复制链接

克隆策略

大盘风控的一些案例¶

思路三:训练一个LSTM神经网络模型预测涨跌，来作为大盘风控的依据¶

1.我们先用模板的lstm+cnn深度学习预测股价涨跌的案例，训练出一个专门用于风控的指标 2.把这个预测的结果跟我们的stockRanker模型的预测结果进行合并 3.在回测引擎中进行编写择时控制逻辑的代码完成风控

大盘风控模块，读取深度学习模型预测涨跌的风控数据¶

    benckmark_risk=ranker_prediction['result'].values[0]
    # 若预测涨跌的概率小于0.49则不开仓 全仓风控

In [116]:

# 本代码由可视化策略环境自动生成 2022年3月20日 23:49
# 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。


# 用户的自定义层需要写到字典中，比如
# {
#   "MyLayer": MyLayer
# }
m23_custom_objects_bigquant_run = {
    
}

# 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 m21_run_bigquant_run(input_1, input_2, input_3):

   
    test_data = input_2.read_pickle()
    pred_label = input_1.read_pickle()
    pred_result = pred_label.reshape(pred_label.shape[0]) 
    dt = input_3.read_df()['date'][-1*len(pred_result):]
    pred_df = pd.Series(pred_result, index=dt)
    ds = DataSource.write_df(pred_df)
    return Outputs(data_1=ds)

# 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 m21_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 m5_run_bigquant_run(input_1, input_2, input_3):
    # 示例代码如下。在这里编写您的代码
    df = input_1.read()
    
    data_2 = DataSource.write_pickle(df)
    #给空列命名 为 result
    df1=pd.DataFrame(df)
    df1.columns=['result']
    #df1
    data_1 = DataSource.write_df(df1)
    return Outputs(data_1=data_1, data_2=data_2, 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 m5_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 = 0.2
    context.options['hold_days'] = 5

# 回测引擎：每日数据处理函数，每天执行一次
def m19_handle_data_bigquant_run(context, data):
   

    #-------------大盘风控模块
    # 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，读取深度学习模型预测涨跌的风控数据    
        benckmark_risk=ranker_prediction['result'].values[0]
        # 若预测涨跌的概率小于0.49则不开仓 全仓风控
        if benckmark_risk < 0.49:
            for instrument in positions.keys():
                context.order_target(context.symbol(instrument), 0)
                print(today,'LSTM预测大盘上涨小于0.49,全仓卖出')
                #如果return 在这里   只会卖出第一支持仓的股票，执行一次后返回，有可能起不到全仓风控的作用
                return
    except:
        print('开仓！')    
    
     #-------------大盘风控模块
    
    
    
    #prediction = context.prediction[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.portfolio.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.portfolio.positions.items()}
        instruments = list(reversed(list(ranker_prediction.instrument[ranker_prediction.instrument.apply(
                lambda x: x in equities)])))

        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


m1 = M.instruments.v2(
    start_date='2018-01-01',
    end_date='2020-12-31',
    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 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)
""",
    start_date='',
    end_date='',
    benchmark='000300.HIX',
    drop_na_label=True,
    cast_label_int=True
)

m3 = 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
return_5
return_10
return_20
avg_amount_0/avg_amount_5
avg_amount_5/avg_amount_20
rank_avg_amount_0/rank_avg_amount_5
rank_avg_amount_5/rank_avg_amount_10
rank_return_0
rank_return_5
rank_return_10
rank_return_0/rank_return_5
rank_return_5/rank_return_10
pe_ttm_0
"""
)

m15 = M.general_feature_extractor.v7(
    instruments=m1.data,
    features=m3.data,
    start_date='',
    end_date='',
    before_start_days=90
)

m16 = M.derived_feature_extractor.v3(
    input_data=m15.data,
    features=m3.data,
    date_col='date',
    instrument_col='instrument',
    drop_na=False,
    remove_extra_columns=False
)

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

m13 = M.dropnan.v1(
    input_data=m7.data
)

m6 = M.stock_ranker_train.v5(
    training_ds=m13.data,
    features=m3.data,
    learning_algorithm='排序',
    number_of_leaves=30,
    minimum_docs_per_leaf=1000,
    number_of_trees=20,
    learning_rate=0.1,
    max_bins=1023,
    feature_fraction=1,
    m_lazy_run=False
)

m9 = M.instruments.v2(
    start_date=T.live_run_param('trading_date', '2021-01-01'),
    end_date=T.live_run_param('trading_date', '2021-12-31'),
    market='CN_STOCK_A',
    instrument_list='',
    max_count=0
)

m17 = M.general_feature_extractor.v7(
    instruments=m9.data,
    features=m3.data,
    start_date='',
    end_date='',
    before_start_days=90
)

m18 = M.derived_feature_extractor.v3(
    input_data=m17.data,
    features=m3.data,
    date_col='date',
    instrument_col='instrument',
    drop_na=False,
    remove_extra_columns=False
)

m14 = M.dropnan.v1(
    input_data=m18.data
)

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

m32 = M.dl_layer_input.v1(
    shape='50,5',
    batch_shape='',
    dtype='float32',
    sparse=False,
    name=''
)

m31 = M.dl_layer_reshape.v1(
    inputs=m32.data,
    target_shape='50,5,1',
    name=''
)

m33 = M.dl_layer_conv2d.v1(
    inputs=m31.data,
    filters=32,
    kernel_size='3,5',
    strides='1,1',
    padding='valid',
    data_format='channels_last',
    dilation_rate='1,1',
    activation='relu',
    use_bias=True,
    kernel_initializer='glorot_uniform',
    bias_initializer='Zeros',
    kernel_regularizer='None',
    kernel_regularizer_l1=0,
    kernel_regularizer_l2=0,
    bias_regularizer='None',
    bias_regularizer_l1=0,
    bias_regularizer_l2=0,
    activity_regularizer='None',
    activity_regularizer_l1=0,
    activity_regularizer_l2=0,
    kernel_constraint='None',
    bias_constraint='None',
    name=''
)

m30 = M.dl_layer_reshape.v1(
    inputs=m33.data,
    target_shape='48,32',
    name=''
)

m29 = M.dl_layer_lstm.v1(
    inputs=m30.data,
    units=32,
    activation='tanh',
    recurrent_activation='hard_sigmoid',
    use_bias=True,
    kernel_initializer='glorot_uniform',
    recurrent_initializer='Orthogonal',
    bias_initializer='Ones',
    unit_forget_bias=True,
    kernel_regularizer='None',
    kernel_regularizer_l1=0,
    kernel_regularizer_l2=0,
    recurrent_regularizer='None',
    recurrent_regularizer_l1=0,
    recurrent_regularizer_l2=0,
    bias_regularizer='None',
    bias_regularizer_l1=0,
    bias_regularizer_l2=0,
    activity_regularizer='None',
    activity_regularizer_l1=0,
    activity_regularizer_l2=0,
    kernel_constraint='None',
    recurrent_constraint='None',
    bias_constraint='None',
    dropout=0,
    recurrent_dropout=0,
    return_sequences=False,
    implementation='2',
    name=''
)

m28 = M.dl_layer_dropout.v1(
    inputs=m29.data,
    rate=0.4,
    noise_shape='',
    name=''
)

m27 = M.dl_layer_dense.v1(
    inputs=m28.data,
    units=32,
    activation='tanh',
    use_bias=True,
    kernel_initializer='glorot_uniform',
    bias_initializer='Zeros',
    kernel_regularizer='None',
    kernel_regularizer_l1=0,
    kernel_regularizer_l2=0,
    bias_regularizer='None',
    bias_regularizer_l1=0,
    bias_regularizer_l2=0,
    activity_regularizer='None',
    activity_regularizer_l1=0,
    activity_regularizer_l2=0,
    kernel_constraint='None',
    bias_constraint='None',
    name=''
)

m26 = M.dl_layer_dropout.v1(
    inputs=m27.data,
    rate=0.8,
    noise_shape='',
    name=''
)

m25 = M.dl_layer_dense.v1(
    inputs=m26.data,
    units=1,
    activation='sigmoid',
    use_bias=True,
    kernel_initializer='glorot_uniform',
    bias_initializer='Zeros',
    kernel_regularizer='None',
    kernel_regularizer_l1=0,
    kernel_regularizer_l2=0,
    bias_regularizer='None',
    bias_regularizer_l1=0,
    bias_regularizer_l2=0,
    activity_regularizer='None',
    activity_regularizer_l1=0,
    activity_regularizer_l2=0,
    kernel_constraint='None',
    bias_constraint='None',
    name=''
)

m24 = M.dl_model_init.v1(
    inputs=m32.data,
    outputs=m25.data
)

m38 = M.instruments.v2(
    start_date='2015-01-02',
    end_date='2021-10-30',
    market='CN_STOCK_A',
    instrument_list="""601398.SHA
600028.SHA
601628.SHA
600029.SHA
600048.SHA""",
    max_count=0
)

m37 = M.advanced_auto_labeler.v2(
    instruments=m38.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/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)
where(shift(close, -10) / close -1>0,1,0)

# 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=True,
    user_functions={}
)

m41 = M.input_features.v1(
    features="""(close_0/close_1-1)*10
(high_0/high_1-1)*10
(low_0/low_1-1)*10
(open_0/open_1-1)*10
(volume_0/volume_1-1)*10"""
)

m39 = M.general_feature_extractor.v7(
    instruments=m38.data,
    features=m41.data,
    start_date='',
    end_date='',
    before_start_days=90
)

m40 = M.derived_feature_extractor.v3(
    input_data=m39.data,
    features=m41.data,
    date_col='date',
    instrument_col='instrument',
    drop_na=False,
    remove_extra_columns=False,
    user_functions={}
)

m36 = M.join.v3(
    data1=m37.data,
    data2=m40.data,
    on='date',
    how='inner',
    sort=True
)

m35 = M.dropnan.v1(
    input_data=m36.data
)

m34 = M.dl_convert_to_bin.v2(
    input_data=m35.data,
    features=m41.data,
    window_size=50,
    feature_clip=5,
    flatten=False,
    window_along_col=''
)

m23 = M.dl_model_train.v1(
    input_model=m24.data,
    training_data=m34.data,
    optimizer='Adam',
    loss='binary_crossentropy',
    metrics='accuracy',
    batch_size=2048,
    epochs=10,
    custom_objects=m23_custom_objects_bigquant_run,
    n_gpus=1,
    verbose='1:输出进度条记录'
)

m4 = M.instruments.v2(
    start_date='2021-01-01',
    end_date='2021-12-31',
    market='CN_STOCK_A',
    instrument_list="""601398.SHA
600028.SHA
601628.SHA
600029.SHA
600048.SHA""",
    max_count=0
)

m43 = M.general_feature_extractor.v7(
    instruments=m4.data,
    features=m41.data,
    start_date='',
    end_date='',
    before_start_days=90
)

m44 = M.derived_feature_extractor.v3(
    input_data=m43.data,
    features=m41.data,
    date_col='date',
    instrument_col='instrument',
    drop_na=False,
    remove_extra_columns=False,
    user_functions={}
)

m45 = M.dropnan.v1(
    input_data=m44.data
)

m46 = M.dl_convert_to_bin.v2(
    input_data=m45.data,
    features=m41.data,
    window_size=50,
    feature_clip=5,
    flatten=False,
    window_along_col=''
)

m22 = M.dl_model_predict.v1(
    trained_model=m23.data,
    input_data=m46.data,
    batch_size=10240,
    n_gpus=0,
    verbose='2:每个epoch输出一行记录'
)

m21 = M.cached.v3(
    input_1=m22.data,
    input_2=m46.data,
    input_3=m45.data,
    run=m21_run_bigquant_run,
    post_run=m21_post_run_bigquant_run,
    input_ports='',
    params='{}',
    output_ports=''
)

m5 = M.cached.v3(
    input_1=m21.data_1,
    run=m5_run_bigquant_run,
    post_run=m5_post_run_bigquant_run,
    input_ports='',
    params='{}',
    output_ports=''
)

m10 = M.join.v3(
    data1=m5.data_1,
    data2=m8.predictions,
    on='date',
    how='outer',
    sort=False
)

m11 = M.sort.v4(
    input_ds=m10.data,
    sort_by='date',
    group_by='--',
    keep_columns='--',
    ascending=True
)

m19 = M.trade.v4(
    instruments=m9.data,
    options_data=m11.sorted_data,
    start_date='',
    end_date='',
    initialize=m19_initialize_bigquant_run,
    handle_data=m19_handle_data_bigquant_run,
    prepare=m19_prepare_bigquant_run,
    volume_limit=0.025,
    order_price_field_buy='open',
    order_price_field_sell='close',
    capital_base=1000000,
    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.HIX'
)

[2022-03-20 22:06:49.030565] INFO: moduleinvoker: instruments.v2 开始运行..

[2022-03-20 22:06:49.046370] INFO: moduleinvoker: 命中缓存

[2022-03-20 22:06:49.049074] INFO: moduleinvoker: instruments.v2 运行完成[0.018482s].

[2022-03-20 22:06:49.058187] INFO: moduleinvoker: advanced_auto_labeler.v2 开始运行..

[2022-03-20 22:06:49.071092] INFO: moduleinvoker: 命中缓存

[2022-03-20 22:06:49.073547] INFO: moduleinvoker: advanced_auto_labeler.v2 运行完成[0.015361s].

[2022-03-20 22:06:49.079537] INFO: moduleinvoker: input_features.v1 开始运行..

[2022-03-20 22:06:49.093972] INFO: moduleinvoker: 命中缓存

[2022-03-20 22:06:49.096303] INFO: moduleinvoker: input_features.v1 运行完成[0.016798s].

[2022-03-20 22:06:49.114009] INFO: moduleinvoker: general_feature_extractor.v7 开始运行..

[2022-03-20 22:06:49.122379] INFO: moduleinvoker: 命中缓存

[2022-03-20 22:06:49.123984] INFO: moduleinvoker: general_feature_extractor.v7 运行完成[0.009996s].

[2022-03-20 22:06:49.131179] INFO: moduleinvoker: derived_feature_extractor.v3 开始运行..

[2022-03-20 22:06:49.141133] INFO: moduleinvoker: 命中缓存

[2022-03-20 22:06:49.142728] INFO: moduleinvoker: derived_feature_extractor.v3 运行完成[0.011545s].

[2022-03-20 22:06:49.153112] INFO: moduleinvoker: join.v3 开始运行..

[2022-03-20 22:06:49.162688] INFO: moduleinvoker: 命中缓存

[2022-03-20 22:06:49.165316] INFO: moduleinvoker: join.v3 运行完成[0.012212s].

[2022-03-20 22:06:49.177308] INFO: moduleinvoker: dropnan.v1 开始运行..

[2022-03-20 22:06:49.187542] INFO: moduleinvoker: 命中缓存

[2022-03-20 22:06:49.190510] INFO: moduleinvoker: dropnan.v1 运行完成[0.013166s].

[2022-03-20 22:06:49.202198] INFO: moduleinvoker: stock_ranker_train.v5 开始运行..

[2022-03-20 22:06:49.215686] INFO: moduleinvoker: 命中缓存

[2022-03-20 22:06:49.314883] INFO: moduleinvoker: stock_ranker_train.v5 运行完成[0.112706s].

[2022-03-20 22:06:49.321512] INFO: moduleinvoker: instruments.v2 开始运行..

[2022-03-20 22:06:49.354380] INFO: moduleinvoker: 命中缓存

[2022-03-20 22:06:49.357057] INFO: moduleinvoker: instruments.v2 运行完成[0.035528s].

[2022-03-20 22:06:49.373744] INFO: moduleinvoker: general_feature_extractor.v7 开始运行..

[2022-03-20 22:06:49.381643] INFO: moduleinvoker: 命中缓存

[2022-03-20 22:06:49.383192] INFO: moduleinvoker: general_feature_extractor.v7 运行完成[0.009466s].

[2022-03-20 22:06:49.394120] INFO: moduleinvoker: derived_feature_extractor.v3 开始运行..

[2022-03-20 22:06:49.433689] INFO: moduleinvoker: 命中缓存

[2022-03-20 22:06:49.435379] INFO: moduleinvoker: derived_feature_extractor.v3 运行完成[0.041265s].

[2022-03-20 22:06:49.443832] INFO: moduleinvoker: dropnan.v1 开始运行..

[2022-03-20 22:06:49.452527] INFO: moduleinvoker: 命中缓存

[2022-03-20 22:06:49.454328] INFO: moduleinvoker: dropnan.v1 运行完成[0.010489s].

[2022-03-20 22:06:49.463870] INFO: moduleinvoker: stock_ranker_predict.v5 开始运行..

[2022-03-20 22:06:49.476375] INFO: moduleinvoker: 命中缓存

[2022-03-20 22:06:49.478898] INFO: moduleinvoker: stock_ranker_predict.v5 运行完成[0.015029s].

[2022-03-20 22:06:49.487742] INFO: moduleinvoker: dl_layer_input.v1 运行完成[0.001739s].

[2022-03-20 22:06:49.507170] INFO: moduleinvoker: dl_layer_reshape.v1 运行完成[0.012405s].

[2022-03-20 22:06:49.528657] INFO: moduleinvoker: dl_layer_conv2d.v1 运行完成[0.014574s].

[2022-03-20 22:06:49.545973] INFO: moduleinvoker: dl_layer_reshape.v1 运行完成[0.010311s].

[2022-03-20 22:06:49.698118] INFO: moduleinvoker: dl_layer_lstm.v1 运行完成[0.14374s].

[2022-03-20 22:06:49.709133] INFO: moduleinvoker: dl_layer_dropout.v1 运行完成[0.003721s].

[2022-03-20 22:06:49.732337] INFO: moduleinvoker: dl_layer_dense.v1 运行完成[0.015952s].

[2022-03-20 22:06:49.743754] INFO: moduleinvoker: dl_layer_dropout.v1 运行完成[0.003474s].

[2022-03-20 22:06:49.759815] INFO: moduleinvoker: dl_layer_dense.v1 运行完成[0.009778s].

[2022-03-20 22:06:49.797190] INFO: moduleinvoker: cached.v3 开始运行..

[2022-03-20 22:06:49.848436] INFO: moduleinvoker: cached.v3 运行完成[0.051249s].

[2022-03-20 22:06:49.850826] INFO: moduleinvoker: dl_model_init.v1 运行完成[0.086046s].

[2022-03-20 22:06:49.857993] INFO: moduleinvoker: instruments.v2 开始运行..

[2022-03-20 22:06:49.866646] INFO: moduleinvoker: 命中缓存

[2022-03-20 22:06:49.868807] INFO: moduleinvoker: instruments.v2 运行完成[0.01082s].

[2022-03-20 22:06:49.880649] INFO: moduleinvoker: advanced_auto_labeler.v2 开始运行..

[2022-03-20 22:06:49.888056] INFO: moduleinvoker: 命中缓存

[2022-03-20 22:06:49.889865] INFO: moduleinvoker: advanced_auto_labeler.v2 运行完成[0.009231s].

[2022-03-20 22:06:49.894675] INFO: moduleinvoker: input_features.v1 开始运行..

[2022-03-20 22:06:49.902103] INFO: moduleinvoker: 命中缓存

[2022-03-20 22:06:49.903742] INFO: moduleinvoker: input_features.v1 运行完成[0.009073s].

[2022-03-20 22:06:49.923883] INFO: moduleinvoker: general_feature_extractor.v7 开始运行..

[2022-03-20 22:06:49.950528] INFO: moduleinvoker: 命中缓存

[2022-03-20 22:06:49.954869] INFO: moduleinvoker: general_feature_extractor.v7 运行完成[0.030991s].

[2022-03-20 22:06:49.969814] INFO: moduleinvoker: derived_feature_extractor.v3 开始运行..

[2022-03-20 22:06:49.978653] INFO: moduleinvoker: 命中缓存

[2022-03-20 22:06:49.980226] INFO: moduleinvoker: derived_feature_extractor.v3 运行完成[0.010481s].

[2022-03-20 22:06:49.993150] INFO: moduleinvoker: join.v3 开始运行..

[2022-03-20 22:06:50.009768] INFO: moduleinvoker: 命中缓存

[2022-03-20 22:06:50.011872] INFO: moduleinvoker: join.v3 运行完成[0.018726s].

[2022-03-20 22:06:50.021346] INFO: moduleinvoker: dropnan.v1 开始运行..

[2022-03-20 22:06:50.045025] INFO: moduleinvoker: 命中缓存

[2022-03-20 22:06:50.047322] INFO: moduleinvoker: dropnan.v1 运行完成[0.025963s].

[2022-03-20 22:06:50.067354] INFO: moduleinvoker: dl_convert_to_bin.v2 开始运行..

[2022-03-20 22:06:50.082404] INFO: moduleinvoker: 命中缓存

[2022-03-20 22:06:50.085036] INFO: moduleinvoker: dl_convert_to_bin.v2 运行完成[0.0177s].

[2022-03-20 22:06:50.091660] INFO: moduleinvoker: dl_model_train.v1 开始运行..

[2022-03-20 22:06:50.476770] INFO: dl_model_train: 准备训练，训练样本个数：41400，迭代次数：10

[2022-03-20 22:07:42.169856] INFO: dl_model_train: 训练结束，耗时：51.69s

[2022-03-20 22:07:42.212801] INFO: moduleinvoker: dl_model_train.v1 运行完成[52.121171s].

[2022-03-20 22:07:42.221037] INFO: moduleinvoker: instruments.v2 开始运行..

[2022-03-20 22:07:42.234095] INFO: moduleinvoker: 命中缓存

[2022-03-20 22:07:42.235868] INFO: moduleinvoker: instruments.v2 运行完成[0.014841s].

[2022-03-20 22:07:42.253268] INFO: moduleinvoker: general_feature_extractor.v7 开始运行..

[2022-03-20 22:07:42.262669] INFO: moduleinvoker: 命中缓存

[2022-03-20 22:07:42.265305] INFO: moduleinvoker: general_feature_extractor.v7 运行完成[0.012059s].

[2022-03-20 22:07:42.276166] INFO: moduleinvoker: derived_feature_extractor.v3 开始运行..

[2022-03-20 22:07:42.283779] INFO: moduleinvoker: 命中缓存

[2022-03-20 22:07:42.286414] INFO: moduleinvoker: derived_feature_extractor.v3 运行完成[0.010236s].

[2022-03-20 22:07:42.298674] INFO: moduleinvoker: dropnan.v1 开始运行..

[2022-03-20 22:07:42.311116] INFO: moduleinvoker: 命中缓存

[2022-03-20 22:07:42.313809] INFO: moduleinvoker: dropnan.v1 运行完成[0.015137s].

[2022-03-20 22:07:42.338533] INFO: moduleinvoker: dl_convert_to_bin.v2 开始运行..

[2022-03-20 22:07:42.348076] INFO: moduleinvoker: 命中缓存

[2022-03-20 22:07:42.350638] INFO: moduleinvoker: dl_convert_to_bin.v2 运行完成[0.012119s].

[2022-03-20 22:07:42.356807] INFO: moduleinvoker: dl_model_predict.v1 开始运行..

[2022-03-20 22:07:43.265684] INFO: moduleinvoker: dl_model_predict.v1 运行完成[0.908882s].

[2022-03-20 22:07:43.293872] INFO: moduleinvoker: cached.v3 开始运行..

[2022-03-20 22:07:43.514284] INFO: moduleinvoker: cached.v3 运行完成[0.220418s].

[2022-03-20 22:07:43.529654] INFO: moduleinvoker: cached.v3 开始运行..

[2022-03-20 22:07:43.664701] INFO: moduleinvoker: cached.v3 运行完成[0.13503s].

[2022-03-20 22:07:43.681966] INFO: moduleinvoker: join.v3 开始运行..

[2022-03-20 22:07:46.172356] INFO: join: 行数=6447515/1515/1289503, 耗时=1.123823s

[2022-03-20 22:07:46.195690] INFO: join: 最终行数: 6447515

[2022-03-20 22:07:46.205850] INFO: moduleinvoker: join.v3 运行完成[2.523871s].

[2022-03-20 22:07:46.221374] INFO: moduleinvoker: sort.v4 开始运行..

[2022-03-20 22:07:53.594831] INFO: moduleinvoker: sort.v4 运行完成[7.373452s].

[2022-03-20 22:07:53.823113] INFO: moduleinvoker: backtest.v8 开始运行..

[2022-03-20 22:07:53.833677] INFO: backtest: biglearning backtest:V8.6.2

[2022-03-20 22:07:53.835137] INFO: backtest: product_type:stock by specified

[2022-03-20 22:07:54.037603] INFO: moduleinvoker: cached.v2 开始运行..

[2022-03-20 22:07:54.048403] INFO: moduleinvoker: 命中缓存

[2022-03-20 22:07:54.051676] INFO: moduleinvoker: cached.v2 运行完成[0.014115s].

[2022-03-20 22:07:56.605117] INFO: algo: TradingAlgorithm V1.8.7

[2022-03-20 22:07:58.554314] INFO: algo: trading transform...

[2022-03-20 22:08:04.959564] INFO: algo: handle_splits get splits [dt:2021-04-20 00:00:00+00:00] [asset:Equity(5603 [603385.SHA]), ratio:0.9798320531845093]

[2022-03-20 22:08:04.961984] INFO: Position: position stock handle split[sid:5603, orig_amount:3100, new_amount:3163.0, orig_cost:11.800004859210205, new_cost:11.562, ratio:0.9798320531845093, last_sale_price:11.660000801086426]

[2022-03-20 22:08:04.963967] INFO: Position: after split: PositionStock(asset:Equity(5603 [603385.SHA]), amount:3163.0, cost_basis:11.562, last_sale_price:11.899999618530273)

[2022-03-20 22:08:04.965560] INFO: Position: returning cash: 9.4155

[2022-03-20 22:08:06.836920] INFO: algo: handle_splits get splits [dt:2021-05-28 00:00:00+00:00] [asset:Equity(514 [300462.SZA]), ratio:0.9876540899276733]

[2022-03-20 22:08:06.838567] INFO: algo: handle_splits get splits [dt:2021-05-28 00:00:00+00:00] [asset:Equity(3749 [603303.SHA]), ratio:0.9806948900222778]

[2022-03-20 22:08:06.839871] INFO: algo: handle_splits get splits [dt:2021-05-28 00:00:00+00:00] [asset:Equity(4678 [300552.SZA]), ratio:0.981345534324646]

[2022-03-20 22:08:06.841108] INFO: Position: position stock handle split[sid:514, orig_amount:7300, new_amount:7391.0, orig_cost:11.510006453083202, new_cost:11.3679, ratio:0.9876540899276733, last_sale_price:11.999998092651367]

[2022-03-20 22:08:06.842252] INFO: Position: after split: PositionStock(asset:Equity(514 [300462.SZA]), amount:7391.0, cost_basis:11.3679, last_sale_price:12.15000057220459)

[2022-03-20 22:08:06.843256] INFO: Position: returning cash: 3.0207

[2022-03-20 22:08:06.844339] INFO: Position: position stock handle split[sid:3749, orig_amount:2800, new_amount:2855.0, orig_cost:13.060001857270144, new_cost:12.8079, ratio:0.9806948900222778, last_sale_price:12.69999885559082]

[2022-03-20 22:08:06.845838] INFO: Position: after split: PositionStock(asset:Equity(3749 [603303.SHA]), amount:2855.0, cost_basis:12.8079, last_sale_price:12.949999809265137)

[2022-03-20 22:08:06.847515] INFO: Position: returning cash: 1.5034

[2022-03-20 22:08:06.849450] INFO: Position: position stock handle split[sid:4678, orig_amount:3800, new_amount:3872.0, orig_cost:32.11000715114687, new_cost:31.511, ratio:0.981345534324646, last_sale_price:32.09000015258789]

[2022-03-20 22:08:06.851400] INFO: Position: after split: PositionStock(asset:Equity(4678 [300552.SZA]), amount:3872.0, cost_basis:31.511, last_sale_price:32.70000076293945)

[2022-03-20 22:08:06.852865] INFO: Position: returning cash: 7.524

[2022-03-20 22:08:07.320042] INFO: algo: handle_splits get splits [dt:2021-06-08 00:00:00+00:00] [asset:Equity(4190 [002730.SZA]), ratio:0.9940072894096375]

[2022-03-20 22:08:07.321577] INFO: algo: handle_splits get splits [dt:2021-06-08 00:00:00+00:00] [asset:Equity(1377 [688529.SHA]), ratio:0.9925429821014404]

[2022-03-20 22:08:07.322834] INFO: Position: position stock handle split[sid:1377, orig_amount:4500, new_amount:4533.0, orig_cost:24.40003055987955, new_cost:24.2181, ratio:0.9925429821014404, last_sale_price:25.289995193481445]

[2022-03-20 22:08:07.323834] INFO: Position: after split: PositionStock(asset:Equity(1377 [688529.SHA]), amount:4533.0, cost_basis:24.2181, last_sale_price:25.479999542236328)

[2022-03-20 22:08:07.324751] INFO: Position: returning cash: 20.4518

[2022-03-20 22:08:07.509969] INFO: algo: handle_splits get splits [dt:2021-06-11 00:00:00+00:00] [asset:Equity(3409 [300523.SZA]), ratio:0.9967781901359558]

[2022-03-20 22:08:07.512018] INFO: algo: handle_splits get splits [dt:2021-06-11 00:00:00+00:00] [asset:Equity(3955 [600798.SHA]), ratio:0.9893805980682373]

[2022-03-20 22:08:07.514116] INFO: Position: position stock handle split[sid:3409, orig_amount:2600, new_amount:2608.0, orig_cost:23.99000310052421, new_cost:23.9127, ratio:0.9967781901359558, last_sale_price:24.750001907348633]

[2022-03-20 22:08:07.516012] INFO: Position: after split: PositionStock(asset:Equity(3409 [300523.SZA]), amount:2608.0, cost_basis:23.9127, last_sale_price:24.829999923706055)

[2022-03-20 22:08:07.517783] INFO: Position: returning cash: 9.9936

[2022-03-20 22:08:07.519452] INFO: Position: position stock handle split[sid:3955, orig_amount:6200, new_amount:6266.0, orig_cost:5.319999717431042, new_cost:5.2635, ratio:0.9893805980682373, last_sale_price:5.590000629425049]

[2022-03-20 22:08:07.520704] INFO: Position: after split: PositionStock(asset:Equity(3955 [600798.SHA]), amount:6266.0, cost_basis:5.2635, last_sale_price:5.650000095367432)

[2022-03-20 22:08:07.521877] INFO: Position: returning cash: 3.0576

[2022-03-20 22:08:08.031062] INFO: algo: handle_splits get splits [dt:2021-06-22 00:00:00+00:00] [asset:Equity(4637 [002698.SZA]), ratio:0.9844356775283813]

[2022-03-20 22:08:08.032670] INFO: Position: position stock handle split[sid:4637, orig_amount:3800, new_amount:3860.0, orig_cost:12.640003140518548, new_cost:12.4433, ratio:0.9844356775283813, last_sale_price:12.649998664855957]

[2022-03-20 22:08:08.033901] INFO: Position: after split: PositionStock(asset:Equity(4637 [002698.SZA]), amount:3860.0, cost_basis:12.4433, last_sale_price:12.850000381469727)

[2022-03-20 22:08:08.035084] INFO: Position: returning cash: 1.006

[2022-03-20 22:08:08.342474] INFO: algo: handle_splits get splits [dt:2021-06-28 00:00:00+00:00] [asset:Equity(4589 [600353.SHA]), ratio:0.9946524500846863]

[2022-03-20 22:08:08.343934] INFO: Position: position stock handle split[sid:4589, orig_amount:20100, new_amount:20208.0, orig_cost:5.320001013778358, new_cost:5.2916, ratio:0.9946524500846863, last_sale_price:5.580000400543213]

[2022-03-20 22:08:08.345471] INFO: Position: after split: PositionStock(asset:Equity(4589 [600353.SHA]), amount:20208.0, cost_basis:5.2916, last_sale_price:5.610000133514404)

[2022-03-20 22:08:08.347274] INFO: Position: returning cash: 0.355

[2022-03-20 22:08:09.043901] INFO: algo: handle_splits get splits [dt:2021-07-12 00:00:00+00:00] [asset:Equity(3307 [600789.SHA]), ratio:0.9931128621101379]

[2022-03-20 22:08:09.135412] INFO: algo: handle_splits get splits [dt:2021-07-13 00:00:00+00:00] [asset:Equity(4356 [600148.SHA]), ratio:0.9861111044883728]

[2022-03-20 22:08:09.136955] INFO: algo: handle_splits get splits [dt:2021-07-13 00:00:00+00:00] [asset:Equity(170 [002835.SZA]), ratio:0.9920635223388672]

[2022-03-20 22:08:09.138223] INFO: Position: position stock handle split[sid:4356, orig_amount:4300, new_amount:4360.0, orig_cost:12.990021075091633, new_cost:12.8096, ratio:0.9861111044883728, last_sale_price:12.779999732971191]

[2022-03-20 22:08:09.139302] INFO: Position: after split: PositionStock(asset:Equity(4356 [600148.SHA]), amount:4360.0, cost_basis:12.8096, last_sale_price:12.960000038146973)

[2022-03-20 22:08:09.140302] INFO: Position: returning cash: 7.2004

[2022-03-20 22:08:09.141345] INFO: Position: position stock handle split[sid:170, orig_amount:5700, new_amount:5745.0, orig_cost:12.180000599411708, new_cost:12.0833, ratio:0.9920635223388672, last_sale_price:12.499999046325684]

[2022-03-20 22:08:09.142375] INFO: Position: after split: PositionStock(asset:Equity(170 [002835.SZA]), amount:5745.0, cost_basis:12.0833, last_sale_price:12.599998474121094)

[2022-03-20 22:08:09.143362] INFO: Position: returning cash: 7.4978

[2022-03-20 22:08:09.378330] INFO: algo: handle_splits get splits [dt:2021-07-16 00:00:00+00:00] [asset:Equity(953 [600900.SHA]), ratio:0.9654833078384399]

[2022-03-20 22:08:11.331594] INFO: algo: handle_splits get splits [dt:2021-08-27 00:00:00+00:00] [asset:Equity(4154 [300560.SZA]), ratio:0.9884614944458008]

[2022-03-20 22:08:11.333127] INFO: Position: position stock handle split[sid:4154, orig_amount:2500, new_amount:2529.0, orig_cost:15.350000115309493, new_cost:15.1729, ratio:0.9884614944458008, last_sale_price:15.419999122619629]

[2022-03-20 22:08:11.334202] INFO: Position: after split: PositionStock(asset:Equity(4154 [300560.SZA]), amount:2529.0, cost_basis:15.1729, last_sale_price:15.59999942779541)

[2022-03-20 22:08:11.335169] INFO: Position: returning cash: 2.8217

[2022-03-20 22:08:17.012333] INFO: Performance: Simulated 243 trading days out of 243.

[2022-03-20 22:08:17.014081] INFO: Performance: first open: 2021-01-04 09:30:00+00:00

[2022-03-20 22:08:17.015467] INFO: Performance: last close: 2021-12-31 15:00:00+00:00

[2022-03-20 22:08:32.243415] INFO: moduleinvoker: backtest.v8 运行完成[38.420309s].

[2022-03-20 22:08:32.245472] INFO: moduleinvoker: trade.v4 运行完成[38.638944s].

设置测试数据集，查看训练迭代过程的NDCG

bigcharts-data-start/{"__type":"tabs","__id":"bigchart-f9c98cb495a340e79c5234756333805c"}/bigcharts-data-end

Epoch 1/10
 1/21 [>.............................] - ETA: 1:36 - loss: 1.0283 - accuracy: 0.4814 2/21 [=>............................] - ETA: 4s - loss: 1.0122 - accuracy: 0.4866   3/21 [===>..........................] - ETA: 4s - loss: 1.0037 - accuracy: 0.4899 4/21 [====>.........................] - ETA: 4s - loss: 1.0009 - accuracy: 0.4910 5/21 [======>.......................] - ETA: 3s - loss: 0.9973 - accuracy: 0.4919 6/21 [=======>......................] - ETA: 3s - loss: 0.9942 - accuracy: 0.4930 7/21 [=========>....................] - ETA: 3s - loss: 0.9909 - accuracy: 0.4937 8/21 [==========>...................] - ETA: 3s - loss: 0.9879 - accuracy: 0.4942 9/21 [===========>..................] - ETA: 2s - loss: 0.9849 - accuracy: 0.494810/21 [=============>................] - ETA: 2s - loss: 0.9821 - accuracy: 0.495011/21 [==============>...............] - ETA: 2s - loss: 0.9795 - accuracy: 0.495312/21 [================>.............] - ETA: 2s - loss: 0.9768 - accuracy: 0.495613/21 [=================>............] - ETA: 1s - loss: 0.9742 - accuracy: 0.496014/21 [===================>..........] - ETA: 1s - loss: 0.9715 - accuracy: 0.496315/21 [====================>.........] - ETA: 1s - loss: 0.9689 - accuracy: 0.496616/21 [=====================>........] - ETA: 1s - loss: 0.9664 - accuracy: 0.496917/21 [=======================>......] - ETA: 0s - loss: 0.9639 - accuracy: 0.497218/21 [========================>.....] - ETA: 0s - loss: 0.9616 - accuracy: 0.497519/21 [==========================>...] - ETA: 0s - loss: 0.9593 - accuracy: 0.497720/21 [===========================>..] - ETA: 0s - loss: 0.9571 - accuracy: 0.497821/21 [==============================] - ETA: 0s - loss: 0.9551 - accuracy: 0.498021/21 [==============================] - 9s 222ms/step - loss: 0.9532 - accuracy: 0.4981
Epoch 2/10
 1/21 [>.............................] - ETA: 4s - loss: 0.8426 - accuracy: 0.4897 2/21 [=>............................] - ETA: 4s - loss: 0.8351 - accuracy: 0.4960 3/21 [===>..........................] - ETA: 3s - loss: 0.8319 - accuracy: 0.4977 4/21 [====>.........................] - ETA: 3s - loss: 0.8300 - accuracy: 0.4984 5/21 [======>.......................] - ETA: 3s - loss: 0.8290 - accuracy: 0.4982 6/21 [=======>......................] - ETA: 3s - loss: 0.8276 - accuracy: 0.4986 7/21 [=========>....................] - ETA: 3s - loss: 0.8262 - accuracy: 0.4989 8/21 [==========>...................] - ETA: 2s - loss: 0.8251 - accuracy: 0.4990 9/21 [===========>..................] - ETA: 2s - loss: 0.8240 - accuracy: 0.499210/21 [=============>................] - ETA: 2s - loss: 0.8227 - accuracy: 0.499611/21 [==============>...............] - ETA: 2s - loss: 0.8216 - accuracy: 0.499712/21 [================>.............] - ETA: 1s - loss: 0.8204 - accuracy: 0.499913/21 [=================>............] - ETA: 1s - loss: 0.8192 - accuracy: 0.499914/21 [===================>..........] - ETA: 1s - loss: 0.8182 - accuracy: 0.500015/21 [====================>.........] - ETA: 1s - loss: 0.8171 - accuracy: 0.500216/21 [=====================>........] - ETA: 1s - loss: 0.8160 - accuracy: 0.500417/21 [=======================>......] - ETA: 0s - loss: 0.8149 - accuracy: 0.500618/21 [========================>.....] - ETA: 0s - loss: 0.8139 - accuracy: 0.500719/21 [==========================>...] - ETA: 0s - loss: 0.8129 - accuracy: 0.500820/21 [===========================>..] - ETA: 0s - loss: 0.8120 - accuracy: 0.500921/21 [==============================] - ETA: 0s - loss: 0.8111 - accuracy: 0.501021/21 [==============================] - 5s 217ms/step - loss: 0.8103 - accuracy: 0.5011
Epoch 3/10
 1/21 [>.............................] - ETA: 4s - loss: 0.7419 - accuracy: 0.5190 2/21 [=>............................] - ETA: 3s - loss: 0.7466 - accuracy: 0.5135 3/21 [===>..........................] - ETA: 3s - loss: 0.7502 - accuracy: 0.5109 4/21 [====>.........................] - ETA: 3s - loss: 0.7525 - accuracy: 0.5085 5/21 [======>.......................] - ETA: 3s - loss: 0.7533 - accuracy: 0.5075 6/21 [=======>......................] - ETA: 3s - loss: 0.7534 - accuracy: 0.5072 7/21 [=========>....................] - ETA: 3s - loss: 0.7536 - accuracy: 0.5066 8/21 [==========>...................] - ETA: 2s - loss: 0.7537 - accuracy: 0.5062 9/21 [===========>..................] - ETA: 2s - loss: 0.7537 - accuracy: 0.506010/21 [=============>................] - ETA: 2s - loss: 0.7536 - accuracy: 0.505811/21 [==============>...............] - ETA: 2s - loss: 0.7534 - accuracy: 0.505712/21 [================>.............] - ETA: 1s - loss: 0.7531 - accuracy: 0.505513/21 [=================>............] - ETA: 1s - loss: 0.7529 - accuracy: 0.505314/21 [===================>..........] - ETA: 1s - loss: 0.7527 - accuracy: 0.505115/21 [====================>.........] - ETA: 1s - loss: 0.7525 - accuracy: 0.504916/21 [=====================>........] - ETA: 1s - loss: 0.7522 - accuracy: 0.504817/21 [=======================>......] - ETA: 0s - loss: 0.7520 - accuracy: 0.504618/21 [========================>.....] - ETA: 0s - loss: 0.7518 - accuracy: 0.504419/21 [==========================>...] - ETA: 0s - loss: 0.7517 - accuracy: 0.504220/21 [===========================>..] - ETA: 0s - loss: 0.7515 - accuracy: 0.504021/21 [==============================] - ETA: 0s - loss: 0.7513 - accuracy: 0.503821/21 [==============================] - 4s 213ms/step - loss: 0.7511 - accuracy: 0.5037
Epoch 4/10
 1/21 [>.............................] - ETA: 4s - loss: 0.7369 - accuracy: 0.4873 2/21 [=>............................] - ETA: 4s - loss: 0.7335 - accuracy: 0.4929 3/21 [===>..........................] - ETA: 3s - loss: 0.7330 - accuracy: 0.4942 4/21 [====>.........................] - ETA: 3s - loss: 0.7322 - accuracy: 0.4957 5/21 [======>.......................] - ETA: 3s - loss: 0.7317 - accuracy: 0.4970 6/21 [=======>......................] - ETA: 3s - loss: 0.7314 - accuracy: 0.4974 7/21 [=========>....................] - ETA: 3s - loss: 0.7311 - accuracy: 0.4976 8/21 [==========>...................] - ETA: 2s - loss: 0.7308 - accuracy: 0.4977 9/21 [===========>..................] - ETA: 2s - loss: 0.7305 - accuracy: 0.498010/21 [=============>................] - ETA: 2s - loss: 0.7302 - accuracy: 0.498211/21 [==============>...............] - ETA: 2s - loss: 0.7300 - accuracy: 0.498412/21 [================>.............] - ETA: 2s - loss: 0.7297 - accuracy: 0.498713/21 [=================>............] - ETA: 1s - loss: 0.7295 - accuracy: 0.499014/21 [===================>..........] - ETA: 1s - loss: 0.7292 - accuracy: 0.499315/21 [====================>.........] - ETA: 1s - loss: 0.7289 - accuracy: 0.499716/21 [=====================>........] - ETA: 1s - loss: 0.7286 - accuracy: 0.500017/21 [=======================>......] - ETA: 0s - loss: 0.7284 - accuracy: 0.500418/21 [========================>.....] - ETA: 0s - loss: 0.7281 - accuracy: 0.500719/21 [==========================>...] - ETA: 0s - loss: 0.7279 - accuracy: 0.501020/21 [===========================>..] - ETA: 0s - loss: 0.7276 - accuracy: 0.501221/21 [==============================] - ETA: 0s - loss: 0.7274 - accuracy: 0.501421/21 [==============================] - 5s 215ms/step - loss: 0.7272 - accuracy: 0.5015
Epoch 5/10
 1/21 [>.............................] - ETA: 4s - loss: 0.7176 - accuracy: 0.5024 2/21 [=>............................] - ETA: 4s - loss: 0.7172 - accuracy: 0.5026 3/21 [===>..........................] - ETA: 4s - loss: 0.7166 - accuracy: 0.5018 4/21 [====>.........................] - ETA: 3s - loss: 0.7163 - accuracy: 0.5023 5/21 [======>.......................] - ETA: 3s - loss: 0.7163 - accuracy: 0.5019 6/21 [=======>......................] - ETA: 3s - loss: 0.7164 - accuracy: 0.5021 7/21 [=========>....................] - ETA: 3s - loss: 0.7164 - accuracy: 0.5022 8/21 [==========>...................] - ETA: 2s - loss: 0.7164 - accuracy: 0.5025 9/21 [===========>..................] - ETA: 2s - loss: 0.7164 - accuracy: 0.502610/21 [=============>................] - ETA: 2s - loss: 0.7163 - accuracy: 0.502511/21 [==============>...............] - ETA: 2s - loss: 0.7162 - accuracy: 0.502412/21 [================>.............] - ETA: 1s - loss: 0.7161 - accuracy: 0.502213/21 [=================>............] - ETA: 1s - loss: 0.7161 - accuracy: 0.502014/21 [===================>..........] - ETA: 1s - loss: 0.7159 - accuracy: 0.502015/21 [====================>.........] - ETA: 1s - loss: 0.7157 - accuracy: 0.502016/21 [=====================>........] - ETA: 1s - loss: 0.7156 - accuracy: 0.502117/21 [=======================>......] - ETA: 0s - loss: 0.7155 - accuracy: 0.502118/21 [========================>.....] - ETA: 0s - loss: 0.7154 - accuracy: 0.502119/21 [==========================>...] - ETA: 0s - loss: 0.7153 - accuracy: 0.502120/21 [===========================>..] - ETA: 0s - loss: 0.7152 - accuracy: 0.502221/21 [==============================] - ETA: 0s - loss: 0.7151 - accuracy: 0.502221/21 [==============================] - 4s 211ms/step - loss: 0.7150 - accuracy: 0.5023
Epoch 6/10
 1/21 [>.............................] - ETA: 4s - loss: 0.7061 - accuracy: 0.5142 2/21 [=>............................] - ETA: 4s - loss: 0.7095 - accuracy: 0.5066 3/21 [===>..........................] - ETA: 4s - loss: 0.7103 - accuracy: 0.5050 4/21 [====>.........................] - ETA: 3s - loss: 0.7104 - accuracy: 0.5047 5/21 [======>.......................] - ETA: 3s - loss: 0.7105 - accuracy: 0.5040 6/21 [=======>......................] - ETA: 3s - loss: 0.7106 - accuracy: 0.5034 7/21 [=========>....................] - ETA: 3s - loss: 0.7107 - accuracy: 0.5026 8/21 [==========>...................] - ETA: 3s - loss: 0.7108 - accuracy: 0.5019 9/21 [===========>..................] - ETA: 2s - loss: 0.7109 - accuracy: 0.501310/21 [=============>................] - ETA: 2s - loss: 0.7109 - accuracy: 0.501011/21 [==============>...............] - ETA: 2s - loss: 0.7108 - accuracy: 0.500912/21 [================>.............] - ETA: 2s - loss: 0.7108 - accuracy: 0.500813/21 [=================>............] - ETA: 1s - loss: 0.7107 - accuracy: 0.500614/21 [===================>..........] - ETA: 1s - loss: 0.7107 - accuracy: 0.500615/21 [====================>.........] - ETA: 1s - loss: 0.7106 - accuracy: 0.500616/21 [=====================>........] - ETA: 1s - loss: 0.7105 - accuracy: 0.500617/21 [=======================>......] - ETA: 0s - loss: 0.7104 - accuracy: 0.500518/21 [========================>.....] - ETA: 0s - loss: 0.7103 - accuracy: 0.500519/21 [==========================>...] - ETA: 0s - loss: 0.7102 - accuracy: 0.500620/21 [===========================>..] - ETA: 0s - loss: 0.7101 - accuracy: 0.500621/21 [==============================] - ETA: 0s - loss: 0.7101 - accuracy: 0.500721/21 [==============================] - 5s 222ms/step - loss: 0.7100 - accuracy: 0.5007
Epoch 7/10
 1/21 [>.............................] - ETA: 4s - loss: 0.7048 - accuracy: 0.5039 2/21 [=>............................] - ETA: 4s - loss: 0.7051 - accuracy: 0.5037 3/21 [===>..........................] - ETA: 4s - loss: 0.7056 - accuracy: 0.5036 4/21 [====>.........................] - ETA: 3s - loss: 0.7057 - accuracy: 0.5028 5/21 [======>.......................] - ETA: 3s - loss: 0.7058 - accuracy: 0.5027 6/21 [=======>......................] - ETA: 3s - loss: 0.7059 - accuracy: 0.5028 7/21 [=========>....................] - ETA: 3s - loss: 0.7059 - accuracy: 0.5028 8/21 [==========>...................] - ETA: 3s - loss: 0.7059 - accuracy: 0.5028 9/21 [===========>..................] - ETA: 2s - loss: 0.7058 - accuracy: 0.503110/21 [=============>................] - ETA: 2s - loss: 0.7057 - accuracy: 0.503211/21 [==============>...............] - ETA: 2s - loss: 0.7056 - accuracy: 0.503312/21 [================>.............] - ETA: 2s - loss: 0.7056 - accuracy: 0.503213/21 [=================>............] - ETA: 1s - loss: 0.7055 - accuracy: 0.503114/21 [===================>..........] - ETA: 1s - loss: 0.7054 - accuracy: 0.503115/21 [====================>.........] - ETA: 1s - loss: 0.7053 - accuracy: 0.503116/21 [=====================>........] - ETA: 1s - loss: 0.7052 - accuracy: 0.503017/21 [=======================>......] - ETA: 0s - loss: 0.7051 - accuracy: 0.503118/21 [========================>.....] - ETA: 0s - loss: 0.7050 - accuracy: 0.503119/21 [==========================>...] - ETA: 0s - loss: 0.7049 - accuracy: 0.503020/21 [===========================>..] - ETA: 0s - loss: 0.7048 - accuracy: 0.503021/21 [==============================] - ETA: 0s - loss: 0.7048 - accuracy: 0.503021/21 [==============================] - 5s 227ms/step - loss: 0.7047 - accuracy: 0.5029
Epoch 8/10
 1/21 [>.............................] - ETA: 4s - loss: 0.7018 - accuracy: 0.5186 2/21 [=>............................] - ETA: 4s - loss: 0.7027 - accuracy: 0.5146 3/21 [===>..........................] - ETA: 4s - loss: 0.7025 - accuracy: 0.5141 4/21 [====>.........................] - ETA: 3s - loss: 0.7022 - accuracy: 0.5142 5/21 [======>.......................] - ETA: 3s - loss: 0.7022 - accuracy: 0.5142 6/21 [=======>......................] - ETA: 3s - loss: 0.7023 - accuracy: 0.5137 7/21 [=========>....................] - ETA: 3s - loss: 0.7024 - accuracy: 0.5131 8/21 [==========>...................] - ETA: 3s - loss: 0.7024 - accuracy: 0.5123 9/21 [===========>..................] - ETA: 2s - loss: 0.7024 - accuracy: 0.511710/21 [=============>................] - ETA: 2s - loss: 0.7024 - accuracy: 0.511311/21 [==============>...............] - ETA: 2s - loss: 0.7024 - accuracy: 0.511012/21 [================>.............] - ETA: 2s - loss: 0.7024 - accuracy: 0.510613/21 [=================>............] - ETA: 1s - loss: 0.7024 - accuracy: 0.510214/21 [===================>..........] - ETA: 1s - loss: 0.7024 - accuracy: 0.509915/21 [====================>.........] - ETA: 1s - loss: 0.7024 - accuracy: 0.509716/21 [=====================>........] - ETA: 1s - loss: 0.7024 - accuracy: 0.509417/21 [=======================>......] - ETA: 0s - loss: 0.7024 - accuracy: 0.509118/21 [========================>.....] - ETA: 0s - loss: 0.7024 - accuracy: 0.508919/21 [==========================>...] - ETA: 0s - loss: 0.7025 - accuracy: 0.508720/21 [===========================>..] - ETA: 0s - loss: 0.7025 - accuracy: 0.508421/21 [==============================] - ETA: 0s - loss: 0.7025 - accuracy: 0.508221/21 [==============================] - 5s 230ms/step - loss: 0.7025 - accuracy: 0.5080
Epoch 9/10
 1/21 [>.............................] - ETA: 4s - loss: 0.7012 - accuracy: 0.5000 2/21 [=>............................] - ETA: 4s - loss: 0.7015 - accuracy: 0.5006 3/21 [===>..........................] - ETA: 4s - loss: 0.7017 - accuracy: 0.5005 4/21 [====>.........................] - ETA: 4s - loss: 0.7015 - accuracy: 0.5013 5/21 [======>.......................] - ETA: 3s - loss: 0.7013 - accuracy: 0.5021 6/21 [=======>......................] - ETA: 3s - loss: 0.7011 - accuracy: 0.5030 7/21 [=========>....................] - ETA: 3s - loss: 0.7010 - accuracy: 0.5034 8/21 [==========>...................] - ETA: 3s - loss: 0.7009 - accuracy: 0.5035 9/21 [===========>..................] - ETA: 3s - loss: 0.7009 - accuracy: 0.503410/21 [=============>................] - ETA: 2s - loss: 0.7009 - accuracy: 0.503311/21 [==============>...............] - ETA: 2s - loss: 0.7009 - accuracy: 0.503012/21 [================>.............] - ETA: 2s - loss: 0.7009 - accuracy: 0.502813/21 [=================>............] - ETA: 2s - loss: 0.7008 - accuracy: 0.502814/21 [===================>..........] - ETA: 1s - loss: 0.7008 - accuracy: 0.502715/21 [====================>.........] - ETA: 1s - loss: 0.7008 - accuracy: 0.502616/21 [=====================>........] - ETA: 1s - loss: 0.7007 - accuracy: 0.502517/21 [=======================>......] - ETA: 0s - loss: 0.7007 - accuracy: 0.502518/21 [========================>.....] - ETA: 0s - loss: 0.7007 - accuracy: 0.502419/21 [==========================>...] - ETA: 0s - loss: 0.7007 - accuracy: 0.502320/21 [===========================>..] - ETA: 0s - loss: 0.7006 - accuracy: 0.502321/21 [==============================] - ETA: 0s - loss: 0.7006 - accuracy: 0.502321/21 [==============================] - 5s 239ms/step - loss: 0.7006 - accuracy: 0.5022
Epoch 10/10
 1/21 [>.............................] - ETA: 4s - loss: 0.7004 - accuracy: 0.5078 2/21 [=>............................] - ETA: 4s - loss: 0.7000 - accuracy: 0.5063 3/21 [===>..........................] - ETA: 4s - loss: 0.6996 - accuracy: 0.5063 4/21 [====>.........................] - ETA: 4s - loss: 0.6995 - accuracy: 0.5066 5/21 [======>.......................] - ETA: 3s - loss: 0.6994 - accuracy: 0.5063 6/21 [=======>......................] - ETA: 3s - loss: 0.6995 - accuracy: 0.5058 7/21 [=========>....................] - ETA: 3s - loss: 0.6996 - accuracy: 0.5056 8/21 [==========>...................] - ETA: 3s - loss: 0.6997 - accuracy: 0.5053 9/21 [===========>..................] - ETA: 2s - loss: 0.6997 - accuracy: 0.504910/21 [=============>................] - ETA: 2s - loss: 0.6997 - accuracy: 0.504611/21 [==============>...............] - ETA: 2s - loss: 0.6997 - accuracy: 0.504112/21 [================>.............] - ETA: 2s - loss: 0.6997 - accuracy: 0.503713/21 [=================>............] - ETA: 2s - loss: 0.6997 - accuracy: 0.503414/21 [===================>..........] - ETA: 1s - loss: 0.6997 - accuracy: 0.503115/21 [====================>.........] - ETA: 1s - loss: 0.6997 - accuracy: 0.502816/21 [=====================>........] - ETA: 1s - loss: 0.6998 - accuracy: 0.502517/21 [=======================>......] - ETA: 0s - loss: 0.6998 - accuracy: 0.502418/21 [========================>.....] - ETA: 0s - loss: 0.6998 - accuracy: 0.502319/21 [==========================>...] - ETA: 0s - loss: 0.6998 - accuracy: 0.502220/21 [===========================>..] - ETA: 0s - loss: 0.6998 - accuracy: 0.502221/21 [==============================] - ETA: 0s - loss: 0.6998 - accuracy: 0.502221/21 [==============================] - 5s 239ms/step - loss: 0.6997 - accuracy: 0.5022

1/1 - 0s
DataSource(0e6d0a6ad257415d99ba0b8c40539b39T)

2021-01-18 LSTM预测大盘上涨小于0.5,全仓卖出
2021-02-23 LSTM预测大盘上涨小于0.5,全仓卖出
2021-02-25 LSTM预测大盘上涨小于0.5,全仓卖出
2021-03-17 LSTM预测大盘上涨小于0.5,全仓卖出

2021-04-26 LSTM预测大盘上涨小于0.5,全仓卖出

2021-08-18 LSTM预测大盘上涨小于0.5,全仓卖出
2021-08-19 LSTM预测大盘上涨小于0.5,全仓卖出

2021-09-02 LSTM预测大盘上涨小于0.5,全仓卖出
2021-09-06 LSTM预测大盘上涨小于0.5,全仓卖出
2021-09-23 LSTM预测大盘上涨小于0.5,全仓卖出
2021-10-12 LSTM预测大盘上涨小于0.5,全仓卖出
2021-12-06 LSTM预测大盘上涨小于0.5,全仓卖出
2021-12-07 LSTM预测大盘上涨小于0.5,全仓卖出

收益率52.88%
年化收益率55.3%
基准收益率-5.2%
阿尔法0.59
贝塔0.38
夏普比率1.95
胜率0.51
盈亏比1.56
收益波动率22.37%
信息比率0.13
最大回撤15.07%

bigcharts-data-start/{"__type":"tabs","__id":"bigchart-4b878be0e6614e69ab31bdfaed133b91"}/bigcharts-data-end

In [84]:

# c1=m21.data_1.read()
# c2=pd.DataFrame(c1)
# c2.columns=['result']
# c2

Out[84]:

	result
date
2020-10-09	0.498755
2020-10-12	0.502414
2020-10-13	0.491644
2020-10-14	0.504739
2020-10-15	0.494695
...	...
2021-12-27	0.494163
2021-12-28	0.499631
2021-12-29	0.491855
2021-12-30	0.503310
2021-12-31	0.496119

1515 rows × 1 columns

大盘风控 的一些案例¶

思路三:训练一个LSTM神经网络模型预测涨跌，来作为大盘风控的依据¶

大盘风控模块，读取深度学习模型预测涨跌的风控数据¶

大盘风控的一些案例¶