复制链接

克隆策略

In [26]:

# 本代码由可视化策略环境自动生成 2020年3月2日 15:31
# 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 m2_run_bigquant_run(input_1, input_2, input_3):
    pred_label = input_1.read_pickle()
    test_data = input_2.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)
    
#    pred_label = np.where(pred_label>0.5,1,0)
#    labels = test_data['y']
#    print('准确率%s'%(np.mean(pred_label==labels)))
    
    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='100,1',
    batch_shape='',
    dtype='float32',
    sparse=False,
    name=''
)

m4 = M.dl_layer_lstm.v1(
    inputs=m3.data,
    units=100,
    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=0.2,
    recurrent_dropout=0,
    return_sequences=True,
    implementation='1',
    name=''
)

m25 = M.dl_layer_lstm.v1(
    inputs=m4.data,
    units=100,
    activation='tanh',
    recurrent_activation='hard_sigmoid',
    use_bias=True,
    kernel_initializer='glorot_uniform',
    recurrent_initializer='Orthogonal',
    bias_initializer='Zeros',
    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=0.2,
    recurrent_dropout=0,
    return_sequences=False,
    implementation='1',
    name=''
)

m9 = M.dl_layer_dense.v1(
    inputs=m25.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=''
)

m12 = M.dl_layer_activation.v1(
    inputs=m9.data,
    activation='tanh',
    name=''
)

m5 = M.dl_model_init.v1(
    inputs=m3.data,
    outputs=m12.data
)

m8 = M.input_features.v1(
    features='shift(close_0,1)'
)

m24 = M.instruments.v2(
    start_date='2015-01-01',
    end_date='2017-03-01',
    market='CN_STOCK_A',
    instrument_list='600009.SHA',
    max_count=0
)

m23 = 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：未来10日上涨标记为1，否则为0
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={}
)

m15 = M.general_feature_extractor_vx1.v1(
    instruments=m24.data,
    features=m8.data,
    start_date='',
    end_date='',
    before_start_days=90
)

m16 = M.derived_feature_extractor.v3(
    input_data=m15.data,
    features=m8.data,
    date_col='date',
    instrument_col='instrument',
    drop_na=False,
    remove_extra_columns=False,
    user_functions={}
)

m17 = M.join.v3(
    data1=m23.data,
    data2=m16.data,
    on='date',
    how='inner',
    sort=True
)

m18 = M.dropnan.v1(
    input_data=m17.data
)

m13 = M.dl_convert_to_bin.v2(
    input_data=m18.data,
    features=m8.data,
    window_size=100,
    feature_clip=5,
    flatten=False,
    window_along_col=''
)

m6 = M.dl_model_train.v1(
    input_model=m5.data,
    training_data=m13.data,
    optimizer='SGD',
    loss='binary_crossentropy',
    metrics='accuracy',
    batch_size=2048,
    epochs=10,
    n_gpus=1,
    verbose='1:输出进度条记录'
)

m19 = M.instruments.v2(
    start_date='2019-01-01',
    end_date='2020-03-01',
    market='CN_STOCK_A',
    instrument_list='600009.SHA',
    max_count=0
)

m21 = M.general_feature_extractor_vx1.v1(
    instruments=m19.data,
    features=m8.data,
    start_date='',
    end_date='',
    before_start_days=90
)

m26 = M.derived_feature_extractor.v3(
    input_data=m21.data,
    features=m8.data,
    date_col='date',
    instrument_col='instrument',
    drop_na=False,
    remove_extra_columns=False,
    user_functions={}
)

m20 = M.dropnan.v1(
    input_data=m26.data
)

m14 = M.dl_convert_to_bin.v2(
    input_data=m20.data,
    features=m8.data,
    window_size=100,
    feature_clip=5,
    flatten=False,
    window_along_col=''
)

m7 = M.dl_model_predict.v1(
    trained_model=m6.data,
    input_data=m14.data,
    batch_size=10240,
    n_gpus=2,
    verbose='2:每个epoch输出一行记录'
)

m2 = M.cached.v3(
    input_1=m7.data,
    input_2=m14.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=m19.data,
    options_data=m2.data_1,
    start_date='2017-04-01',
    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=''
)

[2020-03-02 15:30:02.069365] INFO: bigquant: dl_layer_input.v1 运行完成[0.003092s].

[2020-03-02 15:30:02.521524] INFO: bigquant: dl_layer_lstm.v1 运行完成[0.449405s].

[2020-03-02 15:30:02.977858] INFO: bigquant: dl_layer_lstm.v1 运行完成[0.453976s].

[2020-03-02 15:30:02.992978] INFO: bigquant: dl_layer_dense.v1 运行完成[0.012194s].

[2020-03-02 15:30:02.997217] INFO: bigquant: dl_layer_activation.v1 运行完成[0.002507s].

[2020-03-02 15:30:03.024807] INFO: bigquant: cached.v3 开始运行..

[2020-03-02 15:30:03.126089] INFO: bigquant: cached.v3 运行完成[0.101277s].

[2020-03-02 15:30:03.128202] INFO: bigquant: dl_model_init.v1 运行完成[0.128969s].

[2020-03-02 15:30:03.130719] INFO: bigquant: input_features.v1 开始运行..

[2020-03-02 15:30:03.155013] INFO: bigquant: 命中缓存

[2020-03-02 15:30:03.156623] INFO: bigquant: input_features.v1 运行完成[0.025893s].

[2020-03-02 15:30:03.158858] INFO: bigquant: instruments.v2 开始运行..

[2020-03-02 15:30:03.182734] INFO: bigquant: 命中缓存

[2020-03-02 15:30:03.184444] INFO: bigquant: instruments.v2 运行完成[0.02557s].

[2020-03-02 15:30:03.189618] INFO: bigquant: advanced_auto_labeler.v2 开始运行..

[2020-03-02 15:30:03.218479] INFO: bigquant: 命中缓存

[2020-03-02 15:30:03.220470] INFO: bigquant: advanced_auto_labeler.v2 运行完成[0.030846s].

[2020-03-02 15:30:03.244077] INFO: bigquant: general_feature_extractor_vx1.v1 开始运行..

[2020-03-02 15:30:03.268207] INFO: bigquant: 命中缓存

[2020-03-02 15:30:03.270974] INFO: bigquant: general_feature_extractor_vx1.v1 运行完成[0.026885s].

[2020-03-02 15:30:03.277162] INFO: bigquant: derived_feature_extractor.v3 开始运行..

[2020-03-02 15:30:03.301679] INFO: bigquant: 命中缓存

[2020-03-02 15:30:03.303044] INFO: bigquant: derived_feature_extractor.v3 运行完成[0.025903s].

[2020-03-02 15:30:03.305610] INFO: bigquant: join.v3 开始运行..

[2020-03-02 15:30:03.354700] INFO: bigquant: 命中缓存

[2020-03-02 15:30:03.356244] INFO: bigquant: join.v3 运行完成[0.050632s].

[2020-03-02 15:30:03.358892] INFO: bigquant: dropnan.v1 开始运行..

[2020-03-02 15:30:03.388784] INFO: bigquant: 命中缓存

[2020-03-02 15:30:03.391330] INFO: bigquant: dropnan.v1 运行完成[0.03241s].

[2020-03-02 15:30:03.421884] INFO: bigquant: dl_convert_to_bin.v2 开始运行..

[2020-03-02 15:30:03.452180] INFO: bigquant: 命中缓存

[2020-03-02 15:30:03.453937] INFO: bigquant: dl_convert_to_bin.v2 运行完成[0.032052s].

[2020-03-02 15:30:03.456482] INFO: bigquant: dl_model_train.v1 开始运行..

[2020-03-02 15:30:04.550105] INFO: dl_model_train: 准备训练，训练样本个数：525，迭代次数：10

[2020-03-02 15:31:10.893909] INFO: dl_model_train: 训练结束，耗时：66.34s

[2020-03-02 15:31:10.994721] INFO: bigquant: dl_model_train.v1 运行完成[67.538233s].

[2020-03-02 15:31:10.996944] INFO: bigquant: instruments.v2 开始运行..

[2020-03-02 15:31:11.027987] INFO: bigquant: 命中缓存

[2020-03-02 15:31:11.032525] INFO: bigquant: instruments.v2 运行完成[0.035564s].

[2020-03-02 15:31:11.060030] INFO: bigquant: general_feature_extractor_vx1.v1 开始运行..

[2020-03-02 15:31:11.089693] INFO: bigquant: 命中缓存

[2020-03-02 15:31:11.091085] INFO: bigquant: general_feature_extractor_vx1.v1 运行完成[0.031063s].

[2020-03-02 15:31:11.093100] INFO: bigquant: derived_feature_extractor.v3 开始运行..

[2020-03-02 15:31:11.130458] INFO: bigquant: 命中缓存

[2020-03-02 15:31:11.131781] INFO: bigquant: derived_feature_extractor.v3 运行完成[0.038673s].

[2020-03-02 15:31:11.136765] INFO: bigquant: dropnan.v1 开始运行..

[2020-03-02 15:31:11.162686] INFO: bigquant: 命中缓存

[2020-03-02 15:31:11.164042] INFO: bigquant: dropnan.v1 运行完成[0.02727s].

[2020-03-02 15:31:11.186039] INFO: bigquant: dl_convert_to_bin.v2 开始运行..

[2020-03-02 15:31:11.212126] INFO: bigquant: 命中缓存

[2020-03-02 15:31:11.213505] INFO: bigquant: dl_convert_to_bin.v2 运行完成[0.027471s].

[2020-03-02 15:31:11.217566] INFO: bigquant: dl_model_predict.v1 开始运行..

[2020-03-02 15:31:12.552076] INFO: device_manager: 没有gpu资源，将使用cpu计算

[2020-03-02 15:31:12.554934] INFO: device_manager: 本次操作不使用GPU

[2020-03-02 15:31:14.497730] INFO: bigquant: dl_model_predict.v1 运行完成[3.280141s].

[2020-03-02 15:31:14.504404] INFO: bigquant: cached.v3 开始运行..

[2020-03-02 15:31:14.719161] INFO: bigquant: cached.v3 运行完成[0.21475s].

[2020-03-02 15:31:15.024787] INFO: bigquant: backtest.v8 开始运行..

[2020-03-02 15:31:15.028793] INFO: bigquant: biglearning backtest:V8.3.2

[2020-03-02 15:31:15.030669] INFO: bigquant: product_type:stock by specified

[2020-03-02 15:31:15.372428] INFO: bigquant: cached.v2 开始运行..

[2020-03-02 15:31:33.185619] INFO: bigquant: 读取股票行情完成:952

[2020-03-02 15:31:33.321551] INFO: bigquant: cached.v2 运行完成[17.949125s].

[2020-03-02 15:31:33.368180] INFO: algo: TradingAlgorithm V1.6.5

[2020-03-02 15:31:34.614837] INFO: algo: trading transform...

[2020-03-02 15:31:36.760901] INFO: algo: handle_splits get splits [dt:2019-08-22 00:00:00+00:00] [asset:Equity(0 [600009.SHA]), ratio:0.9922354221343994]

[2020-03-02 15:31:36.762783] INFO: Position: position stock handle split[sid:0, orig_amount:18500, new_amount:18644.0, orig_cost:54.00008007534902, new_cost:53.58, ratio:0.9922354221343994, last_sale_price:84.34001159667969]

[2020-03-02 15:31:36.767335] INFO: Position: after split: PositionStock(asset:Equity(0 [600009.SHA]), amount:18644.0, cost_basis:53.58, last_sale_price:85.0)

[2020-03-02 15:31:36.768844] INFO: Position: returning cash: 64.84

[2020-03-02 15:31:37.276146] INFO: Performance: Simulated 708 trading days out of 708.

[2020-03-02 15:31:37.277611] INFO: Performance: first open: 2017-04-05 09:30:00+00:00

[2020-03-02 15:31:37.278687] INFO: Performance: last close: 2020-02-28 15:00:00+00:00

[2020-03-02 15:31:40.572547] INFO: bigquant: backtest.v8 运行完成[25.54776s].

[2020-03-02 15:31:40.574227] INFO: bigquant: trade.v4 运行完成[25.852802s].

Train on 525 samples
Epoch 1/10
525/525 [==============================] - 17s 32ms/sample - loss: 0.6904 - accuracy: 0.5257
Epoch 2/10
525/525 [==============================] - 5s 10ms/sample - loss: 0.6931 - accuracy: 0.5314
Epoch 3/10
525/525 [==============================] - 5s 10ms/sample - loss: 0.6874 - accuracy: 0.5390
Epoch 4/10
525/525 [==============================] - 5s 9ms/sample - loss: 0.6953 - accuracy: 0.5010
Epoch 5/10
525/525 [==============================] - 6s 11ms/sample - loss: 0.6936 - accuracy: 0.5238
Epoch 6/10
525/525 [==============================] - 6s 11ms/sample - loss: 0.6890 - accuracy: 0.5314
Epoch 7/10
525/525 [==============================] - 6s 12ms/sample - loss: 0.6911 - accuracy: 0.5390
Epoch 8/10
525/525 [==============================] - 5s 9ms/sample - loss: 0.6919 - accuracy: 0.5162
Epoch 9/10
525/525 [==============================] - 5s 10ms/sample - loss: 0.6922 - accuracy: 0.5333
Epoch 10/10
525/525 [==============================] - 7s 12ms/sample - loss: 0.6923 - accuracy: 0.5333

339/339 - 2s
DataSource(9266b7ad07c94f078913ff6da14250e2T, v3)

2018-10-09 15:00:00+00:00 买入！

收益率22.59%
年化收益率7.52%
基准收益率14.0%
阿尔法0.05
贝塔0.63
夏普比率0.3
胜率1.0
盈亏比0.0
收益波动率24.56%
信息比率0.01
最大回撤27.07%

bigcharts-data-start/{"__type":"tabs","__id":"bigchart-9b92f046d2434e4a837d402bd366f03b"}/bigcharts-data-end