_ranker.v2 运行完成[251.914768s].

[2022-02-06 10:24:24.856363] INFO: moduleinvoker: backtest.v8 开始运行..

[2022-02-06 10:24:24.862049] INFO: backtest: biglearning backtest:V8.6.1

[2022-02-06 10:24:24.904777] INFO: backtest: product_type:stock by specified

[2022-02-06 10:24:24.966548] INFO: moduleinvoker: cached.v2 开始运行..

[2022-02-06 10:24:28.711542] INFO: backtest: 读取股票行情完成:1760855

[2022-02-06 10:24:31.282952] INFO: moduleinvoker: cached.v2 运行完成[6.316405s].

[2022-02-06 10:24:32.765249] INFO: algo: TradingAlgorithm V1.8.6

[2022-02-06 10:24:33.387879] INFO: algo: trading transform...

[2022-02-06 10:24:33.825754] ERROR: moduleinvoker: module name: backtest, module version: v8, trackeback: KeyError: 1622505600000000000

During handling of the above exception, another exception occurred:

KeyError: Timestamp('2021-06-01 00:00:00')

The above exception was the direct cause of the following exception:

KeyError: Timestamp('2021-06-01 00:00:00')

The above exception was the direct cause of the following exception:

KeyError: '2021-06-01'

[2022-02-06 10:24:33.844062] ERROR: moduleinvoker: module name: trade, module version: v4, trackeback: KeyError: 1622505600000000000

During handling of the above exception, another exception occurred:

KeyError: Timestamp('2021-06-01 00:00:00')

The above exception was the direct cause of the following exception:

KeyError: Timestamp('2021-06-01 00:00:00')

The above exception was the direct cause of the following exception:

KeyError: '2021-06-01'

bigcharts-data-start/{"__type":"tabs","__id":"bigchart-c6ca77a6809342869a2ce768b8359832"}/bigcharts-data-end

bigcharts-data-start/{"__type":"tabs","__id":"bigchart-4005cc9f0de04bbba21a01e7e8a6a769"}/bigcharts-data-end

---------------------------------------------------------------------------
KeyError                                  Traceback (most recent call last)
KeyError: 1622505600000000000

During handling of the above exception, another exception occurred:

KeyError                                  Traceback (most recent call last)
KeyError: Timestamp('2021-06-01 00:00:00')

The above exception was the direct cause of the following exception:

KeyError                                  Traceback (most recent call last)
KeyError: Timestamp('2021-06-01 00:00:00')

The above exception was the direct cause of the following exception:

KeyError                                  Traceback (most recent call last)
<ipython-input-1-d27544f93c17> in <module>
    244 )
    245 
--> 246 m1 = M.trade.v4(
    247     instruments=m28.data,
    248     options_data=m6.predictions,

<ipython-input-1-d27544f93c17> in m1_handle_data_bigquant_run(context, data)
     27     stock_hold_now = [equity.symbol for equity in context.portfolio.positions ]
     28     #大盘风控模块，读取风控数据
---> 29     benchmark_risk=context.benchmark_risk[today]
     30 
     31     #当risk为1时，市场有风险，全部平仓，不再执行其它操作

KeyError: '2021-06-01'