复制链接

克隆策略

In [1]:

# 本代码由可视化策略环境自动生成 2022年12月15日 15:41
# 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 m30_run_bigquant_run(input_1, input_2, input_3):
    # 示例代码如下。在这里编写您的代码
    from sklearn.model_selection import train_test_split
    data = input_2.read()
    x_train, x_val, y_train, y_val = train_test_split(data["x"], data['y'])
    data_1 = DataSource.write_pickle({'x': x_train, 'y': y_train})
    data_2 = DataSource.write_pickle({'x': x_val, 'y': y_val})
    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 m30_post_run_bigquant_run(outputs):
    return outputs

# 用户的自定义层需要写到字典中，比如
# {
#   "MyLayer": MyLayer
# }
m6_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 m2_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 m2_post_run_bigquant_run(outputs):
    return outputs

# 回测引擎：初始化函数，只执行一次
def m1_initialize_bigquant_run(context):
    # 加载预测数据
    context.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))
# 回测引擎：每日数据处理函数，每天执行一次
def m1_handle_data_bigquant_run(context, data):
    # 按日期过滤得到今日的预测数据
    try:
        prediction = context.prediction[data.current_dt.strftime('%Y-%m-%d')]
    except KeyError as e:
        return
    
    instrument = context.instruments[0]
    sid = context.symbol(instrument)
    cur_position = context.portfolio.positions[sid].amount
    
    # 交易逻辑
    if prediction > 0.5 and cur_position == 0:
        context.order_target_percent(context.symbol(instrument), 1)
        print(data.current_dt, '买入！')
        
    elif prediction < 0.5 and cur_position > 0:
        context.order_target_percent(context.symbol(instrument), 0)
        print(data.current_dt, '卖出！')
    
# 回测引擎：准备数据，只执行一次
def m1_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 m1_before_trading_start_bigquant_run(context, data):
    pass


m3 = M.dl_layer_input.v1(
    shape='50,5',
    batch_shape='',
    dtype='float32',
    sparse=False,
    name=''
)

m13 = M.dl_layer_reshape.v1(
    inputs=m3.data,
    target_shape='50,5,1',
    name=''
)

m14 = M.dl_layer_conv2d.v1(
    inputs=m13.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=''
)

m15 = M.dl_layer_reshape.v1(
    inputs=m14.data,
    target_shape='48,32',
    name=''
)

m4 = M.dl_layer_lstm.v1(
    inputs=m15.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=''
)

m11 = M.dl_layer_dropout.v1(
    inputs=m4.data,
    rate=0.4,
    noise_shape='',
    name=''
)

m10 = M.dl_layer_dense.v1(
    inputs=m11.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=''
)

m12 = M.dl_layer_dropout.v1(
    inputs=m10.data,
    rate=0.8,
    noise_shape='',
    name=''
)

m9 = M.dl_layer_dense.v1(
    inputs=m12.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=''
)

m5 = M.dl_model_init.v1(
    inputs=m3.data,
    outputs=m9.data
)

m8 = 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"""
)

m24 = M.instruments.v2(
    start_date='2015-07-02',
    end_date='2017-10-30',
    market='CN_STOCK_A',
    instrument_list='600009.SHA',
    max_count=0
)

m21 = M.advanced_auto_labeler.v2(
    instruments=m24.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={}
)

m22 = M.general_feature_extractor.v7(
    instruments=m24.data,
    features=m8.data,
    start_date='',
    end_date='',
    before_start_days=90
)

m23 = M.derived_feature_extractor.v3(
    input_data=m22.data,
    features=m8.data,
    date_col='date',
    instrument_col='instrument',
    drop_na=False,
    remove_extra_columns=False,
    user_functions={}
)

m17 = M.join.v3(
    data1=m21.data,
    data2=m23.data,
    on='date',
    how='inner',
    sort=True
)

m19 = M.dropnan.v2(
    input_data=m17.data
)

m18 = M.standardlize.v8(
    input_1=m19.data,
    input_2=m8.data,
    columns_input=''
)

m25 = M.dl_convert_to_bin.v2(
    input_data=m18.data,
    features=m8.data,
    window_size=50,
    feature_clip=5,
    flatten=False,
    window_along_col=''
)

m30 = M.cached.v3(
    input_2=m25.data,
    run=m30_run_bigquant_run,
    post_run=m30_post_run_bigquant_run,
    input_ports='',
    params='{}',
    output_ports=''
)

m6 = M.dl_model_train.v1(
    input_model=m5.data,
    training_data=m30.data_1,
    validation_data=m30.data_2,
    optimizer='Adam',
    loss='binary_crossentropy',
    metrics='accuracy',
    batch_size=2048,
    epochs=10,
    custom_objects=m6_custom_objects_bigquant_run,
    n_gpus=0,
    verbose='1:输出进度条记录'
)

m28 = M.instruments.v2(
    start_date=T.live_run_param('trading_date', '2015-02-11'),
    end_date=T.live_run_param('trading_date', '2019-09-01'),
    market='CN_STOCK_A',
    instrument_list='600009.SHA',
    max_count=0
)

m16 = M.general_feature_extractor.v7(
    instruments=m28.data,
    features=m8.data,
    start_date='',
    end_date='',
    before_start_days=90
)

m26 = M.derived_feature_extractor.v3(
    input_data=m16.data,
    features=m8.data,
    date_col='date',
    instrument_col='instrument',
    drop_na=False,
    remove_extra_columns=False,
    user_functions={}
)

m29 = M.dropnan.v2(
    input_data=m26.data
)

m20 = M.standardlize.v8(
    input_1=m29.data,
    input_2=m8.data,
    columns_input=''
)

m27 = M.dl_convert_to_bin.v2(
    input_data=m20.data,
    features=m8.data,
    window_size=50,
    feature_clip=5,
    flatten=False,
    window_along_col=''
)

m7 = M.dl_model_predict.v1(
    trained_model=m6.data,
    input_data=m27.data,
    batch_size=10240,
    n_gpus=0,
    verbose='2:每个epoch输出一行记录'
)

m2 = M.cached.v3(
    input_1=m7.data,
    input_2=m27.data,
    input_3=m20.data,
    run=m2_run_bigquant_run,
    post_run=m2_post_run_bigquant_run,
    input_ports='',
    params='{}',
    output_ports=''
)

m1 = M.trade.v4(
    instruments=m28.data,
    options_data=m2.data_1,
    start_date='',
    end_date='',
    initialize=m1_initialize_bigquant_run,
    handle_data=m1_handle_data_bigquant_run,
    prepare=m1_prepare_bigquant_run,
    before_trading_start=m1_before_trading_start_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=''
)

[2022-12-15 15:40:57.403816] INFO: moduleinvoker: dl_layer_input.v1 运行完成[4.886038s].

[2022-12-15 15:40:57.434708] INFO: moduleinvoker: dl_layer_reshape.v1 运行完成[0.015537s].

[2022-12-15 15:40:57.553274] INFO: moduleinvoker: dl_layer_conv2d.v1 运行完成[0.105243s].

[2022-12-15 15:40:57.571558] INFO: moduleinvoker: dl_layer_reshape.v1 运行完成[0.010806s].

[2022-12-15 15:40:57.854046] INFO: moduleinvoker: dl_layer_lstm.v1 运行完成[0.269981s].

[2022-12-15 15:40:57.873217] INFO: moduleinvoker: dl_layer_dropout.v1 运行完成[0.004814s].

[2022-12-15 15:40:57.908595] INFO: moduleinvoker: dl_layer_dense.v1 运行完成[0.021511s].

[2022-12-15 15:40:57.932152] INFO: moduleinvoker: dl_layer_dropout.v1 运行完成[0.006318s].

[2022-12-15 15:40:57.954465] INFO: moduleinvoker: dl_layer_dense.v1 运行完成[0.014s].

[2022-12-15 15:40:58.021662] INFO: moduleinvoker: cached.v3 开始运行..

[2022-12-15 15:40:58.035759] INFO: moduleinvoker: 命中缓存

[2022-12-15 15:40:58.038267] INFO: moduleinvoker: cached.v3 运行完成[0.016628s].

[2022-12-15 15:40:58.042799] INFO: moduleinvoker: dl_model_init.v1 运行完成[0.077993s].

[2022-12-15 15:40:58.049904] INFO: moduleinvoker: input_features.v1 开始运行..

[2022-12-15 15:40:58.060610] INFO: moduleinvoker: 命中缓存

[2022-12-15 15:40:58.062966] INFO: moduleinvoker: input_features.v1 运行完成[0.013069s].

[2022-12-15 15:40:58.070766] INFO: moduleinvoker: instruments.v2 开始运行..

[2022-12-15 15:40:58.079947] INFO: moduleinvoker: 命中缓存

[2022-12-15 15:40:58.082019] INFO: moduleinvoker: instruments.v2 运行完成[0.011275s].

[2022-12-15 15:40:58.099062] INFO: moduleinvoker: advanced_auto_labeler.v2 开始运行..

[2022-12-15 15:40:58.110201] INFO: moduleinvoker: 命中缓存

[2022-12-15 15:40:58.541380] INFO: moduleinvoker: advanced_auto_labeler.v2 运行完成[0.442326s].

[2022-12-15 15:40:58.562533] INFO: moduleinvoker: general_feature_extractor.v7 开始运行..

[2022-12-15 15:40:58.577314] INFO: moduleinvoker: 命中缓存

[2022-12-15 15:40:58.579865] INFO: moduleinvoker: general_feature_extractor.v7 运行完成[0.017353s].

[2022-12-15 15:40:58.597051] INFO: moduleinvoker: derived_feature_extractor.v3 开始运行..

[2022-12-15 15:40:58.608078] INFO: moduleinvoker: 命中缓存

[2022-12-15 15:40:58.610775] INFO: moduleinvoker: derived_feature_extractor.v3 运行完成[0.013735s].

[2022-12-15 15:40:58.633753] INFO: moduleinvoker: join.v3 开始运行..

[2022-12-15 15:40:58.643273] INFO: moduleinvoker: 命中缓存

[2022-12-15 15:40:58.646109] INFO: moduleinvoker: join.v3 运行完成[0.012381s].

[2022-12-15 15:40:58.677871] INFO: moduleinvoker: dropnan.v2 开始运行..

[2022-12-15 15:40:58.708520] INFO: moduleinvoker: 命中缓存

[2022-12-15 15:40:58.711598] INFO: moduleinvoker: dropnan.v2 运行完成[0.033708s].

[2022-12-15 15:40:58.724906] INFO: moduleinvoker: standardlize.v8 开始运行..

[2022-12-15 15:40:58.736928] INFO: moduleinvoker: 命中缓存

[2022-12-15 15:40:58.739559] INFO: moduleinvoker: standardlize.v8 运行完成[0.01466s].

[2022-12-15 15:40:58.758164] INFO: moduleinvoker: dl_convert_to_bin.v2 开始运行..

[2022-12-15 15:40:58.768476] INFO: moduleinvoker: 命中缓存

[2022-12-15 15:40:58.770842] INFO: moduleinvoker: dl_convert_to_bin.v2 运行完成[0.012706s].

[2022-12-15 15:40:58.789412] INFO: moduleinvoker: cached.v3 开始运行..

[2022-12-15 15:40:58.803400] INFO: moduleinvoker: 命中缓存

[2022-12-15 15:40:58.806251] INFO: moduleinvoker: cached.v3 运行完成[0.016857s].

[2022-12-15 15:40:58.830984] INFO: moduleinvoker: dl_model_train.v1 开始运行..

[2022-12-15 15:40:58.843515] INFO: moduleinvoker: 命中缓存

[2022-12-15 15:40:58.848253] INFO: moduleinvoker: dl_model_train.v1 运行完成[0.017283s].

[2022-12-15 15:40:58.856198] INFO: moduleinvoker: instruments.v2 开始运行..

[2022-12-15 15:40:58.866230] INFO: moduleinvoker: 命中缓存

[2022-12-15 15:40:58.868795] INFO: moduleinvoker: instruments.v2 运行完成[0.012687s].

[2022-12-15 15:40:58.886490] INFO: moduleinvoker: general_feature_extractor.v7 开始运行..

[2022-12-15 15:40:58.895283] INFO: moduleinvoker: 命中缓存

[2022-12-15 15:40:58.897218] INFO: moduleinvoker: general_feature_extractor.v7 运行完成[0.010758s].

[2022-12-15 15:40:58.906404] INFO: moduleinvoker: derived_feature_extractor.v3 开始运行..

[2022-12-15 15:40:58.916316] INFO: moduleinvoker: 命中缓存

[2022-12-15 15:40:58.918760] INFO: moduleinvoker: derived_feature_extractor.v3 运行完成[0.012365s].

[2022-12-15 15:40:58.929555] INFO: moduleinvoker: dropnan.v2 开始运行..

[2022-12-15 15:40:58.941038] INFO: moduleinvoker: 命中缓存

[2022-12-15 15:40:58.943159] INFO: moduleinvoker: dropnan.v2 运行完成[0.013602s].

[2022-12-15 15:40:58.950058] INFO: moduleinvoker: standardlize.v8 开始运行..

[2022-12-15 15:40:58.959451] INFO: moduleinvoker: 命中缓存

[2022-12-15 15:40:58.961631] INFO: moduleinvoker: standardlize.v8 运行完成[0.011569s].

[2022-12-15 15:40:58.980076] INFO: moduleinvoker: dl_convert_to_bin.v2 开始运行..

[2022-12-15 15:40:58.988306] INFO: moduleinvoker: 命中缓存

[2022-12-15 15:40:58.990076] INFO: moduleinvoker: dl_convert_to_bin.v2 运行完成[0.010028s].

[2022-12-15 15:40:58.999126] INFO: moduleinvoker: dl_model_predict.v1 开始运行..

[2022-12-15 15:40:59.008021] INFO: moduleinvoker: 命中缓存

[2022-12-15 15:40:59.010372] INFO: moduleinvoker: dl_model_predict.v1 运行完成[0.011271s].

[2022-12-15 15:40:59.025946] INFO: moduleinvoker: cached.v3 开始运行..

[2022-12-15 15:40:59.037549] INFO: moduleinvoker: 命中缓存

[2022-12-15 15:40:59.040142] INFO: moduleinvoker: cached.v3 运行完成[0.014216s].

[2022-12-15 15:41:03.191780] INFO: moduleinvoker: backtest.v8 开始运行..

[2022-12-15 15:41:03.256835] INFO: moduleinvoker: 命中缓存

[2022-12-15 15:41:04.821160] INFO: moduleinvoker: backtest.v8 运行完成[1.629379s].

[2022-12-15 15:41:04.823640] INFO: moduleinvoker: trade.v4 运行完成[5.772099s].

DataSource(40af4ed7616c4867b280e42d961c472cT)

收益率0.0%
年化收益率0.0%
基准收益率11.52%
阿尔法-0.03
贝塔0.0
夏普比率n/a
胜率0.0
盈亏比0.0
收益波动率0.0%
信息比率-0.01
最大回撤0.0%

bigcharts-data-start/{"__type":"tabs","__id":"bigchart-09fb2986928145cb91b45330c8fcb8c5"}/bigcharts-data-end

In [2]:

# 方法一：手动绘制曲线
from matplotlib import pyplot as plt

train_loss = m6.data.read()["history"]["loss"]
val_loss = m6.data.read()["history"]["val_loss"]

plt.plot(train_loss, label="train")
plt.plot(val_loss, label="validation")
plt.legend()
plt.show()

In [3]:

train_acc = m6.data.read()["history"]["accuracy"]
val_acc = m6.data.read()["history"]["val_accuracy"]

plt.plot(train_acc, label="train")
plt.plot(val_acc, label="validation")
plt.legend()
plt.show()