请问这个Bidirectional层的线应该怎么连

a1750271086 · 2020-09-23 09:37:08 UTC

克隆策略

In [4]:

# 本代码由可视化策略环境自动生成 2020年9月23日 16:34
# 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 m24_run_bigquant_run(input_1, input_2, input_3):
    # 示例代码如下。在这里编写您的代码
    pred_label = input_1.read_pickle()
    df = input_2.read_df()
    df = pd.DataFrame({'pred_label':pred_label[:,0], 'instrument':df.instrument, 'date':df.date})
    df.sort_values(['date','pred_label'],inplace=True, ascending=[True,False])
    return Outputs(data_1=DataSource.write_df(df), 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 m24_post_run_bigquant_run(outputs):
    return outputs

# 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 = 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 = 0.1
    context.options['hold_days'] = 1
# 回测引擎：每日数据处理函数，每天执行一次
def m19_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
    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 m19_prepare_bigquant_run(context):
    pass


g = T.Graph({

    'm1': 'M.instruments.v2',
    'm1.start_date': '2018-01-01',
    '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/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)
shift(close, -5) / 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))

# 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,

    'm13': 'M.standardlize.v8',
    'm13.input_1': T.Graph.OutputPort('m2.data'),
    'm13.columns_input': 'label',

    'm3': 'M.input_features.v1',
    'm3.features': """(close_0-mean(close_0,12))/mean(close_0,12)*100
rank(std(amount_0,15))
rank_avg_amount_0/rank_avg_amount_8
ts_argmin(low_0,20)
rank_return_30
(low_1-close_0)/close_0
ta_bbands_lowerband_14_0
mean(mf_net_pct_s_0,4)
amount_0/avg_amount_3
return_0/return_5
return_1/return_5
rank_avg_amount_7/rank_avg_amount_10
ta_sma_10_0/close_0
sqrt(high_0*low_0)-amount_0/volume_0*adjust_factor_0
avg_turn_15/(turn_0+1e-5)
return_10
mf_net_pct_s_0
(close_0-open_0)/close_1""",

    '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': 30,

    'm16': 'M.derived_feature_extractor.v3',
    'm16.input_data': T.Graph.OutputPort('m15.data'),
    'm16.features': T.Graph.OutputPort('m3.data'),
    'm16.date_col': 'date',
    'm16.instrument_col': 'instrument',
    'm16.drop_na': True,
    'm16.remove_extra_columns': False,

    'm14': 'M.standardlize.v8',
    'm14.input_1': T.Graph.OutputPort('m16.data'),
    'm14.input_2': T.Graph.OutputPort('m3.data'),
    'm14.columns_input': '',

    'm7': 'M.join.v3',
    'm7.data1': T.Graph.OutputPort('m13.data'),
    'm7.data2': T.Graph.OutputPort('m14.data'),
    'm7.on': 'date,instrument',
    'm7.how': 'inner',
    'm7.sort': False,

    'm26': 'M.dl_convert_to_bin.v2',
    'm26.input_data': T.Graph.OutputPort('m7.data'),
    'm26.features': T.Graph.OutputPort('m3.data'),
    'm26.window_size': 2,
    'm26.feature_clip': 18,
    'm26.flatten': False,
    'm26.window_along_col': 'instrument',

    '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', '2020-09-18'),
    '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': 30,

    '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': True,
    'm18.remove_extra_columns': False,

    'm25': 'M.standardlize.v8',
    'm25.input_1': T.Graph.OutputPort('m18.data'),
    'm25.input_2': T.Graph.OutputPort('m3.data'),
    'm25.columns_input': '',

    'm27': 'M.dl_convert_to_bin.v2',
    'm27.input_data': T.Graph.OutputPort('m25.data'),
    'm27.features': T.Graph.OutputPort('m3.data'),
    'm27.window_size': 2,
    'm27.feature_clip': 18,
    'm27.flatten': False,
    'm27.window_along_col': 'instrument',

    'm6': 'M.dl_layer_input.v1',
    'm6.shape': '2,18',
    'm6.batch_shape': '',
    'm6.dtype': 'float32',
    'm6.sparse': False,
    'm6.name': '',

    'm10': 'M.dl_layer_lstm.v1',
    'm10.inputs': T.Graph.OutputPort('m6.data'),
    'm10.units': 128,
    'm10.activation': 'tanh',
    'm10.recurrent_activation': 'hard_sigmoid',
    'm10.use_bias': True,
    'm10.kernel_initializer': 'glorot_uniform',
    'm10.recurrent_initializer': 'Orthogonal',
    'm10.bias_initializer': 'Zeros',
    'm10.unit_forget_bias': True,
    'm10.kernel_regularizer': 'None',
    'm10.kernel_regularizer_l1': 0,
    'm10.kernel_regularizer_l2': 0,
    'm10.recurrent_regularizer': 'None',
    'm10.recurrent_regularizer_l1': 0,
    'm10.recurrent_regularizer_l2': 0,
    'm10.bias_regularizer': 'None',
    'm10.bias_regularizer_l1': 0,
    'm10.bias_regularizer_l2': 0,
    'm10.activity_regularizer': 'None',
    'm10.activity_regularizer_l1': 0,
    'm10.activity_regularizer_l2': 0,
    'm10.kernel_constraint': 'None',
    'm10.recurrent_constraint': 'None',
    'm10.bias_constraint': 'None',
    'm10.dropout': 0,
    'm10.recurrent_dropout': 0,
    'm10.return_sequences': True,
    'm10.implementation': '1',
    'm10.name': '',

    'm12': 'M.dl_layer_dropout.v1',
    'm12.inputs': T.Graph.OutputPort('m10.data'),
    'm12.rate': 0.2,
    'm12.noise_shape': '',
    'm12.name': '',

    'm8': 'M.dl_layer_lstm.v1',
    'm8.inputs': T.Graph.OutputPort('m12.data'),
    'm8.units': 128,
    'm8.activation': 'tanh',
    'm8.recurrent_activation': 'hard_sigmoid',
    'm8.use_bias': True,
    'm8.kernel_initializer': 'glorot_uniform',
    'm8.recurrent_initializer': 'Orthogonal',
    'm8.bias_initializer': 'Zeros',
    'm8.unit_forget_bias': True,
    'm8.kernel_regularizer': 'None',
    'm8.kernel_regularizer_l1': 0,
    'm8.kernel_regularizer_l2': 0,
    'm8.recurrent_regularizer': 'None',
    'm8.recurrent_regularizer_l1': 0,
    'm8.recurrent_regularizer_l2': 0,
    'm8.bias_regularizer': 'None',
    'm8.bias_regularizer_l1': 0,
    'm8.bias_regularizer_l2': 0,
    'm8.activity_regularizer': 'None',
    'm8.activity_regularizer_l1': 0,
    'm8.activity_regularizer_l2': 0,
    'm8.kernel_constraint': 'None',
    'm8.recurrent_constraint': 'None',
    'm8.bias_constraint': 'None',
    'm8.dropout': 0,
    'm8.recurrent_dropout': 0,
    'm8.return_sequences': True,
    'm8.implementation': '2',
    'm8.name': '',

    'm22': 'M.dl_layer_bidirectional.v1',
    'm22.layer': T.Graph.OutputPort('m8.data'),
    'm22.inputs': T.Graph.OutputPort('m6.data'),
    'm22.merge_mode': 'concat',
    'm22.weights': '',
    'm22.name': '',

    'm21': 'M.dl_layer_dropout.v1',
    'm21.inputs': T.Graph.OutputPort('m22.data'),
    'm21.rate': 0.2,
    'm21.noise_shape': '',
    'm21.name': '',

    'm20': 'M.dl_layer_dense.v1',
    'm20.inputs': T.Graph.OutputPort('m21.data'),
    'm20.units': 1,
    'm20.activation': 'tanh',
    'm20.use_bias': True,
    'm20.kernel_initializer': 'glorot_uniform',
    'm20.bias_initializer': 'Zeros',
    'm20.kernel_regularizer': 'None',
    'm20.kernel_regularizer_l1': 0,
    'm20.kernel_regularizer_l2': 0,
    'm20.bias_regularizer': 'None',
    'm20.bias_regularizer_l1': 0,
    'm20.bias_regularizer_l2': 0,
    'm20.activity_regularizer': 'None',
    'm20.activity_regularizer_l1': 0,
    'm20.activity_regularizer_l2': 0,
    'm20.kernel_constraint': 'None',
    'm20.bias_constraint': 'None',
    'm20.name': '',

    'm34': 'M.dl_model_init.v1',
    'm34.inputs': T.Graph.OutputPort('m6.data'),
    'm34.outputs': T.Graph.OutputPort('m20.data'),

    'm5': 'M.dl_model_train.v1',
    'm5.input_model': T.Graph.OutputPort('m34.data'),
    'm5.training_data': T.Graph.OutputPort('m26.data'),
    'm5.validation_data': T.Graph.OutputPort('m26.data'),
    'm5.optimizer': 'RMSprop',
    'm5.loss': 'mean_squared_error',
    'm5.metrics': 'mae',
    'm5.batch_size': 2048,
    'm5.epochs': 200,
    'm5.n_gpus': 0,
    'm5.verbose': '2:每个epoch输出一行记录',

    'm11': 'M.dl_model_predict.v1',
    'm11.trained_model': T.Graph.OutputPort('m5.data'),
    'm11.input_data': T.Graph.OutputPort('m27.data'),
    'm11.batch_size': 1024,
    'm11.n_gpus': 0,
    'm11.verbose': '2:每个epoch输出一行记录',

    'm24': 'M.cached.v3',
    'm24.input_1': T.Graph.OutputPort('m11.data'),
    'm24.input_2': T.Graph.OutputPort('m18.data'),
    'm24.run': m24_run_bigquant_run,
    'm24.post_run': m24_post_run_bigquant_run,
    'm24.input_ports': '',
    'm24.params': '{}',
    'm24.output_ports': '',

    'm19': 'M.trade.v4',
    'm19.instruments': T.Graph.OutputPort('m9.data'),
    'm19.options_data': T.Graph.OutputPort('m24.data_1'),
    'm19.start_date': '',
    'm19.end_date': '',
    'm19.initialize': m19_initialize_bigquant_run,
    'm19.handle_data': m19_handle_data_bigquant_run,
    'm19.prepare': m19_prepare_bigquant_run,
    'm19.volume_limit': 0,
    'm19.order_price_field_buy': 'open',
    'm19.order_price_field_sell': 'close',
    'm19.capital_base': 1000000,
    'm19.auto_cancel_non_tradable_orders': True,
    'm19.data_frequency': 'daily',
    'm19.price_type': '后复权',
    'm19.product_type': '股票',
    'm19.plot_charts': True,
    'm19.backtest_only': False,
    'm19.benchmark': '000300.SHA',
})

# g.run({})


def m4_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='m24', # 预测 模块id
    trade_mid='m19', # 回测 模块id
    start_date='2019-01-01', # 数据开始日期
    end_date=T.live_run_param('trading_date', '2020-09-18'), # 数据结束日期
    train_update_days=125, # 更新周期，按交易日计算，每多少天更新一次
    train_update_days_for_live=125, #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=125, # 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').ix[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].data_1, 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}


m4 = M.hyper_rolling_train.v1(
    run=m4_run_bigquant_run,
    run_now=True,
    bq_graph=g
)

[2020-09-23 16:32:06.830741] INFO: moduleinvoker: instruments.v2 开始运行..

[2020-09-23 16:32:06.837343] INFO: moduleinvoker: 命中缓存

[2020-09-23 16:32:06.838849] INFO: moduleinvoker: instruments.v2 运行完成[0.008111s].

[2020-09-23 16:32:06.841000] INFO: moduleinvoker: input_features.v1 开始运行..

[2020-09-23 16:32:06.846437] INFO: moduleinvoker: 命中缓存

[2020-09-23 16:32:06.849565] INFO: moduleinvoker: input_features.v1 运行完成[0.008544s].

[2020-09-23 16:32:06.851939] INFO: moduleinvoker: instruments.v2 开始运行..

[2020-09-23 16:32:06.857314] INFO: moduleinvoker: 命中缓存

[2020-09-23 16:32:06.859014] INFO: moduleinvoker: instruments.v2 运行完成[0.007064s].

[2020-09-23 16:32:06.865170] INFO: moduleinvoker: dl_layer_input.v1 运行完成[0.003352s].

[2020-09-23 16:32:06.867655] INFO: moduleinvoker: advanced_auto_labeler.v2 开始运行..

[2020-09-23 16:32:06.872599] INFO: moduleinvoker: 命中缓存

[2020-09-23 16:32:06.874164] INFO: moduleinvoker: advanced_auto_labeler.v2 运行完成[0.006501s].

[2020-09-23 16:32:06.879767] INFO: moduleinvoker: general_feature_extractor.v7 开始运行..

[2020-09-23 16:32:06.884909] INFO: moduleinvoker: 命中缓存

[2020-09-23 16:32:06.886112] INFO: moduleinvoker: general_feature_extractor.v7 运行完成[0.006343s].

[2020-09-23 16:32:06.892410] INFO: moduleinvoker: general_feature_extractor.v7 开始运行..

[2020-09-23 16:32:06.897261] INFO: moduleinvoker: 命中缓存

[2020-09-23 16:32:06.898634] INFO: moduleinvoker: general_feature_extractor.v7 运行完成[0.006223s].

[2020-09-23 16:32:07.395368] INFO: moduleinvoker: dl_layer_lstm.v1 运行完成[0.494655s].

[2020-09-23 16:32:07.397812] INFO: moduleinvoker: standardlize.v8 开始运行..

[2020-09-23 16:32:07.405324] INFO: moduleinvoker: 命中缓存

[2020-09-23 16:32:07.406824] INFO: moduleinvoker: standardlize.v8 运行完成[0.009015s].

[2020-09-23 16:32:07.408814] INFO: moduleinvoker: derived_feature_extractor.v3 开始运行..

[2020-09-23 16:32:07.426239] INFO: moduleinvoker: 命中缓存

[2020-09-23 16:32:07.427644] INFO: moduleinvoker: derived_feature_extractor.v3 运行完成[0.018823s].

[2020-09-23 16:32:07.429288] INFO: moduleinvoker: derived_feature_extractor.v3 开始运行..

[2020-09-23 16:32:07.448952] INFO: moduleinvoker: 命中缓存

[2020-09-23 16:32:07.450723] INFO: moduleinvoker: derived_feature_extractor.v3 运行完成[0.021405s].

[2020-09-23 16:32:07.475984] INFO: moduleinvoker: dl_layer_dropout.v1 运行完成[0.023009s].

[2020-09-23 16:32:07.477653] INFO: moduleinvoker: standardlize.v8 开始运行..

[2020-09-23 16:32:07.482916] INFO: moduleinvoker: 命中缓存

[2020-09-23 16:32:07.484359] INFO: moduleinvoker: standardlize.v8 运行完成[0.006703s].

[2020-09-23 16:32:07.486525] INFO: moduleinvoker: standardlize.v8 开始运行..

[2020-09-23 16:32:07.491697] INFO: moduleinvoker: 命中缓存

[2020-09-23 16:32:07.492692] INFO: moduleinvoker: standardlize.v8 运行完成[0.006165s].

[2020-09-23 16:32:07.666621] INFO: moduleinvoker: dl_layer_lstm.v1 运行完成[0.171397s].

[2020-09-23 16:32:07.669142] INFO: moduleinvoker: join.v3 开始运行..

[2020-09-23 16:32:07.674901] INFO: moduleinvoker: 命中缓存

[2020-09-23 16:32:07.676596] INFO: moduleinvoker: join.v3 运行完成[0.00744s].

[2020-09-23 16:32:07.682950] INFO: moduleinvoker: dl_convert_to_bin.v2 开始运行..

[2020-09-23 16:32:07.688854] INFO: moduleinvoker: 命中缓存

[2020-09-23 16:32:07.690494] INFO: moduleinvoker: dl_convert_to_bin.v2 运行完成[0.007561s].

[2020-09-23 16:32:07.692871] ERROR: moduleinvoker: module name: dl_layer_bidirectional, module version: v1, trackeback: Traceback (most recent call last):
ValueError: Please initialize `Bidirectional` layer with a `Layer` instance. You passed: Tensor("lstm_7/Identity:0", shape=(None, 2, 128), dtype=float32)

[2020-09-23 16:32:07.696777] ERROR: moduleinvoker: module name: hyper_rolling_train, module version: v1, trackeback: Traceback (most recent call last):
ValueError: Please initialize `Bidirectional` layer with a `Layer` instance. You passed: Tensor("lstm_7/Identity:0", shape=(None, 2, 128), dtype=float32)

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
<ipython-input-4-8a259d5feaa9> in <module>()
    457     run=m4_run_bigquant_run,
    458     run_now=True,
--> 459     bq_graph=g
    460 )

<ipython-input-4-8a259d5feaa9> in m4_run_bigquant_run(bq_graph, inputs, trading_days_market, train_instruments_mid, test_instruments_mid, predict_mid, trade_mid, start_date, end_date, train_update_days, train_update_days_for_live, train_data_min_days, train_data_max_days, rolling_count_for_live)
    439         parameters[test_instruments_mid + '.end_date'] = rolling['test_end_date']
    440         # print('------ rolling_train:', parameters)
--> 441         results.append(g.run(parameters))
    442 
    443     # 合并预测结果并回测

ValueError: Please initialize `Bidirectional` layer with a `Layer` instance. You passed: Tensor("lstm_7/Identity:0", shape=(None, 2, 128), dtype=float32)

a1750271086 · 2020-09-23 09:46:33 UTC

还有一个获取L2数据的模块的两个input应该连哪两条线？谢谢老师