复制链接

克隆策略

In [7]:

# 本代码由可视化策略环境自动生成 2022年8月9日 10:20
# 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。


# Python 代码入口函数，input_1/2/3 对应三个输入端，data_1/2/3 对应三个输出端
def m6_run_bigquant_run(input_1, input_2, input_3):
    df = input_1.read_df()
    
    # 数据格式统一：float
    col_names = list(df.columns)
    col_names.remove('instrument')
    col_names.remove('date')
    df.loc[:,col_names] = df.loc[:,col_names].astype(float)
    
    # 处理数据：inf_
    import numpy as np
    df.loc[:,col_names] = df.loc[:,col_names].where(~np.isinf(df.loc[:,col_names].values),other=np.nan)
    data_1 = DataSource.write_df(df)
    return Outputs(data_1=data_1, data_2=None, data_3=None)

# 后处理函数，可选。输入是主函数的输出，可以在这里对数据做处理，或者返回更友好的outputs数据格式。此函数输出不会被缓存。
def m6_post_run_bigquant_run(outputs):
    return outputs

# Python 代码入口函数，input_1/2/3 对应三个输入端，data_1/2/3 对应三个输出端
def m8_run_bigquant_run(input_1, input_2, input_3):
    df = input_1.read_df()
    
    # 数据格式统一：float
    col_names = list(df.columns)
    col_names.remove('instrument')
    col_names.remove('date')
    df.loc[:,col_names] = df.loc[:,col_names].astype(float)
    
    # 处理数据：inf_
    import numpy as np
    df.loc[:,col_names] = df.loc[:,col_names].where(~np.isinf(df.loc[:,col_names].values),other=np.nan)
    data_1 = DataSource.write_df(df)
    return Outputs(data_1=data_1, data_2=None, data_3=None)

# 后处理函数，可选。输入是主函数的输出，可以在这里对数据做处理，或者返回更友好的outputs数据格式。此函数输出不会被缓存。
def m8_post_run_bigquant_run(outputs):
    return outputs

# 回测引擎：初始化函数，只执行一次
def m11_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 = 5
    # 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 = 0.2
    context.options['hold_days'] = 5

# 回测引擎：每日数据处理函数，每天执行一次
def m11_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')]
    
    ranker_prediction = ranker_prediction.sort_values('pred_label',ascending=False)
    
    # 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)

# 回测引擎：准备数据，只执行一次
def m11_prepare_bigquant_run(context):
    pass

# 回测引擎：每个单位时间开始前调用一次，即每日开盘前调用一次。
def m11_before_trading_start_bigquant_run(context, data):
    pass


g = T.Graph({

    'm1': 'M.instruments.v2',
    'm1.start_date': '2016-01-01',
    'm1.end_date': '2019-01-01',
    'm1.market': 'CN_STOCK_A',
    'm1.instrument_list': '',
    'm1.max_count': 0,

    'm2': 'M.advanced_auto_labeler.v2',
    'm2.instruments': T.Graph.OutputPort('m1.data'),
    'm2.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/develop/datasource/deprecated/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/develop/bigexpr/usage.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)
shift(close, -5) / shift(open, -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)
""",
    'm2.start_date': '',
    'm2.end_date': '',
    'm2.benchmark': '000300.SHA',
    'm2.drop_na_label': True,
    'm2.cast_label_int': False,

    'm4': 'M.standardlize.v9',
    'm4.input_1': T.Graph.OutputPort('m2.data'),
    'm4.standard_func': 'ZScoreNorm',
    'm4.columns_input': 'label',

    'm3': 'M.input_features.v1',
    'm3.features': """pb_lf_0
return_20
ta_rsi_14_0
ta_ema_30_0
std(deal_number_0,30)
100*ta_bbi(close_30, 'long', timeperiod_1=3, timeperiod_2=6, timeperiod_3=12, timeperiod_4=24)/ta_bbi(close_30, 'short', timeperiod_1=3, timeperiod_2=6, timeperiod_3=12, timeperiod_4=24)
# ta_atr_14_0



""",

    'm15': 'M.general_feature_extractor.v7',
    'm15.instruments': T.Graph.OutputPort('m1.data'),
    'm15.features': T.Graph.OutputPort('m3.data'),
    'm15.start_date': '',
    'm15.end_date': '',
    'm15.before_start_days': 90,

    'm21': 'M.chinaa_stock_filter.v1',
    'm21.input_data': T.Graph.OutputPort('m15.data'),
    'm21.index_constituent_cond': ['全部'],
    'm21.board_cond': ['全部'],
    'm21.industry_cond': ['全部'],
    'm21.st_cond': ['正常'],
    'm21.delist_cond': ['非退市'],
    'm21.output_left_data': False,

    'm16': 'M.derived_feature_extractor.v3',
    'm16.input_data': T.Graph.OutputPort('m21.data'),
    'm16.features': T.Graph.OutputPort('m3.data'),
    'm16.date_col': 'date',
    'm16.instrument_col': 'instrument',
    'm16.drop_na': False,
    'm16.remove_extra_columns': False,

    'm7': 'M.join.v3',
    'm7.data1': T.Graph.OutputPort('m4.data'),
    'm7.data2': T.Graph.OutputPort('m16.data'),
    'm7.on': 'date,instrument',
    'm7.how': 'inner',
    'm7.sort': True,

    'm6': 'M.cached.v3',
    'm6.input_1': T.Graph.OutputPort('m7.data'),
    'm6.run': m6_run_bigquant_run,
    'm6.post_run': m6_post_run_bigquant_run,
    'm6.input_ports': '',
    'm6.params': '{}',
    'm6.output_ports': '',

    'm13': 'M.dropnan.v1',
    'm13.input_data': T.Graph.OutputPort('m6.data_1'),

    'm9': 'M.instruments.v2',
    'm9.start_date': T.live_run_param('trading_date', '2019-01-01'),
    'm9.end_date': T.live_run_param('trading_date', '2022-01-01'),
    'm9.market': 'CN_STOCK_A',
    'm9.instrument_list': '',
    'm9.max_count': 0,

    'm17': 'M.general_feature_extractor.v7',
    'm17.instruments': T.Graph.OutputPort('m9.data'),
    'm17.features': T.Graph.OutputPort('m3.data'),
    'm17.start_date': '',
    'm17.end_date': '',
    'm17.before_start_days': 90,

    'm18': 'M.derived_feature_extractor.v3',
    'm18.input_data': T.Graph.OutputPort('m17.data'),
    'm18.features': T.Graph.OutputPort('m3.data'),
    'm18.date_col': 'date',
    'm18.instrument_col': 'instrument',
    'm18.drop_na': False,
    'm18.remove_extra_columns': False,

    'm5': 'M.advanced_auto_labeler.v2',
    'm5.instruments': T.Graph.OutputPort('m9.data'),
    'm5.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/develop/datasource/deprecated/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/develop/bigexpr/usage.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)
shift(close, -5) / shift(open, -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)
""",
    'm5.start_date': '',
    'm5.end_date': '',
    'm5.benchmark': '000300.SHA',
    'm5.drop_na_label': True,
    'm5.cast_label_int': False,

    'm19': 'M.standardlize.v9',
    'm19.input_1': T.Graph.OutputPort('m5.data'),
    'm19.standard_func': 'ZScoreNorm',
    'm19.columns_input': 'label',

    'm20': 'M.join.v3',
    'm20.data1': T.Graph.OutputPort('m19.data'),
    'm20.data2': T.Graph.OutputPort('m18.data'),
    'm20.on': 'date,instrument',
    'm20.how': 'inner',
    'm20.sort': True,

    'm8': 'M.cached.v3',
    'm8.input_1': T.Graph.OutputPort('m20.data'),
    'm8.run': m8_run_bigquant_run,
    'm8.post_run': m8_post_run_bigquant_run,
    'm8.input_ports': '',
    'm8.params': '{}',
    'm8.output_ports': '',

    'm14': 'M.dropnan.v1',
    'm14.input_data': T.Graph.OutputPort('m8.data_1'),

    'm10': 'M.random_forest_regressor.v1',
    'm10.training_ds': T.Graph.OutputPort('m13.data'),
    'm10.features': T.Graph.OutputPort('m3.data'),
    'm10.predict_ds': T.Graph.OutputPort('m14.data'),
    'm10.iterations': 20,
    'm10.feature_fraction': 0.7,
    'm10.max_depth': 30,
    'm10.min_samples_per_leaf': 200,
    'm10.key_cols': 'date,instrument',
    'm10.workers': 4,
    'm10.random_state': 0,
    'm10.other_train_parameters': {
    'criterion':'mse',
},

    'm11': 'M.trade.v4',
    'm11.instruments': T.Graph.OutputPort('m9.data'),
    'm11.options_data': T.Graph.OutputPort('m10.predictions'),
    'm11.start_date': '',
    'm11.end_date': '',
    'm11.initialize': m11_initialize_bigquant_run,
    'm11.handle_data': m11_handle_data_bigquant_run,
    'm11.prepare': m11_prepare_bigquant_run,
    'm11.before_trading_start': m11_before_trading_start_bigquant_run,
    'm11.volume_limit': 0.025,
    'm11.order_price_field_buy': 'open',
    'm11.order_price_field_sell': 'close',
    'm11.capital_base': 1000000,
    'm11.auto_cancel_non_tradable_orders': True,
    'm11.data_frequency': 'daily',
    'm11.price_type': '后复权',
    'm11.product_type': '股票',
    'm11.plot_charts': True,
    'm11.backtest_only': False,
    'm11.benchmark': '000300.HIX',
})

# g.run({})


def m12_run_bigquant_run(
    bq_graph,
    inputs,
    trading_days_market='CN', # 使用那个市场的交易日历, TODO
    train_instruments_mid='m1', # 训练数据 证券代码列表 模块id
    test_instruments_mid='m9', # 测试数据 证券代码列表 模块id
    predict_mid='m10', # 预测 模块id
    trade_mid='m11', # 回测 模块id
    start_date='2019-01-04', # 数据开始日期
    end_date=T.live_run_param('trading_date', '2022-01-01'), # 数据结束日期
    train_update_days=250, # 更新周期，按交易日计算，每多少天更新一次
    train_update_days_for_live=None, #模拟实盘模式下的更新周期，按交易日计算，每多少天更新一次。如果需要在模拟实盘阶段使用不同的模型更新周期，可以设置这个参数
    train_data_min_days=230, # 最小数据天数，按交易日计算，所以第一个滚动的结束日期是 从开始日期到开始日期+最小数据天数
    train_data_max_days=250, # 最大数据天数，按交易日计算，0，表示没有限制，否则每一个滚动的开始日期=max(此滚动的结束日期-最大数据天数, 开始日期
    rolling_count_for_live=1, #实盘模式下滚动次数，模拟实盘模式下，取最后多少次滚动。一般在模拟实盘模式下，只用到最后一次滚动训练的模型，这里可以设置为1；如果你的滚动训练数据时间段很短，以至于期间可能没有训练数据，这里可以设置大一点。0表示没有限制
):
    def merge_datasources(input_1):
        df_list = [ds[0].read_df().set_index('date').loc[ds[1]:].reset_index() for ds in input_1]
        df = pd.concat(df_list)
        instrument_data = {
            'start_date': df['date'].min().strftime('%Y-%m-%d'),
            'end_date': df['date'].max().strftime('%Y-%m-%d'),
            'instruments': list(set(df['instrument'])),
        }
        return Outputs(data=DataSource.write_df(df), instrument_data=DataSource.write_pickle(instrument_data))

    def gen_rolling_dates(trading_days_market, start_date, end_date, train_update_days, train_update_days_for_live, train_data_min_days, train_data_max_days, rolling_count_for_live):
        # 是否实盘模式
        tdays = list(D.trading_days(market=trading_days_market, start_date=start_date, end_date=end_date)['date'])
        is_live_run = T.live_run_param('trading_date', None) is not None

        if is_live_run and train_update_days_for_live:
            train_update_days = train_update_days_for_live

        rollings = []
        train_end_date = train_data_min_days
        while train_end_date < len(tdays):
            if train_data_max_days is not None and train_data_max_days > 0:
                train_start_date = max(train_end_date - train_data_max_days, 0)
            else:
                train_start_date = 0
            rollings.append({
                'train_start_date': tdays[train_start_date].strftime('%Y-%m-%d'),
                'train_end_date': tdays[train_end_date - 1].strftime('%Y-%m-%d'),
                'test_start_date': tdays[train_end_date].strftime('%Y-%m-%d'),
                'test_end_date': tdays[min(train_end_date + train_update_days, len(tdays)) - 1].strftime('%Y-%m-%d'),
            })
            train_end_date += train_update_days

        if not rollings:
            raise Exception('没有滚动需要执行，请检查配置')

        if is_live_run and rolling_count_for_live:
            rollings = rollings[-rolling_count_for_live:]

        return rollings

    g = bq_graph

    rolling_dates = gen_rolling_dates(
        trading_days_market, start_date, end_date, train_update_days, train_update_days_for_live, train_data_min_days, train_data_max_days, rolling_count_for_live)

    # 训练和预测
    results = []
    for rolling in rolling_dates:
        parameters = {}
        # 先禁用回测
        parameters[trade_mid + '.__enabled__'] = False
        parameters[train_instruments_mid + '.start_date'] = rolling['train_start_date']
        parameters[train_instruments_mid + '.end_date'] = rolling['train_end_date']
        parameters[test_instruments_mid + '.start_date'] = rolling['test_start_date']
        parameters[test_instruments_mid + '.end_date'] = rolling['test_end_date']
        # print('------ rolling_train:', parameters)
        results.append(g.run(parameters))

    # 合并预测结果并回测
    mx = M.cached.v3(run=merge_datasources, input_1=[[result[predict_mid].predictions, result[test_instruments_mid].data.read_pickle()['start_date']] for result in results])
    parameters = {}
    parameters['*.__enabled__'] = False
    parameters[trade_mid + '.__enabled__'] = True
    parameters[trade_mid + '.instruments'] = mx.instrument_data
    parameters[trade_mid + '.options_data'] = mx.data

    trade = g.run(parameters)

    return {'rollings': results, 'trade': trade}


m12 = M.hyper_rolling_train.v1(
    run=m12_run_bigquant_run,
    run_now=True,
    bq_graph=g
)

[2022-08-08 16:47:38.781052] INFO: moduleinvoker: instruments.v2 开始运行..

[2022-08-08 16:47:38.789431] INFO: moduleinvoker: 命中缓存

[2022-08-08 16:47:38.791091] INFO: moduleinvoker: instruments.v2 运行完成[0.010043s].

[2022-08-08 16:47:38.795641] INFO: moduleinvoker: input_features.v1 开始运行..

[2022-08-08 16:47:38.823946] INFO: moduleinvoker: input_features.v1 运行完成[0.028292s].

[2022-08-08 16:47:38.829987] INFO: moduleinvoker: instruments.v2 开始运行..

[2022-08-08 16:47:38.835885] INFO: moduleinvoker: 命中缓存

[2022-08-08 16:47:38.837284] INFO: moduleinvoker: instruments.v2 运行完成[0.007296s].

[2022-08-08 16:47:38.846631] INFO: moduleinvoker: advanced_auto_labeler.v2 开始运行..

[2022-08-08 16:47:38.853137] INFO: moduleinvoker: 命中缓存

[2022-08-08 16:47:38.854462] INFO: moduleinvoker: advanced_auto_labeler.v2 运行完成[0.007828s].

[2022-08-08 16:47:38.867440] INFO: moduleinvoker: general_feature_extractor.v7 开始运行..

[2022-08-08 16:47:38.875987] INFO: moduleinvoker: 命中缓存

[2022-08-08 16:47:38.877534] INFO: moduleinvoker: general_feature_extractor.v7 运行完成[0.01009s].

[2022-08-08 16:47:38.885897] INFO: moduleinvoker: advanced_auto_labeler.v2 开始运行..

[2022-08-08 16:47:38.893352] INFO: moduleinvoker: 命中缓存

[2022-08-08 16:47:38.894837] INFO: moduleinvoker: advanced_auto_labeler.v2 运行完成[0.008937s].

[2022-08-08 16:47:38.907197] INFO: moduleinvoker: general_feature_extractor.v7 开始运行..

[2022-08-08 16:47:38.913013] INFO: moduleinvoker: 命中缓存

[2022-08-08 16:47:38.914403] INFO: moduleinvoker: general_feature_extractor.v7 运行完成[0.00721s].

[2022-08-08 16:47:38.919432] INFO: moduleinvoker: standardlize.v9 开始运行..

[2022-08-08 16:47:38.925553] INFO: moduleinvoker: 命中缓存

[2022-08-08 16:47:38.927145] INFO: moduleinvoker: standardlize.v9 运行完成[0.007707s].

[2022-08-08 16:47:38.935038] INFO: moduleinvoker: chinaa_stock_filter.v1 开始运行..

[2022-08-08 16:47:38.941568] INFO: moduleinvoker: 命中缓存

[2022-08-08 16:47:38.943383] INFO: moduleinvoker: chinaa_stock_filter.v1 运行完成[0.008343s].

[2022-08-08 16:47:38.948093] INFO: moduleinvoker: standardlize.v9 开始运行..

[2022-08-08 16:47:38.955966] INFO: moduleinvoker: 命中缓存

[2022-08-08 16:47:38.957263] INFO: moduleinvoker: standardlize.v9 运行完成[0.009167s].

[2022-08-08 16:47:38.984809] INFO: moduleinvoker: derived_feature_extractor.v3 开始运行..

[2022-08-08 16:47:41.712187] INFO: derived_feature_extractor: 提取完成 std(deal_number_0,30), 0.826s

[2022-08-08 16:47:54.076980] INFO: derived_feature_extractor: 提取完成 100*ta_bbi(close_30, 'long', timeperiod_1=3, timeperiod_2=6, timeperiod_3=12, timeperiod_4=24)/ta_bbi(close_30, 'short', timeperiod_1=3, timeperiod_2=6, timeperiod_3=12, timeperiod_4=24), 12.363s

[2022-08-08 16:47:54.607436] INFO: derived_feature_extractor: /y_2019, 262339

[2022-08-08 16:47:55.928608] INFO: derived_feature_extractor: /y_2020, 925405

[2022-08-08 16:47:56.286675] INFO: moduleinvoker: derived_feature_extractor.v3 运行完成[17.301846s].

[2022-08-08 16:47:56.294481] INFO: moduleinvoker: derived_feature_extractor.v3 开始运行..

[2022-08-08 16:47:59.054627] INFO: derived_feature_extractor: 提取完成 std(deal_number_0,30), 0.745s

[2022-08-08 16:48:11.619049] INFO: derived_feature_extractor: 提取完成 100*ta_bbi(close_30, 'long', timeperiod_1=3, timeperiod_2=6, timeperiod_3=12, timeperiod_4=24)/ta_bbi(close_30, 'short', timeperiod_1=3, timeperiod_2=6, timeperiod_3=12, timeperiod_4=24), 12.563s

[2022-08-08 16:48:13.026879] INFO: derived_feature_extractor: /y_2018, 205974

[2022-08-08 16:48:14.845708] INFO: derived_feature_extractor: /y_2019, 813785

[2022-08-08 16:48:15.918344] INFO: moduleinvoker: derived_feature_extractor.v3 运行完成[19.623852s].

[2022-08-08 16:48:15.948379] INFO: moduleinvoker: join.v3 开始运行..

[2022-08-08 16:48:20.757943] INFO: join: /y_2019, 行数=44663/262339, 耗时=1.691993s

[2022-08-08 16:48:26.317208] INFO: join: /y_2020, 行数=898935/925405, 耗时=5.556716s

[2022-08-08 16:48:26.399071] INFO: join: 最终行数: 943598

[2022-08-08 16:48:26.424372] INFO: moduleinvoker: join.v3 运行完成[10.475995s].

[2022-08-08 16:48:26.434446] INFO: moduleinvoker: join.v3 开始运行..

[2022-08-08 16:48:30.743677] INFO: join: /y_2018, 行数=0/205974, 耗时=0.956953s

[2022-08-08 16:48:33.359951] INFO: join: /y_2019, 行数=786448/813785, 耗时=2.613955s

[2022-08-08 16:48:33.649753] INFO: join: 最终行数: 786448

[2022-08-08 16:48:33.668439] INFO: moduleinvoker: join.v3 运行完成[7.233984s].

[2022-08-08 16:48:33.686628] INFO: moduleinvoker: cached.v3 开始运行..

[2022-08-08 16:48:36.748265] INFO: moduleinvoker: cached.v3 运行完成[3.061645s].

[2022-08-08 16:48:36.759580] INFO: moduleinvoker: cached.v3 开始运行..

[2022-08-08 16:48:40.722237] INFO: moduleinvoker: cached.v3 运行完成[3.962648s].

[2022-08-08 16:48:40.734406] INFO: moduleinvoker: dropnan.v1 开始运行..

[2022-08-08 16:48:43.542785] INFO: dropnan: /data, 470215/943598

[2022-08-08 16:48:43.660694] INFO: dropnan: 行数: 470215/943598

[2022-08-08 16:48:43.669711] INFO: moduleinvoker: dropnan.v1 运行完成[2.935293s].

[2022-08-08 16:48:43.686452] INFO: moduleinvoker: dropnan.v1 开始运行..

[2022-08-08 16:48:45.635341] INFO: dropnan: /data, 396926/786448

[2022-08-08 16:48:45.714315] INFO: dropnan: 行数: 396926/786448

[2022-08-08 16:48:45.726345] INFO: moduleinvoker: dropnan.v1 运行完成[2.039894s].

[2022-08-08 16:48:45.735467] INFO: moduleinvoker: random_forest_regressor.v1 开始运行..

[2022-08-08 16:49:00.526356] INFO: moduleinvoker: random_forest_regressor.v1 运行完成[14.790873s].

[2022-08-08 16:49:00.533553] INFO: moduleinvoker: instruments.v2 开始运行..

[2022-08-08 16:49:00.540704] INFO: moduleinvoker: 命中缓存

[2022-08-08 16:49:00.542716] INFO: moduleinvoker: instruments.v2 运行完成[0.009162s].

[2022-08-08 16:49:00.546958] INFO: moduleinvoker: input_features.v1 开始运行..

[2022-08-08 16:49:00.552250] INFO: moduleinvoker: 命中缓存

[2022-08-08 16:49:00.553665] INFO: moduleinvoker: input_features.v1 运行完成[0.006723s].

[2022-08-08 16:49:00.558632] INFO: moduleinvoker: instruments.v2 开始运行..

[2022-08-08 16:49:00.564793] INFO: moduleinvoker: 命中缓存

[2022-08-08 16:49:00.566230] INFO: moduleinvoker: instruments.v2 运行完成[0.007595s].

[2022-08-08 16:49:00.587856] INFO: moduleinvoker: advanced_auto_labeler.v2 开始运行..

[2022-08-08 16:49:00.595515] INFO: moduleinvoker: 命中缓存

[2022-08-08 16:49:00.598088] INFO: moduleinvoker: advanced_auto_labeler.v2 运行完成[0.010249s].

[2022-08-08 16:49:00.614056] INFO: moduleinvoker: general_feature_extractor.v7 开始运行..

[2022-08-08 16:49:00.622153] INFO: moduleinvoker: 命中缓存

[2022-08-08 16:49:00.624134] INFO: moduleinvoker: general_feature_extractor.v7 运行完成[0.010092s].

[2022-08-08 16:49:00.632903] INFO: moduleinvoker: advanced_auto_labeler.v2 开始运行..

[2022-08-08 16:49:00.648991] INFO: moduleinvoker: 命中缓存

[2022-08-08 16:49:00.650951] INFO: moduleinvoker: advanced_auto_labeler.v2 运行完成[0.018054s].

[2022-08-08 16:49:00.662941] INFO: moduleinvoker: general_feature_extractor.v7 开始运行..

[2022-08-08 16:49:00.677060] INFO: moduleinvoker: 命中缓存

[2022-08-08 16:49:00.679388] INFO: moduleinvoker: general_feature_extractor.v7 运行完成[0.016439s].

[2022-08-08 16:49:00.686673] INFO: moduleinvoker: standardlize.v9 开始运行..

[2022-08-08 16:49:00.693418] INFO: moduleinvoker: 命中缓存

[2022-08-08 16:49:00.695173] INFO: moduleinvoker: standardlize.v9 运行完成[0.008506s].

[2022-08-08 16:49:00.705233] INFO: moduleinvoker: chinaa_stock_filter.v1 开始运行..

[2022-08-08 16:49:00.712059] INFO: moduleinvoker: 命中缓存

[2022-08-08 16:49:00.713798] INFO: moduleinvoker: chinaa_stock_filter.v1 运行完成[0.008568s].

[2022-08-08 16:49:00.719220] INFO: moduleinvoker: standardlize.v9 开始运行..

[2022-08-08 16:49:00.727728] INFO: moduleinvoker: 命中缓存

[2022-08-08 16:49:00.729407] INFO: moduleinvoker: standardlize.v9 运行完成[0.010183s].

[2022-08-08 16:49:00.736552] INFO: moduleinvoker: derived_feature_extractor.v3 开始运行..

[2022-08-08 16:49:03.955661] INFO: derived_feature_extractor: 提取完成 std(deal_number_0,30), 0.953s

[2022-08-08 16:49:18.145579] INFO: derived_feature_extractor: 提取完成 100*ta_bbi(close_30, 'long', timeperiod_1=3, timeperiod_2=6, timeperiod_3=12, timeperiod_4=24)/ta_bbi(close_30, 'short', timeperiod_1=3, timeperiod_2=6, timeperiod_3=12, timeperiod_4=24), 14.188s

[2022-08-08 16:49:18.930136] INFO: derived_feature_extractor: /y_2020, 255775

[2022-08-08 16:49:20.549467] INFO: derived_feature_extractor: /y_2021, 1061527

[2022-08-08 16:49:21.149779] INFO: moduleinvoker: derived_feature_extractor.v3 运行完成[20.413206s].

[2022-08-08 16:49:21.158011] INFO: moduleinvoker: derived_feature_extractor.v3 开始运行..

[2022-08-08 16:49:23.790143] INFO: derived_feature_extractor: 提取完成 std(deal_number_0,30), 0.797s

[2022-08-08 16:49:35.933538] INFO: derived_feature_extractor: 提取完成 100*ta_bbi(close_30, 'long', timeperiod_1=3, timeperiod_2=6, timeperiod_3=12, timeperiod_4=24)/ta_bbi(close_30, 'short', timeperiod_1=3, timeperiod_2=6, timeperiod_3=12, timeperiod_4=24), 12.141s

[2022-08-08 16:49:36.469722] INFO: derived_feature_extractor: /y_2019, 253180

[2022-08-08 16:49:37.713263] INFO: derived_feature_extractor: /y_2020, 883534

[2022-08-08 16:49:38.042911] INFO: moduleinvoker: derived_feature_extractor.v3 运行完成[16.884886s].

[2022-08-08 16:49:38.053468] INFO: moduleinvoker: join.v3 开始运行..

[2022-08-08 16:49:40.835919] INFO: join: /y_2020, 行数=20490/255775, 耗时=0.894199s

[2022-08-08 16:49:43.904376] INFO: join: /y_2021, 行数=1035254/1061527, 耗时=3.065812s

[2022-08-08 16:49:43.995145] INFO: join: 最终行数: 1055744

[2022-08-08 16:49:44.008897] INFO: moduleinvoker: join.v3 运行完成[5.955419s].

[2022-08-08 16:49:44.022266] INFO: moduleinvoker: join.v3 开始运行..

[2022-08-08 16:49:46.434525] INFO: join: /y_2019, 行数=43147/253180, 耗时=0.831003s

[2022-08-08 16:49:48.878233] INFO: join: /y_2020, 行数=859430/883534, 耗时=2.440957s

[2022-08-08 16:49:48.949638] INFO: join: 最终行数: 902577

[2022-08-08 16:49:48.961943] INFO: moduleinvoker: join.v3 运行完成[4.939667s].

[2022-08-08 16:49:48.990118] INFO: moduleinvoker: cached.v3 开始运行..

[2022-08-08 16:49:52.241956] INFO: moduleinvoker: cached.v3 运行完成[3.251837s].

[2022-08-08 16:49:52.256418] INFO: moduleinvoker: cached.v3 开始运行..

[2022-08-08 16:49:54.995293] INFO: moduleinvoker: cached.v3 运行完成[2.738878s].

[2022-08-08 16:49:55.004650] INFO: moduleinvoker: dropnan.v1 开始运行..

[2022-08-08 16:49:57.794537] INFO: dropnan: /data, 562593/1055744

[2022-08-08 16:49:57.850700] INFO: dropnan: 行数: 562593/1055744

[2022-08-08 16:49:57.859611] INFO: moduleinvoker: dropnan.v1 运行完成[2.854954s].

[2022-08-08 16:49:57.882475] INFO: moduleinvoker: dropnan.v1 开始运行..

[2022-08-08 16:50:00.396112] INFO: dropnan: /data, 447303/902577

[2022-08-08 16:50:00.495105] INFO: dropnan: 行数: 447303/902577

[2022-08-08 16:50:00.570060] INFO: moduleinvoker: dropnan.v1 运行完成[2.687578s].

[2022-08-08 16:50:00.586487] INFO: moduleinvoker: random_forest_regressor.v1 开始运行..

[2022-08-08 16:50:19.770553] INFO: moduleinvoker: random_forest_regressor.v1 运行完成[19.18406s].

[2022-08-08 16:50:19.807013] INFO: moduleinvoker: cached.v3 开始运行..

[2022-08-08 16:50:21.840793] INFO: moduleinvoker: cached.v3 运行完成[2.033785s].

[2022-08-08 16:50:21.901986] INFO: moduleinvoker: backtest.v8 开始运行..

[2022-08-08 16:50:21.907506] INFO: backtest: biglearning backtest:V8.6.2

[2022-08-08 16:50:21.908876] INFO: backtest: product_type:stock by specified

[2022-08-08 16:50:22.093229] INFO: moduleinvoker: cached.v2 开始运行..

[2022-08-08 16:50:29.407889] INFO: backtest: 读取股票行情完成:3271945

[2022-08-08 16:50:32.851888] INFO: moduleinvoker: cached.v2 运行完成[10.758666s].

[2022-08-08 16:50:53.334769] INFO: algo: TradingAlgorithm V1.8.8

[2022-08-08 16:50:55.747936] INFO: algo: trading transform...

[2022-08-08 16:51:00.406627] WARNING: Performance: maybe_close_position no price for asset:Equity(5567 [300156.SZA]), field:price, dt:2020-08-25 15:00:00+00:00

[2022-08-08 16:51:08.517164] INFO: Performance: Simulated 493 trading days out of 493.

[2022-08-08 16:51:08.518983] INFO: Performance: first open: 2019-12-16 09:30:00+00:00

[2022-08-08 16:51:08.520312] INFO: Performance: last close: 2021-12-24 15:00:00+00:00

[2022-08-08 16:51:14.170721] INFO: moduleinvoker: backtest.v8 运行完成[52.26874s].

[2022-08-08 16:51:14.172521] INFO: moduleinvoker: trade.v4 运行完成[52.323782s].

[2022-08-08 16:51:14.174154] INFO: moduleinvoker: hyper_rolling_train.v1 运行完成[215.430826s].

收益率73.95%
年化收益率32.71%
基准收益率24.02%
阿尔法0.22
贝塔0.84
夏普比率1.1
胜率0.51
盈亏比1.24
收益波动率26.07%
信息比率0.06
最大回撤21.3%

bigcharts-data-start/{"__type":"tabs","__id":"bigchart-c6aee084584845b0922c48a8e3ef98ee"}/bigcharts-data-end

In [8]:

# 第一次训练的训练集
print(m12.result['rollings'][0]['m1'].data.read_pickle()['start_date'])
print(m12.result['rollings'][0]['m1'].data.read_pickle()['end_date'])
# 第一次训练的模型的特征得分
# m12.result['rollings'][0]['m6'].feature_gains.read_df()
# 第一次训练的预测集
print(m12.result['rollings'][0]['m9'].data.read_pickle()['start_date'])
print(m12.result['rollings'][0]['m9'].data.read_pickle()['end_date'])
# 训练的第一个模型的预测结果
# m12.result['rollings'][0]['m8'].predictions.read_df().set_index('date').ix['2017-02-14'].head()
# 第二次训练的训练集
print(m12.result['rollings'][1]['m1'].data.read_pickle()['start_date'])
print(m12.result['rollings'][1]['m1'].data.read_pickle()['end_date'])

In [9]:

m12.result['rollings'][0]['m10'].feature_gains()

Out[9]:

	feature	gain
1	return_20	0.270511
3	ta_ema_30_0	0.205657
4	std(deal_number_0,30)	0.186926
0	pb_lf_0	0.175962
2	ta_rsi_14_0	0.160944
5	100*ta_bbi(close_30, 'long', timeperiod_1=3, t...	0.000000

In [10]:

len(m12.result['rollings'])

Out[10]:

In [ ]: