复制链接

克隆策略

In [1]:

import math
import torch
import torch.nn as nn

from bigmodels.models.base import BaseModel

In [2]:

class GATModel(BaseModel):
    def __init__(self, d_feat=98, hidden_size=64, num_layers=2, dropout=0.1, base_model="GRU"):
        super().__init__()

        if base_model == "GRU":
            self.rnn = nn.GRU(
                input_size=d_feat,
                hidden_size=hidden_size,
                num_layers=num_layers,
                batch_first=True,
                dropout=dropout,
            )
        elif base_model == "LSTM":
            self.rnn = nn.LSTM(
                input_size=d_feat,
                hidden_size=hidden_size,
                num_layers=num_layers,
                batch_first=True,
                dropout=dropout,
            )
        else:
            raise ValueError("unknown base model name `%s`" % base_model)

        self.hidden_size = hidden_size
        self.d_feat = d_feat
        self.transformation = nn.Linear(self.hidden_size, self.hidden_size)
        self.a = nn.Parameter(torch.randn(self.hidden_size * 2, 1))
        self.a.requires_grad = True
        self.fc = nn.Linear(self.hidden_size, self.hidden_size)
        self.fc_out = nn.Linear(hidden_size, 1)
        self.leaky_relu = nn.LeakyReLU()
        self.softmax = nn.Softmax(dim=1)

    def cal_attention(self, x, y):
        x = self.transformation(x)
        y = self.transformation(y)

        sample_num = x.shape[0]
        dim = x.shape[1]
        e_x = x.expand(sample_num, sample_num, dim)
        e_y = torch.transpose(e_x, 0, 1)
        attention_in = torch.cat((e_x, e_y), 2).view(-1, dim * 2)
        self.a_t = torch.t(self.a)
        attention_out = self.a_t.mm(torch.t(attention_in)).view(sample_num, sample_num)
        attention_out = self.leaky_relu(attention_out)
        att_weight = self.softmax(attention_out)
        return att_weight

    def forward(self, x):
        out, _ = self.rnn(x)
        hidden = out[:, -1, :]
        att_weight = self.cal_attention(hidden, hidden)
        hidden = att_weight.mm(hidden) + hidden
        hidden = self.fc(hidden)
        hidden = self.leaky_relu(hidden)
        return self.fc_out(hidden).squeeze()

In [4]:

# 本代码由可视化策略环境自动生成 2021年12月18日 15:32
# 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 m33_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
    # train data
    train_data = input_1.read()
    x_train, x_val, y_train, y_val = train_test_split(train_data["x"], train_data['y'], shuffle=True, random_state=2021)
    # val data
    test_data = input_2.read()
    x_test = test_data["x"]
    
    model = GATModel(d_feat=98, hidden_size=64, num_layers=2, dropout=0.1, base_model="GRU")
    model.compile(device="cuda:0")
    model.fit(x_train, y_train, val_data=(x_val, y_val), batch_size=1024, epochs=10, verbose=1, num_workers=2)
    
    # model.fit(train_data["x"], train_data['y'], batch_size=1024, epochs=2, verbose=1, num_workers=2)
    output = model.predict(x_test)
    
    data_1 = DataSource.write_pickle(output)
    return Outputs(data_1=data_1, data_2=None, data_3=None)
# 后处理函数，可选。输入是主函数的输出，可以在这里对数据做处理，或者返回更友好的outputs数据格式。此函数输出不会被缓存。
def m33_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 m41_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[:], 'instrument':df.instrument, 'date':df.date})
    df.sort_values(['date','pred_label'],inplace=True, ascending=[True,False])
    return Outputs(data_1=DataSource.write_df(df), data_2=None, data_3=None)

# 后处理函数，可选。输入是主函数的输出，可以在这里对数据做处理，或者返回更友好的outputs数据格式。此函数输出不会被缓存。
def m41_post_run_bigquant_run(outputs):
    return outputs

# 回测引擎：初始化函数，只执行一次
def m42_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=0.001, sell_cost=0.001, 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 = 20
    # 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 m42_handle_data_bigquant_run(context, data):
    # 按日期过滤得到今日的预测数据
    ranker_prediction = context.ranker_prediction[
        context.ranker_prediction.date == data.current_dt.strftime('%Y-%m-%d')]

    # 1. 资金分配
    # 平均持仓时间是hold_days，每日都将买入股票，每日预期使用 1/hold_days 的资金
    # 实际操作中，会存在一定的买入误差，所以在前hold_days天，等量使用资金；之后，尽量使用剩余资金（这里设置最多用等量的1.5倍）
    is_staging = context.trading_day_index < context.options['hold_days'] # 是否在建仓期间（前 hold_days 天）
    cash_avg = context.portfolio.portfolio_value / context.options['hold_days']
    cash_for_buy = min(context.portfolio.cash, (1 if is_staging else 1.5) * cash_avg)
    cash_for_sell = cash_avg - (context.portfolio.cash - cash_for_buy)
    positions = {e.symbol: p.amount * p.last_sale_price
                 for e, p in context.perf_tracker.position_tracker.positions.items()}

    # 2. 生成卖出订单：hold_days天之后才开始卖出；对持仓的股票，按机器学习算法预测的排序末位淘汰
    if not is_staging and cash_for_sell > 0:
        equities = {e.symbol: e for e, p in context.perf_tracker.position_tracker.positions.items()}
        instruments = list(reversed(list(ranker_prediction.instrument[ranker_prediction.instrument.apply(
                lambda x: x in equities and not context.has_unfinished_sell_order(equities[x]))])))
        # print('rank order for sell %s' % instruments)
        for instrument in instruments:
            context.order_target(context.symbol(instrument), 0)
            cash_for_sell -= positions[instrument]
            if cash_for_sell <= 0:
                break

    # 3. 生成买入订单：按机器学习算法预测的排序，买入前面的stock_count只股票
    buy_cash_weights = context.stock_weights
    buy_instruments = list(ranker_prediction.instrument[:len(buy_cash_weights)])
    max_cash_per_instrument = context.portfolio.portfolio_value * context.max_cash_per_instrument
    for i, instrument in enumerate(buy_instruments):
        cash = cash_for_buy * buy_cash_weights[i]
        if cash > max_cash_per_instrument - positions.get(instrument, 0):
            # 确保股票持仓量不会超过每次股票最大的占用资金量
            cash = max_cash_per_instrument - positions.get(instrument, 0)
        if cash > 0:
            context.order_value(context.symbol(instrument), cash)

# 回测引擎：准备数据，只执行一次
def m42_prepare_bigquant_run(context):
    pass


m22 = M.instruments.v2(
    start_date='2018-01-01',
    end_date='2020-12-31',
    market='CN_STOCK_A',
    instrument_list=''
)

m23 = M.advanced_auto_labeler.v2(
    instruments=m22.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
)

m1 = M.standardlize.v8(
    input_1=m23.data,
    columns_input='label'
)

m24 = M.input_features.v1(
    features="""close_0
open_0
high_0
low_0 
amount_0
turn_0 
return_0
 
close_1
open_1
high_1
low_1
return_1
amount_1
turn_1
 
close_2
open_2
high_2
low_2
amount_2
turn_2
return_2
 
close_3
open_3
high_3
low_3
amount_3
turn_3
return_3
 
close_4
open_4
high_4
low_4
amount_4
turn_4
return_4
 
mean(close_0, 5)
mean(low_0, 5)
mean(open_0, 5)
mean(high_0, 5)
mean(turn_0, 5)
mean(amount_0, 5)
mean(return_0, 5)
 
ts_max(close_0, 5)
ts_max(low_0, 5)
ts_max(open_0, 5)
ts_max(high_0, 5)
ts_max(turn_0, 5)
ts_max(amount_0, 5)
ts_max(return_0, 5)
 
ts_min(close_0, 5)
ts_min(low_0, 5)
ts_min(open_0, 5)
ts_min(high_0, 5)
ts_min(turn_0, 5)
ts_min(amount_0, 5)
ts_min(return_0, 5) 
 
std(close_0, 5)
std(low_0, 5)
std(open_0, 5)
std(high_0, 5)
std(turn_0, 5)
std(amount_0, 5)
std(return_0, 5)
 
ts_rank(close_0, 5)
ts_rank(low_0, 5)
ts_rank(open_0, 5)
ts_rank(high_0, 5)
ts_rank(turn_0, 5)
ts_rank(amount_0, 5)
ts_rank(return_0, 5)
 
decay_linear(close_0, 5)
decay_linear(low_0, 5)
decay_linear(open_0, 5)
decay_linear(high_0, 5)
decay_linear(turn_0, 5)
decay_linear(amount_0, 5)
decay_linear(return_0, 5)
 
correlation(volume_0, return_0, 5)
correlation(volume_0, high_0, 5)
correlation(volume_0, low_0, 5)
correlation(volume_0, close_0, 5)
correlation(volume_0, open_0, 5)
correlation(volume_0, turn_0, 5)
  
correlation(return_0, high_0, 5)
correlation(return_0, low_0, 5)
correlation(return_0, close_0, 5)
correlation(return_0, open_0, 5)
correlation(return_0, turn_0, 5)
 
correlation(high_0, low_0, 5)
correlation(high_0, close_0, 5)
correlation(high_0, open_0, 5)
correlation(high_0, turn_0, 5)
 
correlation(low_0, close_0, 5)
correlation(low_0, open_0, 5)
correlation(low_0, turn_0, 5)
 
correlation(close_0, open_0, 5)
correlation(close_0, turn_0, 5)

correlation(open_0, turn_0, 5)"""
)

m27 = M.general_feature_extractor.v7(
    instruments=m22.data,
    features=m24.data,
    start_date='',
    end_date='',
    before_start_days=10
)

m3 = M.chinaa_stock_filter.v1(
    input_data=m27.data,
    index_constituent_cond=['中证500'],
    board_cond=['全部'],
    industry_cond=['全部'],
    st_cond=['全部'],
    delist_cond=['全部'],
    output_left_data=False
)

m28 = M.derived_feature_extractor.v3(
    input_data=m3.data,
    features=m24.data,
    date_col='date',
    instrument_col='instrument',
    drop_na=True,
    remove_extra_columns=False
)

m34 = M.standardlize.v8(
    input_1=m28.data,
    input_2=m24.data,
    columns_input='[]'
)

m35 = M.fillnan.v1(
    input_data=m34.data,
    features=m24.data,
    fill_value='0.0'
)

m25 = M.join.v3(
    data1=m1.data,
    data2=m35.data,
    on='date,instrument',
    how='inner',
    sort=False
)

m2 = M.dl_convert_to_bin.v2(
    input_data=m25.data,
    features=m24.data,
    window_size=5,
    feature_clip=3,
    flatten=False,
    window_along_col='instrument'
)

m26 = M.instruments.v2(
    start_date='2021-01-01',
    end_date='2021-11-01',
    market='CN_STOCK_A',
    instrument_list='',
    max_count=0
)

m29 = M.general_feature_extractor.v7(
    instruments=m26.data,
    features=m24.data,
    start_date='',
    end_date='',
    before_start_days=10
)

m4 = M.chinaa_stock_filter.v1(
    input_data=m29.data,
    index_constituent_cond=['中证500'],
    board_cond=['全部'],
    industry_cond=['全部'],
    st_cond=['全部'],
    delist_cond=['全部'],
    output_left_data=False
)

m30 = M.derived_feature_extractor.v3(
    input_data=m4.data,
    features=m24.data,
    date_col='date',
    instrument_col='instrument',
    drop_na=True,
    remove_extra_columns=False
)

m37 = M.standardlize.v8(
    input_1=m30.data,
    input_2=m24.data,
    columns_input='[]'
)

m36 = M.fillnan.v1(
    input_data=m37.data,
    features=m24.data,
    fill_value='0.0'
)

m32 = M.dl_convert_to_bin.v2(
    input_data=m36.data,
    features=m24.data,
    window_size=5,
    feature_clip=3,
    flatten=False,
    window_along_col='instrument'
)

m33 = M.cached.v3(
    input_1=m2.data,
    input_2=m32.data,
    run=m33_run_bigquant_run,
    post_run=m33_post_run_bigquant_run,
    input_ports='',
    params='{}',
    output_ports='',
    m_cached=False
)

m41 = M.cached.v3(
    input_1=m33.data_1,
    input_2=m30.data,
    run=m41_run_bigquant_run,
    post_run=m41_post_run_bigquant_run,
    input_ports='',
    params='{}',
    output_ports='',
    m_cached=False
)

m42 = M.trade.v4(
    instruments=m26.data,
    options_data=m41.data_1,
    start_date='',
    end_date='',
    initialize=m42_initialize_bigquant_run,
    handle_data=m42_handle_data_bigquant_run,
    prepare=m42_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='后复权',
    product_type='股票',
    plot_charts=True,
    backtest_only=False,
    benchmark='000300.SHA'
)

[2021-12-08 18:35:34.162179] INFO: moduleinvoker: instruments.v2 开始运行..

[2021-12-08 18:35:35.665534] INFO: moduleinvoker: instruments.v2 运行完成[1.503377s].

[2021-12-08 18:35:35.749840] INFO: moduleinvoker: advanced_auto_labeler.v2 开始运行..

[2021-12-08 18:35:42.668375] INFO: 自动标注(股票): 加载历史数据: 2647809 行

[2021-12-08 18:35:42.672912] INFO: 自动标注(股票): 开始标注 ..

[2021-12-08 18:35:59.237544] INFO: moduleinvoker: advanced_auto_labeler.v2 运行完成[23.487706s].

[2021-12-08 18:35:59.296786] INFO: moduleinvoker: standardlize.v8 开始运行..

[2021-12-08 18:37:04.693031] INFO: moduleinvoker: standardlize.v8 运行完成[65.396251s].

[2021-12-08 18:37:04.713108] INFO: moduleinvoker: input_features.v1 开始运行..

[2021-12-08 18:37:04.726718] INFO: moduleinvoker: 命中缓存

[2021-12-08 18:37:04.730649] INFO: moduleinvoker: input_features.v1 运行完成[0.01758s].

[2021-12-08 18:37:04.769771] INFO: moduleinvoker: general_feature_extractor.v7 开始运行..

[2021-12-08 18:37:16.849540] INFO: 基础特征抽取: 年份 2017, 特征行数=19477

[2021-12-08 18:39:30.260325] INFO: 基础特征抽取: 年份 2018, 特征行数=816987

[2021-12-08 18:41:58.815400] INFO: 基础特征抽取: 年份 2019, 特征行数=884867

[2021-12-08 18:43:23.526887] INFO: 基础特征抽取: 年份 2020, 特征行数=945961

[2021-12-08 18:43:24.342805] INFO: 基础特征抽取: 总行数: 2667292

[2021-12-08 18:43:24.373377] INFO: moduleinvoker: general_feature_extractor.v7 运行完成[379.603641s].

[2021-12-08 18:43:24.506675] INFO: moduleinvoker: chinaa_stock_filter.v1 开始运行..

[2021-12-08 18:43:25.867761] INFO: A股股票过滤: 过滤 /y_2017, 2813/0/19477

[2021-12-08 18:43:52.048517] INFO: A股股票过滤: 过滤 /y_2018, 115585/0/816987

[2021-12-08 18:44:35.911794] INFO: A股股票过滤: 过滤 /y_2019, 121771/0/884867

[2021-12-08 18:45:22.103120] INFO: A股股票过滤: 过滤 /y_2020, 121275/0/945961

[2021-12-08 18:45:22.114820] INFO: A股股票过滤: 过滤完成, 361444 + 0

[2021-12-08 18:45:22.178525] INFO: moduleinvoker: chinaa_stock_filter.v1 运行完成[117.671824s].

[2021-12-08 18:45:22.309935] INFO: moduleinvoker: derived_feature_extractor.v3 开始运行..

[2021-12-08 18:45:24.899429] INFO: derived_feature_extractor: 提取完成 mean(close_0, 5), 0.378s

[2021-12-08 18:45:25.541743] INFO: derived_feature_extractor: 提取完成 mean(low_0, 5), 0.640s

[2021-12-08 18:45:25.891809] INFO: derived_feature_extractor: 提取完成 mean(open_0, 5), 0.348s

[2021-12-08 18:45:26.922156] INFO: derived_feature_extractor: 提取完成 mean(high_0, 5), 1.027s

[2021-12-08 18:45:27.280453] INFO: derived_feature_extractor: 提取完成 mean(turn_0, 5), 0.356s

[2021-12-08 18:45:27.898101] INFO: derived_feature_extractor: 提取完成 mean(amount_0, 5), 0.615s

[2021-12-08 18:45:28.170643] INFO: derived_feature_extractor: 提取完成 mean(return_0, 5), 0.270s

[2021-12-08 18:45:28.534006] INFO: derived_feature_extractor: 提取完成 ts_max(close_0, 5), 0.362s

[2021-12-08 18:45:28.829424] INFO: derived_feature_extractor: 提取完成 ts_max(low_0, 5), 0.293s

[2021-12-08 18:45:29.145836] INFO: derived_feature_extractor: 提取完成 ts_max(open_0, 5), 0.314s

[2021-12-08 18:45:29.475467] INFO: derived_feature_extractor: 提取完成 ts_max(high_0, 5), 0.328s

[2021-12-08 18:45:29.880952] INFO: derived_feature_extractor: 提取完成 ts_max(turn_0, 5), 0.388s

[2021-12-08 18:45:30.295385] INFO: derived_feature_extractor: 提取完成 ts_max(amount_0, 5), 0.388s

[2021-12-08 18:45:30.672768] INFO: derived_feature_extractor: 提取完成 ts_max(return_0, 5), 0.374s

[2021-12-08 18:45:31.051408] INFO: derived_feature_extractor: 提取完成 ts_min(close_0, 5), 0.375s

[2021-12-08 18:45:31.428206] INFO: derived_feature_extractor: 提取完成 ts_min(low_0, 5), 0.372s

[2021-12-08 18:45:31.809638] INFO: derived_feature_extractor: 提取完成 ts_min(open_0, 5), 0.377s

[2021-12-08 18:45:32.192781] INFO: derived_feature_extractor: 提取完成 ts_min(high_0, 5), 0.377s

[2021-12-08 18:45:32.590829] INFO: derived_feature_extractor: 提取完成 ts_min(turn_0, 5), 0.383s

[2021-12-08 18:45:32.922991] INFO: derived_feature_extractor: 提取完成 ts_min(amount_0, 5), 0.328s

[2021-12-08 18:45:33.297381] INFO: derived_feature_extractor: 提取完成 ts_min(return_0, 5), 0.372s

[2021-12-08 18:45:33.655490] INFO: derived_feature_extractor: 提取完成 std(close_0, 5), 0.355s

[2021-12-08 18:45:33.963414] INFO: derived_feature_extractor: 提取完成 std(low_0, 5), 0.306s

[2021-12-08 18:45:34.765797] INFO: derived_feature_extractor: 提取完成 std(open_0, 5), 0.798s

[2021-12-08 18:45:35.246016] INFO: derived_feature_extractor: 提取完成 std(high_0, 5), 0.433s

[2021-12-08 18:45:36.112846] INFO: derived_feature_extractor: 提取完成 std(turn_0, 5), 0.853s

[2021-12-08 18:45:36.455234] INFO: derived_feature_extractor: 提取完成 std(amount_0, 5), 0.341s

[2021-12-08 18:45:36.784415] INFO: derived_feature_extractor: 提取完成 std(return_0, 5), 0.327s

[2021-12-08 18:45:38.234140] INFO: derived_feature_extractor: 提取完成 ts_rank(close_0, 5), 1.448s

[2021-12-08 18:45:40.160120] INFO: derived_feature_extractor: 提取完成 ts_rank(low_0, 5), 1.924s

[2021-12-08 18:45:41.964863] INFO: derived_feature_extractor: 提取完成 ts_rank(open_0, 5), 1.803s

[2021-12-08 18:45:43.466923] INFO: derived_feature_extractor: 提取完成 ts_rank(high_0, 5), 1.500s

[2021-12-08 18:45:45.094152] INFO: derived_feature_extractor: 提取完成 ts_rank(turn_0, 5), 1.625s

[2021-12-08 18:45:46.758337] INFO: derived_feature_extractor: 提取完成 ts_rank(amount_0, 5), 1.662s

[2021-12-08 18:45:48.311756] INFO: derived_feature_extractor: 提取完成 ts_rank(return_0, 5), 1.551s

[2021-12-08 18:45:49.391597] INFO: derived_feature_extractor: 提取完成 decay_linear(close_0, 5), 1.078s

[2021-12-08 18:45:50.445079] INFO: derived_feature_extractor: 提取完成 decay_linear(low_0, 5), 1.051s

[2021-12-08 18:45:51.461807] INFO: derived_feature_extractor: 提取完成 decay_linear(open_0, 5), 1.015s

[2021-12-08 18:45:52.510593] INFO: derived_feature_extractor: 提取完成 decay_linear(high_0, 5), 1.046s

[2021-12-08 18:45:53.620522] INFO: derived_feature_extractor: 提取完成 decay_linear(turn_0, 5), 1.103s

[2021-12-08 18:45:54.698793] INFO: derived_feature_extractor: 提取完成 decay_linear(amount_0, 5), 1.074s

[2021-12-08 18:45:55.922694] INFO: derived_feature_extractor: 提取完成 decay_linear(return_0, 5), 1.218s

[2021-12-08 18:46:02.416047] INFO: derived_feature_extractor: 提取完成 correlation(volume_0, return_0, 5), 6.491s

[2021-12-08 18:46:09.356128] INFO: derived_feature_extractor: 提取完成 correlation(volume_0, high_0, 5), 6.921s

[2021-12-08 18:46:16.893555] INFO: derived_feature_extractor: 提取完成 correlation(volume_0, low_0, 5), 7.535s

[2021-12-08 18:46:24.082550] INFO: derived_feature_extractor: 提取完成 correlation(volume_0, close_0, 5), 7.186s

[2021-12-08 18:46:31.046405] INFO: derived_feature_extractor: 提取完成 correlation(volume_0, open_0, 5), 6.959s

[2021-12-08 18:46:38.038806] INFO: derived_feature_extractor: 提取完成 correlation(volume_0, turn_0, 5), 6.989s

[2021-12-08 18:46:45.106680] INFO: derived_feature_extractor: 提取完成 correlation(return_0, high_0, 5), 7.065s

[2021-12-08 18:46:51.916499] INFO: derived_feature_extractor: 提取完成 correlation(return_0, low_0, 5), 6.807s

[2021-12-08 18:46:58.594911] INFO: derived_feature_extractor: 提取完成 correlation(return_0, close_0, 5), 6.676s

[2021-12-08 18:47:06.534193] INFO: derived_feature_extractor: 提取完成 correlation(return_0, open_0, 5), 7.930s

[2021-12-08 18:47:14.611706] INFO: derived_feature_extractor: 提取完成 correlation(return_0, turn_0, 5), 8.075s

[2021-12-08 18:47:22.327145] INFO: derived_feature_extractor: 提取完成 correlation(high_0, low_0, 5), 7.713s

[2021-12-08 18:47:30.122393] INFO: derived_feature_extractor: 提取完成 correlation(high_0, close_0, 5), 7.793s

[2021-12-08 18:47:37.210419] INFO: derived_feature_extractor: 提取完成 correlation(high_0, open_0, 5), 7.086s

[2021-12-08 18:47:44.104590] INFO: derived_feature_extractor: 提取完成 correlation(high_0, turn_0, 5), 6.889s

[2021-12-08 18:47:50.961833] INFO: derived_feature_extractor: 提取完成 correlation(low_0, close_0, 5), 6.854s

[2021-12-08 18:47:58.399665] INFO: derived_feature_extractor: 提取完成 correlation(low_0, open_0, 5), 7.435s

[2021-12-08 18:48:05.073024] INFO: derived_feature_extractor: 提取完成 correlation(low_0, turn_0, 5), 6.671s

[2021-12-08 18:48:12.439489] INFO: derived_feature_extractor: 提取完成 correlation(close_0, open_0, 5), 7.359s

[2021-12-08 18:48:19.986197] INFO: derived_feature_extractor: 提取完成 correlation(close_0, turn_0, 5), 7.544s

[2021-12-08 18:48:26.792925] INFO: derived_feature_extractor: 提取完成 correlation(open_0, turn_0, 5), 6.805s

[2021-12-08 18:48:44.987070] INFO: derived_feature_extractor: /y_2017, 2813

[2021-12-08 18:48:45.697383] INFO: derived_feature_extractor: /y_2018, 115585

[2021-12-08 18:48:56.816059] INFO: derived_feature_extractor: /y_2019, 121771

[2021-12-08 18:49:00.641762] INFO: derived_feature_extractor: /y_2020, 121275

[2021-12-08 18:49:08.899595] INFO: moduleinvoker: derived_feature_extractor.v3 运行完成[226.589642s].

[2021-12-08 18:49:09.068309] INFO: moduleinvoker: standardlize.v8 开始运行..

[2021-12-08 18:52:42.902794] INFO: moduleinvoker: standardlize.v8 运行完成[213.834474s].

[2021-12-08 18:52:43.014984] INFO: moduleinvoker: fillnan.v1 开始运行..

[2021-12-08 18:53:46.080296] INFO: moduleinvoker: fillnan.v1 运行完成[63.065306s].

[2021-12-08 18:53:46.187809] INFO: moduleinvoker: join.v3 开始运行..

[2021-12-08 18:54:26.296627] INFO: join: /data, 行数=353942/358185, 耗时=14.554688s

[2021-12-08 18:54:26.429239] INFO: join: 最终行数: 353942

[2021-12-08 18:54:26.462578] INFO: moduleinvoker: join.v3 运行完成[40.274776s].

[2021-12-08 18:54:26.565483] INFO: moduleinvoker: dl_convert_to_bin.v2 开始运行..

[2021-12-08 18:55:39.420132] INFO: moduleinvoker: dl_convert_to_bin.v2 运行完成[72.854723s].

[2021-12-08 18:55:39.479538] INFO: moduleinvoker: instruments.v2 开始运行..

[2021-12-08 18:55:39.502662] INFO: moduleinvoker: 命中缓存

[2021-12-08 18:55:39.504851] INFO: moduleinvoker: instruments.v2 运行完成[0.025325s].

[2021-12-08 18:55:39.533326] INFO: moduleinvoker: general_feature_extractor.v7 开始运行..

[2021-12-08 18:55:45.068428] INFO: 基础特征抽取: 年份 2020, 特征行数=32852

[2021-12-08 18:55:53.326105] INFO: 基础特征抽取: 年份 2021, 特征行数=858486

[2021-12-08 18:55:53.469715] INFO: 基础特征抽取: 总行数: 891338

[2021-12-08 18:55:53.480508] INFO: moduleinvoker: general_feature_extractor.v7 运行完成[13.9472s].

[2021-12-08 18:55:53.510980] INFO: moduleinvoker: chinaa_stock_filter.v1 开始运行..

[2021-12-08 18:55:54.034809] INFO: A股股票过滤: 过滤 /y_2020, 3998/0/32852

[2021-12-08 18:55:56.697570] INFO: A股股票过滤: 过滤 /y_2021, 99330/0/858486

[2021-12-08 18:55:56.706232] INFO: A股股票过滤: 过滤完成, 103328 + 0

[2021-12-08 18:55:56.775655] INFO: moduleinvoker: chinaa_stock_filter.v1 运行完成[3.264664s].

[2021-12-08 18:55:56.788115] INFO: moduleinvoker: derived_feature_extractor.v3 开始运行..

[2021-12-08 18:55:57.588827] INFO: derived_feature_extractor: 提取完成 mean(close_0, 5), 0.113s

[2021-12-08 18:55:57.694421] INFO: derived_feature_extractor: 提取完成 mean(low_0, 5), 0.104s

[2021-12-08 18:55:57.810368] INFO: derived_feature_extractor: 提取完成 mean(open_0, 5), 0.114s

[2021-12-08 18:55:57.932423] INFO: derived_feature_extractor: 提取完成 mean(high_0, 5), 0.120s

[2021-12-08 18:55:58.054916] INFO: derived_feature_extractor: 提取完成 mean(turn_0, 5), 0.120s

[2021-12-08 18:55:58.166674] INFO: derived_feature_extractor: 提取完成 mean(amount_0, 5), 0.110s

[2021-12-08 18:55:58.285795] INFO: derived_feature_extractor: 提取完成 mean(return_0, 5), 0.116s

[2021-12-08 18:55:58.446523] INFO: derived_feature_extractor: 提取完成 ts_max(close_0, 5), 0.159s

[2021-12-08 18:55:58.614368] INFO: derived_feature_extractor: 提取完成 ts_max(low_0, 5), 0.165s

[2021-12-08 18:55:58.770950] INFO: derived_feature_extractor: 提取完成 ts_max(open_0, 5), 0.154s

[2021-12-08 18:55:58.901794] INFO: derived_feature_extractor: 提取完成 ts_max(high_0, 5), 0.129s

[2021-12-08 18:55:59.125903] INFO: derived_feature_extractor: 提取完成 ts_max(turn_0, 5), 0.222s

[2021-12-08 18:55:59.250891] INFO: derived_feature_extractor: 提取完成 ts_max(amount_0, 5), 0.122s

[2021-12-08 18:55:59.409233] INFO: derived_feature_extractor: 提取完成 ts_max(return_0, 5), 0.156s

[2021-12-08 18:55:59.574073] INFO: derived_feature_extractor: 提取完成 ts_min(close_0, 5), 0.160s

[2021-12-08 18:55:59.730447] INFO: derived_feature_extractor: 提取完成 ts_min(low_0, 5), 0.154s

[2021-12-08 18:55:59.886494] INFO: derived_feature_extractor: 提取完成 ts_min(open_0, 5), 0.153s

[2021-12-08 18:56:00.048239] INFO: derived_feature_extractor: 提取完成 ts_min(high_0, 5), 0.158s

[2021-12-08 18:56:00.277949] INFO: derived_feature_extractor: 提取完成 ts_min(turn_0, 5), 0.227s

[2021-12-08 18:56:00.389884] INFO: derived_feature_extractor: 提取完成 ts_min(amount_0, 5), 0.109s

[2021-12-08 18:56:00.491411] INFO: derived_feature_extractor: 提取完成 ts_min(return_0, 5), 0.100s

[2021-12-08 18:56:00.587624] INFO: derived_feature_extractor: 提取完成 std(close_0, 5), 0.095s

[2021-12-08 18:56:00.743509] INFO: derived_feature_extractor: 提取完成 std(low_0, 5), 0.154s

[2021-12-08 18:56:00.902096] INFO: derived_feature_extractor: 提取完成 std(open_0, 5), 0.156s

[2021-12-08 18:56:01.056073] INFO: derived_feature_extractor: 提取完成 std(high_0, 5), 0.152s

[2021-12-08 18:56:01.215096] INFO: derived_feature_extractor: 提取完成 std(turn_0, 5), 0.157s

[2021-12-08 18:56:01.367804] INFO: derived_feature_extractor: 提取完成 std(amount_0, 5), 0.150s

[2021-12-08 18:56:01.510937] INFO: derived_feature_extractor: 提取完成 std(return_0, 5), 0.141s

[2021-12-08 18:56:01.931699] INFO: derived_feature_extractor: 提取完成 ts_rank(close_0, 5), 0.418s

[2021-12-08 18:56:02.378898] INFO: derived_feature_extractor: 提取完成 ts_rank(low_0, 5), 0.445s

[2021-12-08 18:56:02.994321] INFO: derived_feature_extractor: 提取完成 ts_rank(open_0, 5), 0.612s

[2021-12-08 18:56:03.496569] INFO: derived_feature_extractor: 提取完成 ts_rank(high_0, 5), 0.499s

[2021-12-08 18:56:04.060903] INFO: derived_feature_extractor: 提取完成 ts_rank(turn_0, 5), 0.562s

[2021-12-08 18:56:04.629201] INFO: derived_feature_extractor: 提取完成 ts_rank(amount_0, 5), 0.566s

[2021-12-08 18:56:05.115930] INFO: derived_feature_extractor: 提取完成 ts_rank(return_0, 5), 0.482s

[2021-12-08 18:56:05.517809] INFO: derived_feature_extractor: 提取完成 decay_linear(close_0, 5), 0.398s

[2021-12-08 18:56:05.891405] INFO: derived_feature_extractor: 提取完成 decay_linear(low_0, 5), 0.371s

[2021-12-08 18:56:06.215948] INFO: derived_feature_extractor: 提取完成 decay_linear(open_0, 5), 0.322s

[2021-12-08 18:56:06.599150] INFO: derived_feature_extractor: 提取完成 decay_linear(high_0, 5), 0.380s

[2021-12-08 18:56:06.984284] INFO: derived_feature_extractor: 提取完成 decay_linear(turn_0, 5), 0.382s

[2021-12-08 18:56:07.328263] INFO: derived_feature_extractor: 提取完成 decay_linear(amount_0, 5), 0.340s

[2021-12-08 18:56:07.660341] INFO: derived_feature_extractor: 提取完成 decay_linear(return_0, 5), 0.328s

[2021-12-08 18:56:12.604491] INFO: derived_feature_extractor: 提取完成 correlation(volume_0, return_0, 5), 4.942s

[2021-12-08 18:56:17.719978] INFO: derived_feature_extractor: 提取完成 correlation(volume_0, high_0, 5), 5.111s

[2021-12-08 18:56:22.804369] INFO: derived_feature_extractor: 提取完成 correlation(volume_0, low_0, 5), 5.081s

[2021-12-08 18:56:27.728391] INFO: derived_feature_extractor: 提取完成 correlation(volume_0, close_0, 5), 4.922s

[2021-12-08 18:56:32.302315] INFO: derived_feature_extractor: 提取完成 correlation(volume_0, open_0, 5), 4.572s

[2021-12-08 18:56:36.927372] INFO: derived_feature_extractor: 提取完成 correlation(volume_0, turn_0, 5), 4.623s

[2021-12-08 18:56:41.498291] INFO: derived_feature_extractor: 提取完成 correlation(return_0, high_0, 5), 4.569s

[2021-12-08 18:56:46.531843] INFO: derived_feature_extractor: 提取完成 correlation(return_0, low_0, 5), 5.031s

[2021-12-08 18:56:51.889082] INFO: derived_feature_extractor: 提取完成 correlation(return_0, close_0, 5), 5.354s

[2021-12-08 18:56:57.426880] INFO: derived_feature_extractor: 提取完成 correlation(return_0, open_0, 5), 5.535s

[2021-12-08 18:57:03.001423] INFO: derived_feature_extractor: 提取完成 correlation(return_0, turn_0, 5), 5.572s

[2021-12-08 18:57:08.043295] INFO: derived_feature_extractor: 提取完成 correlation(high_0, low_0, 5), 5.025s

[2021-12-08 18:57:12.580052] INFO: derived_feature_extractor: 提取完成 correlation(high_0, close_0, 5), 4.535s

[2021-12-08 18:57:17.384471] INFO: derived_feature_extractor: 提取完成 correlation(high_0, open_0, 5), 4.801s

[2021-12-08 18:57:21.841010] INFO: derived_feature_extractor: 提取完成 correlation(high_0, turn_0, 5), 4.455s

[2021-12-08 18:57:26.535419] INFO: derived_feature_extractor: 提取完成 correlation(low_0, close_0, 5), 4.690s

[2021-12-08 18:57:31.309682] INFO: derived_feature_extractor: 提取完成 correlation(low_0, open_0, 5), 4.769s

[2021-12-08 18:57:35.701492] INFO: derived_feature_extractor: 提取完成 correlation(low_0, turn_0, 5), 4.390s

[2021-12-08 18:57:40.416286] INFO: derived_feature_extractor: 提取完成 correlation(close_0, open_0, 5), 4.713s

[2021-12-08 18:57:44.819902] INFO: derived_feature_extractor: 提取完成 correlation(close_0, turn_0, 5), 4.401s

[2021-12-08 18:57:49.549750] INFO: derived_feature_extractor: 提取完成 correlation(open_0, turn_0, 5), 4.727s

[2021-12-08 18:57:54.261725] INFO: derived_feature_extractor: /y_2020, 3998

[2021-12-08 18:57:54.863185] INFO: derived_feature_extractor: /y_2021, 99330

[2021-12-08 18:58:28.627144] INFO: moduleinvoker: derived_feature_extractor.v3 运行完成[151.839012s].

[2021-12-08 18:58:28.714651] INFO: moduleinvoker: standardlize.v8 开始运行..

[2021-12-08 18:59:03.038677] INFO: moduleinvoker: standardlize.v8 运行完成[34.324026s].

[2021-12-08 18:59:03.081042] INFO: moduleinvoker: fillnan.v1 开始运行..

[2021-12-08 18:59:05.822271] INFO: moduleinvoker: fillnan.v1 运行完成[2.741217s].

[2021-12-08 18:59:05.850665] INFO: moduleinvoker: dl_convert_to_bin.v2 开始运行..

[2021-12-08 18:59:15.627262] INFO: moduleinvoker: dl_convert_to_bin.v2 运行完成[9.776616s].

[2021-12-08 18:59:15.648118] INFO: moduleinvoker: cached.v3 开始运行..

[2021-12-08 19:02:33.256510] INFO: moduleinvoker: cached.v3 运行完成[197.608355s].

[2021-12-08 19:02:33.277429] INFO: moduleinvoker: cached.v3 开始运行..

[2021-12-08 19:02:38.859297] INFO: moduleinvoker: cached.v3 运行完成[5.581888s].

[2021-12-08 19:02:41.703605] INFO: moduleinvoker: backtest.v8 开始运行..

[2021-12-08 19:02:41.710177] INFO: backtest: biglearning backtest:V8.6.0

[2021-12-08 19:02:41.712238] INFO: backtest: product_type:stock by specified

[2021-12-08 19:02:41.984124] INFO: moduleinvoker: cached.v2 开始运行..

[2021-12-08 19:03:13.709848] INFO: backtest: 读取股票行情完成:1943228

[2021-12-08 19:03:24.682135] INFO: moduleinvoker: cached.v2 运行完成[42.698013s].

[2021-12-08 19:03:32.186943] INFO: algo: TradingAlgorithm V1.8.5

[2021-12-08 19:03:34.567562] INFO: algo: trading transform...

[2021-12-08 19:03:49.020919] INFO: Performance: Simulated 199 trading days out of 199.

[2021-12-08 19:03:49.023920] INFO: Performance: first open: 2021-01-04 09:30:00+00:00

[2021-12-08 19:03:49.026302] INFO: Performance: last close: 2021-11-01 15:00:00+00:00

[2021-12-08 19:04:12.094728] INFO: moduleinvoker: backtest.v8 运行完成[90.391131s].

[2021-12-08 19:04:12.098664] INFO: moduleinvoker: trade.v4 运行完成[93.188994s].

epoch 0   |  train_loss 0.82538|  vall_loss 0.81766|  0:00:19s
epoch 1   |  train_loss 0.82187|  vall_loss 0.81627|  0:00:31s
epoch 2   |  train_loss 0.82091|  vall_loss 0.81366|  0:00:44s
epoch 3   |  train_loss 0.81965|  vall_loss 0.81315|  0:00:56s
epoch 4   |  train_loss 0.81726|  vall_loss 0.81356|  0:01:08s
early stop
best loss: 0.813150 @ 3

收益率32.17%
年化收益率42.36%
基准收益率-6.15%
阿尔法0.51
贝塔0.63
夏普比率1.41
胜率0.53
盈亏比1.28
收益波动率25.2%
信息比率0.12
最大回撤16.31%

bigcharts-data-start/{"__type":"tabs","__id":"bigchart-a0661f8539cc421a969d99c2640eb101"}/bigcharts-data-end

In [ ]: