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3月26日 08:46
# 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。


#import random
# 回测引擎：每日数据处理函数，每天执行一次
def m12_handle_data_bigquant_run(context, data):
    # 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
    #random.shuffle(buy_cash_weights)
    #buy_cash_weights=buy_cash_weights[:3]
    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)

# 回测引擎：准备数据，只执行一次
def m12_prepare_bigquant_run(context):
    pass

# 回测引擎：初始化函数，只执行一次
def m12_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 = 3
    # 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 = 0.33
    context.options['hold_days'] = 20


m1 = M.instruments.v2(
    start_date='2005-01-01',
    end_date='2017-12-31',
    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 {{web_host_url}}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 <{{web_host_url}}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)
#rank(where(shift(close, -5) / shift(open, -1)>1, rank(shift(close, -5) / shift(open, -1)), 0))
where(shift(close, -20) / shift(open, -1)>=1, 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, 2)

# 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.HIX',
    drop_na_label=True,
    cast_label_int=True
)

m3 = M.input_features.v1(
    features="""open_0
close_0
high_0
low_0
turn_0
return_0"""
)

m15 = M.input_features.v1(
    features_ds=m3.data,
    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
#ST状态：0：正常股票，1：ST，2：*ST，11：暂停上市
st_status_0
#上市板，主板：1，中小企业板：2，创业板：3
list_board_0
#已上市的天数
list_days_0"""
)

m4 = M.general_feature_extractor.v6(
    instruments=m1.data,
    features=m15.data,
    start_date='',
    end_date='',
    before_start_days=50
)

m5 = M.derived_feature_extractor.v2(
    input_data=m4.data,
    features=m15.data,
    date_col='date',
    instrument_col='instrument'
)

m7 = M.join.v3(
    data1=m2.data,
    data2=m5.data,
    on='date,instrument',
    how='inner',
    sort=False
)

m16 = M.filter.v3(
    input_data=m7.data,
    expr='list_board_0==1 and list_days_0>=365',
    output_left_data=False
)

m18 = M.filter_delist_stock.v6(
    input_1=m16.data
)

m20 = M.filtet_st_stock_tomo.v3(
    input_1=m18.data_1
)

m19 = M.filtet_st_stock.v7(
    input_1=m20.data_1
)

m21 = M.filter_stockmarket.v2(
    input_1=m19.data_1,
    start='3'
)

m13 = M.dropnan.v1(
    input_data=m21.data_1
)

m22 = M.standardlize.v9(
    input_1=m13.data,
    input_2=m3.data,
    standard_func='RobustZScoreNorm',
    columns_input=''
)

m28 = M.fillnan.v1(
    input_data=m22.data,
    features=m3.data,
    fill_value='0.0'
)

m30 = M.gplearn_train.v3(
    input_1=m28.data,
    input_2=m3.data,
    IS_USE_BEST_GP=True,
    gnrtn=3,
    ppltn=80,
    max_samples=0.9,
    metric='mse',
    stopping_criteria=0.01,
    init_depth_min=2,
    init_depth_max=6,
    p_crossover=0.9,
    p_hoist_mutation=0.01,
    p_point_mutation=0.05,
    p_point_replace=0.01,
    p_subtree_mutation=0.01
)

m9 = M.instruments.v2(
    start_date=T.live_run_param('trading_date', '2018-01-01'),
    end_date=T.live_run_param('trading_date', '2022-03-24'),
    market='CN_STOCK_A',
    instrument_list='',
    max_count=0
)

m10 = M.general_feature_extractor.v6(
    instruments=m9.data,
    features=m15.data,
    start_date='',
    end_date='',
    before_start_days=0
)

m11 = M.derived_feature_extractor.v2(
    input_data=m10.data,
    features=m15.data,
    date_col='date',
    instrument_col='instrument'
)

m17 = M.filter.v3(
    input_data=m11.data,
    expr='list_board_0==1 and list_days_0>=365 ',
    output_left_data=False
)

m24 = M.filter_delist_stock.v6(
    input_1=m17.data
)

m25 = M.filtet_st_stock_tomo.v3(
    input_1=m24.data_1
)

m26 = M.filtet_st_stock.v7(
    input_1=m25.data_1
)

m23 = M.filter_stockmarket.v2(
    input_1=m26.data_1,
    start='3'
)

m14 = M.dropnan.v1(
    input_data=m23.data_1
)

m27 = M.standardlize.v9(
    input_1=m14.data,
    input_2=m3.data,
    standard_func='RobustZScoreNorm',
    columns_input=''
)

m29 = M.fillnan.v1(
    input_data=m27.data,
    features=m3.data,
    fill_value='0.0'
)

m31 = M.GP_predict.v2(
    input_1=m30.data_1,
    input_2=m3.data,
    input_3=m29.data
)

m12 = M.trade.v3(
    instruments=m9.data,
    options_data=m31.data_1,
    start_date='',
    end_date='',
    handle_data=m12_handle_data_bigquant_run,
    prepare=m12_prepare_bigquant_run,
    initialize=m12_initialize_bigquant_run,
    volume_limit=0,
    order_price_field_buy='open',
    order_price_field_sell='close',
    capital_base=150000,
    benchmark='000300.HIX',
    auto_cancel_non_tradable_orders=True,
    data_frequency='daily',
    price_type='真实价格',
    plot_charts=True,
    backtest_only=False
)

[2022-03-25 21:34:29.329943] INFO: moduleinvoker: instruments.v2 开始运行..

[2022-03-25 21:34:29.350817] INFO: moduleinvoker: 命中缓存

[2022-03-25 21:34:29.352563] INFO: moduleinvoker: instruments.v2 运行完成[0.022641s].

[2022-03-25 21:34:29.439088] INFO: moduleinvoker: advanced_auto_labeler.v2 开始运行..

[2022-03-25 21:34:29.451241] INFO: moduleinvoker: 命中缓存

[2022-03-25 21:34:29.455226] INFO: moduleinvoker: advanced_auto_labeler.v2 运行完成[0.016134s].

[2022-03-25 21:34:29.505378] INFO: moduleinvoker: input_features.v1 开始运行..

[2022-03-25 21:34:29.563233] INFO: moduleinvoker: input_features.v1 运行完成[0.057861s].

[2022-03-25 21:34:29.572831] INFO: moduleinvoker: input_features.v1 开始运行..

[2022-03-25 21:34:29.647272] INFO: moduleinvoker: input_features.v1 运行完成[0.074475s].

[2022-03-25 21:34:29.669968] INFO: moduleinvoker: general_feature_extractor.v6 开始运行..

[2022-03-25 21:34:29.993547] INFO: 基础特征抽取: 年份 2004, 特征行数=0

[2022-03-25 21:34:44.345391] INFO: 基础特征抽取: 年份 2005, 特征行数=314144

[2022-03-25 21:34:46.541478] INFO: 基础特征抽取: 年份 2006, 特征行数=287870

[2022-03-25 21:34:49.606205] INFO: 基础特征抽取: 年份 2007, 特征行数=323371

[2022-03-25 21:34:52.285143] INFO: 基础特征抽取: 年份 2008, 特征行数=360328

[2022-03-25 21:35:20.241646] INFO: 基础特征抽取: 年份 2009, 特征行数=375308

[2022-03-25 21:35:27.670625] INFO: 基础特征抽取: 年份 2010, 特征行数=431567

[2022-03-25 21:36:51.328681] INFO: 基础特征抽取: 年份 2011, 特征行数=511455

[2022-03-25 21:36:57.681171] INFO: 基础特征抽取: 年份 2012, 特征行数=565675

[2022-03-25 21:37:02.234134] INFO: 基础特征抽取: 年份 2013, 特征行数=564168

[2022-03-25 21:37:06.773632] INFO: 基础特征抽取: 年份 2014, 特征行数=569948

[2022-03-25 21:37:11.202534] INFO: 基础特征抽取: 年份 2015, 特征行数=569698

[2022-03-25 21:37:16.025521] INFO: 基础特征抽取: 年份 2016, 特征行数=641546

[2022-03-25 21:37:21.819398] INFO: 基础特征抽取: 年份 2017, 特征行数=743233

[2022-03-25 21:37:22.080086] INFO: 基础特征抽取: 总行数: 6258311

[2022-03-25 21:37:22.095108] INFO: moduleinvoker: general_feature_extractor.v6 运行完成[172.425139s].

[2022-03-25 21:37:22.123329] INFO: moduleinvoker: derived_feature_extractor.v2 开始运行..

[2022-03-25 21:37:36.266336] INFO: derived_feature_extractor: /y_2005, 314144

[2022-03-25 21:37:36.950021] INFO: derived_feature_extractor: /y_2006, 287870

[2022-03-25 21:37:37.788357] INFO: derived_feature_extractor: /y_2007, 323371

[2022-03-25 21:37:38.615871] INFO: derived_feature_extractor: /y_2008, 360328

[2022-03-25 21:37:39.388576] INFO: derived_feature_extractor: /y_2009, 375308

[2022-03-25 21:37:40.470001] INFO: derived_feature_extractor: /y_2010, 431567

[2022-03-25 21:37:41.743946] INFO: derived_feature_extractor: /y_2011, 511455

[2022-03-25 21:37:43.005023] INFO: derived_feature_extractor: /y_2012, 565675

[2022-03-25 21:37:44.426191] INFO: derived_feature_extractor: /y_2013, 564168

[2022-03-25 21:37:45.861538] INFO: derived_feature_extractor: /y_2014, 569948

[2022-03-25 21:37:47.135564] INFO: derived_feature_extractor: /y_2015, 569698

[2022-03-25 21:37:48.837313] INFO: derived_feature_extractor: /y_2016, 641546

[2022-03-25 21:37:50.355647] INFO: derived_feature_extractor: /y_2017, 743233

[2022-03-25 21:37:50.827533] INFO: moduleinvoker: derived_feature_extractor.v2 运行完成[28.704196s].

[2022-03-25 21:37:50.841708] INFO: moduleinvoker: join.v3 开始运行..

[2022-03-25 21:38:04.134992] INFO: join: /y_2005, 行数=313138/314144, 耗时=2.794607s

[2022-03-25 21:38:07.116262] INFO: join: /y_2006, 行数=286781/287870, 耗时=2.977438s

[2022-03-25 21:38:10.237886] INFO: join: /y_2007, 行数=320511/323371, 耗时=3.11846s

[2022-03-25 21:38:13.561743] INFO: join: /y_2008, 行数=358830/360328, 耗时=3.32051s

[2022-03-25 21:38:16.655592] INFO: join: /y_2009, 行数=374495/375308, 耗时=3.090479s

[2022-03-25 21:38:20.033040] INFO: join: /y_2010, 行数=431045/431567, 耗时=3.373738s

[2022-03-25 21:38:23.564370] INFO: join: /y_2011, 行数=510937/511455, 耗时=3.524471s

[2022-03-25 21:38:27.204716] INFO: join: /y_2012, 行数=564597/565675, 耗时=3.634664s

[2022-03-25 21:38:30.781573] INFO: join: /y_2013, 行数=563157/564168, 耗时=3.571521s

[2022-03-25 21:38:34.796497] INFO: join: /y_2014, 行数=567889/569948, 耗时=4.007076s

[2022-03-25 21:38:38.620611] INFO: join: /y_2015, 行数=560453/569698, 耗时=3.817226s

[2022-03-25 21:38:42.594019] INFO: join: /y_2016, 行数=637492/641546, 耗时=3.966364s

[2022-03-25 21:38:47.214895] INFO: join: /y_2017, 行数=738390/743233, 耗时=4.613767s

[2022-03-25 21:38:47.369656] INFO: join: 最终行数: 6227715

[2022-03-25 21:38:47.406972] INFO: moduleinvoker: join.v3 运行完成[56.565255s].

[2022-03-25 21:38:47.418873] INFO: moduleinvoker: filter.v3 开始运行..

[2022-03-25 21:38:47.437322] INFO: filter: 使用表达式 list_board_0==1 and list_days_0>=365 过滤

[2022-03-25 21:38:47.934827] INFO: filter: 过滤 /y_2005, 298015/0/313138

[2022-03-25 21:38:48.339371] INFO: filter: 过滤 /y_2006, 271216/0/286781

[2022-03-25 21:38:48.701013] INFO: filter: 过滤 /y_2007, 280507/0/320511

[2022-03-25 21:38:49.035546] INFO: filter: 过滤 /y_2008, 295521/0/358830

[2022-03-25 21:38:49.393087] INFO: filter: 过滤 /y_2009, 304293/0/374495

[2022-03-25 21:38:49.759092] INFO: filter: 过滤 /y_2010, 301195/0/431045

[2022-03-25 21:38:50.188640] INFO: filter: 过滤 /y_2011, 306294/0/510937

[2022-03-25 21:38:50.636097] INFO: filter: 过滤 /y_2012, 316366/0/564597

[2022-03-25 21:38:51.076752] INFO: filter: 过滤 /y_2013, 319032/0/563157

[2022-03-25 21:38:51.533027] INFO: filter: 过滤 /y_2014, 319017/0/567889

[2022-03-25 21:38:51.962655] INFO: filter: 过滤 /y_2015, 299946/0/560453

[2022-03-25 21:38:52.521063] INFO: filter: 过滤 /y_2016, 336462/0/637492

[2022-03-25 21:38:53.203954] INFO: filter: 过滤 /y_2017, 355961/0/738390

[2022-03-25 21:38:53.235947] INFO: moduleinvoker: filter.v3 运行完成[5.817057s].

[2022-03-25 21:38:53.361127] INFO: moduleinvoker: filter_delist_stock.v6 开始运行..

[2022-03-25 21:39:08.349855] INFO: moduleinvoker: filter_delist_stock.v6 运行完成[14.988726s].

[2022-03-25 21:39:08.406002] INFO: moduleinvoker: filtet_st_stock_tomo.v3 开始运行..

[2022-03-25 21:39:22.891566] INFO: moduleinvoker: filtet_st_stock_tomo.v3 运行完成[14.485555s].

[2022-03-25 21:39:22.936852] INFO: moduleinvoker: filtet_st_stock.v7 开始运行..

[2022-03-25 21:39:39.734107] INFO: moduleinvoker: filtet_st_stock.v7 运行完成[16.797261s].

[2022-03-25 21:39:39.795020] INFO: moduleinvoker: filter_stockmarket.v2 开始运行..

[2022-03-25 21:39:43.724163] INFO: moduleinvoker: filter_stockmarket.v2 运行完成[3.929181s].

[2022-03-25 21:39:43.739443] INFO: moduleinvoker: dropnan.v1 开始运行..

[2022-03-25 21:39:47.924079] INFO: dropnan: /data, 3587437/3588467

[2022-03-25 21:39:48.018597] INFO: dropnan: 行数: 3587437/3588467

[2022-03-25 21:39:48.049430] INFO: moduleinvoker: dropnan.v1 运行完成[4.309987s].

[2022-03-25 21:39:48.064013] INFO: moduleinvoker: standardlize.v9 开始运行..

[2022-03-25 21:41:01.331813] INFO: moduleinvoker: standardlize.v9 运行完成[73.26779s].

[2022-03-25 21:41:01.350337] INFO: moduleinvoker: fillnan.v1 开始运行..

[2022-03-25 21:41:04.847542] INFO: moduleinvoker: fillnan.v1 运行完成[3.497203s].

[2022-03-25 21:41:04.936473] INFO: moduleinvoker: gplearn_train.v3 开始运行..

[2022-03-25 21:43:07.184901] ERROR: moduleinvoker: module name: gplearn_train, module version: v3, trackeback: AttributeError: 'DataFrame' object has no attribute 'price'

参与训练模型的因子有：  ['open_0', 'close_0', 'high_0', 'low_0', 'turn_0', 'return_0']
    |   Population Average    |             Best Individual              |
---- ------------------------- ------------------------------------------ ----------
 Gen   Length          Fitness   Length          Fitness      OOB Fitness  Time Left
   0    14.49      5.69083e+22        2         0.518789         0.516984      1.66m
   1     4.04      6.01263e+14        3         0.326106         0.326368     30.92s
   2     6.54      5.98697e+14        4         0.299645         0.299022      0.00s
############################################# 末代最佳表达式：  abs(add(0.203, -0.963)) 筛选后最佳表达式： sqrt(sqrt(min(min(add(0.203, -0.963), log(X2)), abs(X3))))
############################################# 因子有效性: True
############################################# 使用全局最优表达式 True

---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
<ipython-input-1-74e9b346f46e> in <module>
    189 )
    190 
--> 191 m30 = M.gplearn_train.v3(
    192     input_1=m28.data,
    193     input_2=m3.data,

AttributeError: 'DataFrame' object has no attribute 'price'