复制链接

克隆策略

In [6]:

# 本代码由可视化策略环境自动生成 2021年12月14日 20:38
# 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
# }
m5_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 m28_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])
    df_value = pd.merge(left=df,right=input_3.read(),on=['date','instrument'],how='inner')
    return Outputs(data_1=DataSource.write_df(df), data_2=DataSource.write_df(df_value), data_3=None)

#     # 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])
#     df_value = pd.merge(left=df,right=input_3.read(),on=['date','instrument'],how='inner')
#     return Outputs(data_1=DataSource.write_df(df), data_2=DataSource.write_df(df_value), 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 m28_post_run_bigquant_run(outputs):
    return outputs


m1 = M.instruments.v2(
    start_date='2010-01-01',
    end_date='2015-12-31',
    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/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)
""",
    start_date='',
    end_date='',
    benchmark='000300.SHA',
    drop_na_label=True,
    cast_label_int=False
)

m13 = M.standardlize.v8(
    input_1=m2.data,
    columns_input='label',
    m_cached=False
)

m3 = M.input_features.v1(
    features="""close_0/mean(close_0,5)
close_0/mean(close_0,10)
close_0/mean(close_0,20)
close_0/open_0
open_0/mean(close_0,5)
open_0/mean(close_0,10)
open_0/mean(close_0,20)""",
    m_cached=False
)

m15 = M.general_feature_extractor.v7(
    instruments=m1.data,
    features=m3.data,
    start_date='',
    end_date='',
    before_start_days=0
)

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

m14 = M.standardlize.v8(
    input_1=m16.data,
    input_2=m3.data,
    columns_input='[]'
)

m7 = M.join.v3(
    data1=m13.data,
    data2=m14.data,
    on='date,instrument',
    how='inner',
    sort=False
)

m26 = M.dl_convert_to_bin.v2(
    input_data=m7.data,
    features=m3.data,
    window_size=1,
    feature_clip=5,
    flatten=True,
    window_along_col='instrument'
)

m9 = M.instruments.v2(
    start_date=T.live_run_param('trading_date', '2016-01-01'),
    end_date=T.live_run_param('trading_date', '2019-04-20'),
    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=0
)

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

m25 = M.standardlize.v8(
    input_1=m18.data,
    input_2=m3.data,
    columns_input='[]'
)

m27 = M.dl_convert_to_bin.v2(
    input_data=m25.data,
    features=m3.data,
    window_size=1,
    feature_clip=5,
    flatten=True,
    window_along_col='instrument'
)

m10 = M.advanced_auto_labeler.v2(
    instruments=m9.data,
    label_expr="""# #号开始的表示注释
# 0. 每行一个，顺序执行，从第二个开始，可以使用label字段
# 1. 可用数据字段见 https://bigquant.com/docs/data_history_data.html
#   添加benchmark_前缀，可使用对应的benchmark数据
# 2. 可用操作符和函数见 `表达式引擎 <https://bigquant.com/docs/big_expr.html>`_

# 计算收益：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)
""",
    start_date='',
    end_date='',
    benchmark='000300.SHA',
    drop_na_label=True,
    cast_label_int=False
)

m12 = M.standardlize.v8(
    input_1=m10.data,
    columns_input='label',
    m_cached=False
)

m6 = M.dl_layer_input.v1(
    shape='7',
    batch_shape='',
    dtype='float32',
    sparse=False,
    name=''
)

m8 = M.dl_layer_dense.v1(
    inputs=m6.data,
    units=256,
    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=''
)

m21 = M.dl_layer_dropout.v1(
    inputs=m8.data,
    rate=0.1,
    noise_shape='',
    name=''
)

m20 = M.dl_layer_dense.v1(
    inputs=m21.data,
    units=128,
    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=''
)

m22 = M.dl_layer_dropout.v1(
    inputs=m20.data,
    rate=0.1,
    noise_shape='',
    name=''
)

m23 = M.dl_layer_dense.v1(
    inputs=m22.data,
    units=1,
    activation='linear',
    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=''
)

m4 = M.dl_model_init.v1(
    inputs=m6.data,
    outputs=m23.data
)

m5 = M.dl_model_train.v1(
    input_model=m4.data,
    training_data=m26.data,
    optimizer='Adam',
    loss='mean_squared_error',
    metrics='mse',
    batch_size=1024,
    epochs=5,
    custom_objects=m5_custom_objects_bigquant_run,
    n_gpus=0,
    verbose='2:每个epoch输出一行记录'
)

m11 = M.dl_model_predict.v1(
    trained_model=m5.data,
    input_data=m27.data,
    batch_size=1024,
    n_gpus=0,
    verbose='2:每个epoch输出一行记录'
)

m28 = M.cached.v3(
    input_1=m11.data,
    input_2=m18.data,
    input_3=m12.data,
    run=m28_run_bigquant_run,
    post_run=m28_post_run_bigquant_run,
    input_ports='',
    params='{}',
    output_ports=''
)

m24 = M.metrics_regression.v1(
    predictions=m28.data_2,
    explained_variance_score=True,
    mean_absolute_error=True,
    mean_squared_error=True,
    mean_squared_log_error=False,
    median_absolute_error=True,
    r2_score=True
)

[2021-12-14 20:30:49.877647] INFO: moduleinvoker: instruments.v2 开始运行..

[2021-12-14 20:30:49.892846] INFO: moduleinvoker: 命中缓存

[2021-12-14 20:30:49.895762] INFO: moduleinvoker: instruments.v2 运行完成[0.018077s].

[2021-12-14 20:30:49.909783] INFO: moduleinvoker: advanced_auto_labeler.v2 开始运行..

[2021-12-14 20:30:49.924346] INFO: moduleinvoker: 命中缓存

[2021-12-14 20:30:49.926018] INFO: moduleinvoker: advanced_auto_labeler.v2 运行完成[0.016256s].

[2021-12-14 20:30:49.933517] INFO: moduleinvoker: standardlize.v8 开始运行..

[2021-12-14 20:31:25.290048] INFO: moduleinvoker: standardlize.v8 运行完成[35.356502s].

[2021-12-14 20:31:25.331496] INFO: moduleinvoker: input_features.v1 开始运行..

[2021-12-14 20:31:25.370607] INFO: moduleinvoker: input_features.v1 运行完成[0.039135s].

[2021-12-14 20:31:25.388463] INFO: moduleinvoker: general_feature_extractor.v7 开始运行..

[2021-12-14 20:31:25.398111] INFO: moduleinvoker: 命中缓存

[2021-12-14 20:31:25.400888] INFO: moduleinvoker: general_feature_extractor.v7 运行完成[0.012459s].

[2021-12-14 20:31:25.422226] INFO: moduleinvoker: derived_feature_extractor.v3 开始运行..

[2021-12-14 20:31:35.640060] INFO: derived_feature_extractor: 提取完成 close_0/mean(close_0,5), 2.703s

[2021-12-14 20:31:38.285768] INFO: derived_feature_extractor: 提取完成 close_0/mean(close_0,10), 2.644s

[2021-12-14 20:31:40.832744] INFO: derived_feature_extractor: 提取完成 close_0/mean(close_0,20), 2.545s

[2021-12-14 20:31:40.839800] INFO: derived_feature_extractor: 提取完成 close_0/open_0, 0.005s

[2021-12-14 20:31:43.547386] INFO: derived_feature_extractor: 提取完成 open_0/mean(close_0,5), 2.706s

[2021-12-14 20:31:46.147334] INFO: derived_feature_extractor: 提取完成 open_0/mean(close_0,10), 2.598s

[2021-12-14 20:31:48.751093] INFO: derived_feature_extractor: 提取完成 open_0/mean(close_0,20), 2.602s

[2021-12-14 20:31:50.115337] INFO: derived_feature_extractor: /y_2010, 431567

[2021-12-14 20:31:51.594580] INFO: derived_feature_extractor: /y_2011, 511455

[2021-12-14 20:31:53.067801] INFO: derived_feature_extractor: /y_2012, 565675

[2021-12-14 20:31:54.555441] INFO: derived_feature_extractor: /y_2013, 564168

[2021-12-14 20:31:56.135676] INFO: derived_feature_extractor: /y_2014, 569948

[2021-12-14 20:31:57.621015] INFO: derived_feature_extractor: /y_2015, 569698

[2021-12-14 20:31:58.478901] INFO: moduleinvoker: derived_feature_extractor.v3 运行完成[33.05668s].

[2021-12-14 20:31:58.489326] INFO: moduleinvoker: standardlize.v8 开始运行..

[2021-12-14 20:32:56.614599] INFO: moduleinvoker: standardlize.v8 运行完成[58.125281s].

[2021-12-14 20:32:56.629844] INFO: moduleinvoker: join.v3 开始运行..

[2021-12-14 20:34:07.224436] INFO: join: /data, 行数=3133725/3159055, 耗时=56.036422s

[2021-12-14 20:34:07.497437] INFO: join: 最终行数: 3133725

[2021-12-14 20:34:07.625931] INFO: moduleinvoker: join.v3 运行完成[70.996068s].

[2021-12-14 20:34:07.658428] INFO: moduleinvoker: dl_convert_to_bin.v2 开始运行..

[2021-12-14 20:34:13.482888] INFO: moduleinvoker: dl_convert_to_bin.v2 运行完成[5.824489s].

[2021-12-14 20:34:13.497081] INFO: moduleinvoker: instruments.v2 开始运行..

[2021-12-14 20:34:13.512183] INFO: moduleinvoker: 命中缓存

[2021-12-14 20:34:13.514397] INFO: moduleinvoker: instruments.v2 运行完成[0.01731s].

[2021-12-14 20:34:13.542525] INFO: moduleinvoker: general_feature_extractor.v7 开始运行..

[2021-12-14 20:34:13.557025] INFO: moduleinvoker: 命中缓存

[2021-12-14 20:34:13.558731] INFO: moduleinvoker: general_feature_extractor.v7 运行完成[0.016234s].

[2021-12-14 20:34:13.568420] INFO: moduleinvoker: derived_feature_extractor.v3 开始运行..

[2021-12-14 20:34:21.312322] INFO: derived_feature_extractor: 提取完成 close_0/mean(close_0,5), 2.267s

[2021-12-14 20:34:23.504969] INFO: derived_feature_extractor: 提取完成 close_0/mean(close_0,10), 2.185s

[2021-12-14 20:34:25.705907] INFO: derived_feature_extractor: 提取完成 close_0/mean(close_0,20), 2.199s

[2021-12-14 20:34:25.719441] INFO: derived_feature_extractor: 提取完成 close_0/open_0, 0.011s

[2021-12-14 20:34:27.959577] INFO: derived_feature_extractor: 提取完成 open_0/mean(close_0,5), 2.239s

[2021-12-14 20:34:30.126921] INFO: derived_feature_extractor: 提取完成 open_0/mean(close_0,10), 2.166s

[2021-12-14 20:34:32.300843] INFO: derived_feature_extractor: 提取完成 open_0/mean(close_0,20), 2.171s

[2021-12-14 20:34:34.080828] INFO: derived_feature_extractor: /y_2016, 641546

[2021-12-14 20:34:35.952070] INFO: derived_feature_extractor: /y_2017, 743233

[2021-12-14 20:34:38.128232] INFO: derived_feature_extractor: /y_2018, 816987

[2021-12-14 20:34:39.803844] INFO: derived_feature_extractor: /y_2019, 256890

[2021-12-14 20:34:40.876921] INFO: moduleinvoker: derived_feature_extractor.v3 运行完成[27.308532s].

[2021-12-14 20:34:40.885798] INFO: moduleinvoker: standardlize.v8 开始运行..

[2021-12-14 20:35:19.447176] INFO: moduleinvoker: standardlize.v8 运行完成[38.56137s].

[2021-12-14 20:35:19.463265] INFO: moduleinvoker: dl_convert_to_bin.v2 开始运行..

[2021-12-14 20:35:21.662101] INFO: moduleinvoker: dl_convert_to_bin.v2 运行完成[2.198829s].

[2021-12-14 20:35:21.673702] INFO: moduleinvoker: advanced_auto_labeler.v2 开始运行..

[2021-12-14 20:35:21.685284] INFO: moduleinvoker: 命中缓存

[2021-12-14 20:35:21.687795] INFO: moduleinvoker: advanced_auto_labeler.v2 运行完成[0.014103s].

[2021-12-14 20:35:21.694680] INFO: moduleinvoker: standardlize.v8 开始运行..

[2021-12-14 20:35:43.165702] INFO: moduleinvoker: standardlize.v8 运行完成[21.471019s].

[2021-12-14 20:35:43.181395] INFO: moduleinvoker: dl_layer_input.v1 运行完成[0.001842s].

[2021-12-14 20:35:43.211536] INFO: moduleinvoker: dl_layer_dense.v1 运行完成[0.021737s].

[2021-12-14 20:35:43.224114] INFO: moduleinvoker: dl_layer_dropout.v1 运行完成[0.004622s].

[2021-12-14 20:35:43.242120] INFO: moduleinvoker: dl_layer_dense.v1 运行完成[0.011297s].

[2021-12-14 20:35:43.256552] INFO: moduleinvoker: dl_layer_dropout.v1 运行完成[0.005375s].

[2021-12-14 20:35:43.274052] INFO: moduleinvoker: dl_layer_dense.v1 运行完成[0.010951s].

[2021-12-14 20:35:43.340431] INFO: moduleinvoker: cached.v3 开始运行..

[2021-12-14 20:35:43.351647] INFO: moduleinvoker: 命中缓存

[2021-12-14 20:35:43.353642] INFO: moduleinvoker: cached.v3 运行完成[0.013252s].

[2021-12-14 20:35:43.355785] INFO: moduleinvoker: dl_model_init.v1 运行完成[0.049258s].

[2021-12-14 20:35:43.361445] INFO: moduleinvoker: dl_model_train.v1 开始运行..

[2021-12-14 20:35:43.706959] INFO: dl_model_train: 准备训练，训练样本个数：3133725，迭代次数：5

[2021-12-14 20:36:17.119301] INFO: dl_model_train: 训练结束，耗时：33.41s

[2021-12-14 20:36:17.416234] INFO: moduleinvoker: dl_model_train.v1 运行完成[34.054753s].

[2021-12-14 20:36:17.426424] INFO: moduleinvoker: dl_model_predict.v1 开始运行..

[2021-12-14 20:36:22.529228] INFO: moduleinvoker: dl_model_predict.v1 运行完成[5.102829s].

[2021-12-14 20:36:22.543690] INFO: moduleinvoker: cached.v3 开始运行..

[2021-12-14 20:36:36.902159] INFO: moduleinvoker: cached.v3 运行完成[14.358476s].

[2021-12-14 20:36:37.996145] INFO: moduleinvoker: metrics_regression.v1 运行完成[1.070822s].

Epoch 1/5
3061/3061 - 11s - loss: 0.9903 - mse: 0.9903
Epoch 2/5
3061/3061 - 5s - loss: 0.9891 - mse: 0.9891
Epoch 3/5
3061/3061 - 5s - loss: 0.9888 - mse: 0.9888
Epoch 4/5
3061/3061 - 5s - loss: 0.9886 - mse: 0.9886
Epoch 5/5
3061/3061 - 7s - loss: 0.9885 - mse: 0.9885

2334/2334 - 3s
DataSource(255c79349df041069a94b36aac7b4224T)

可解释方差: 0.002733191821018943
平均绝对误差: 0.707599771603117
均方误差: 0.9853414225028186
均方绝对误差: 0.5094940363276789
确定系数(r^2): 0.0027071536646356975

In [8]:

m24.report

Out[8]:

{'explained_variance_score': 0.002733191821018943,
 'mean_absolute_error': 0.707599771603117,
 'mean_squared_error': 0.9853414225028186,
 'median_absolute_error': 0.5094940363276789,
 'r2_score': 0.0027071536646356975}