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 [3]:

# 本代码由可视化策略环境自动生成 2023年2月14日 14:09
# 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'
)

[2023-02-14 14:07:55.529641] INFO: moduleinvoker: instruments.v2 开始运行..

[2023-02-14 14:07:55.541947] INFO: moduleinvoker: 命中缓存

[2023-02-14 14:07:55.544134] INFO: moduleinvoker: instruments.v2 运行完成[0.014516s].

[2023-02-14 14:07:55.553888] INFO: moduleinvoker: advanced_auto_labeler.v2 开始运行..

[2023-02-14 14:07:55.561271] INFO: moduleinvoker: 命中缓存

[2023-02-14 14:07:57.254713] INFO: moduleinvoker: advanced_auto_labeler.v2 运行完成[1.700817s].

[2023-02-14 14:07:57.264280] INFO: moduleinvoker: standardlize.v8 开始运行..

[2023-02-14 14:07:57.273339] INFO: moduleinvoker: 命中缓存

[2023-02-14 14:07:57.275604] INFO: moduleinvoker: standardlize.v8 运行完成[0.011325s].

[2023-02-14 14:07:57.283424] INFO: moduleinvoker: input_features.v1 开始运行..

[2023-02-14 14:07:57.291590] INFO: moduleinvoker: 命中缓存

[2023-02-14 14:07:57.293595] INFO: moduleinvoker: input_features.v1 运行完成[0.010188s].

[2023-02-14 14:07:57.339003] INFO: moduleinvoker: general_feature_extractor.v7 开始运行..

[2023-02-14 14:07:57.346844] INFO: moduleinvoker: 命中缓存

[2023-02-14 14:07:57.349222] INFO: moduleinvoker: general_feature_extractor.v7 运行完成[0.01024s].

[2023-02-14 14:07:57.362284] INFO: moduleinvoker: chinaa_stock_filter.v1 开始运行..

[2023-02-14 14:07:57.369665] INFO: moduleinvoker: 命中缓存

[2023-02-14 14:07:57.371886] INFO: moduleinvoker: chinaa_stock_filter.v1 运行完成[0.009611s].

[2023-02-14 14:07:57.382963] INFO: moduleinvoker: derived_feature_extractor.v3 开始运行..

[2023-02-14 14:07:57.390292] INFO: moduleinvoker: 命中缓存

[2023-02-14 14:07:57.392073] INFO: moduleinvoker: derived_feature_extractor.v3 运行完成[0.00912s].

[2023-02-14 14:07:57.397767] INFO: moduleinvoker: standardlize.v8 开始运行..

[2023-02-14 14:07:57.413434] INFO: moduleinvoker: 命中缓存

[2023-02-14 14:07:57.415587] INFO: moduleinvoker: standardlize.v8 运行完成[0.017816s].

[2023-02-14 14:07:57.427901] INFO: moduleinvoker: fillnan.v1 开始运行..

[2023-02-14 14:07:57.434017] INFO: moduleinvoker: 命中缓存

[2023-02-14 14:07:57.436201] INFO: moduleinvoker: fillnan.v1 运行完成[0.008301s].

[2023-02-14 14:07:57.448415] INFO: moduleinvoker: join.v3 开始运行..

[2023-02-14 14:07:57.455866] INFO: moduleinvoker: 命中缓存

[2023-02-14 14:07:57.457970] INFO: moduleinvoker: join.v3 运行完成[0.009557s].

[2023-02-14 14:07:57.480720] INFO: moduleinvoker: dl_convert_to_bin.v2 开始运行..

[2023-02-14 14:07:57.487375] INFO: moduleinvoker: 命中缓存

[2023-02-14 14:07:57.489764] INFO: moduleinvoker: dl_convert_to_bin.v2 运行完成[0.009058s].

[2023-02-14 14:07:57.496563] INFO: moduleinvoker: instruments.v2 开始运行..

[2023-02-14 14:07:57.502718] INFO: moduleinvoker: 命中缓存

[2023-02-14 14:07:57.505043] INFO: moduleinvoker: instruments.v2 运行完成[0.008487s].

[2023-02-14 14:07:57.525474] INFO: moduleinvoker: general_feature_extractor.v7 开始运行..

[2023-02-14 14:07:57.531673] INFO: moduleinvoker: 命中缓存

[2023-02-14 14:07:57.533767] INFO: moduleinvoker: general_feature_extractor.v7 运行完成[0.008315s].

[2023-02-14 14:07:57.543785] INFO: moduleinvoker: chinaa_stock_filter.v1 开始运行..

[2023-02-14 14:07:57.556330] INFO: moduleinvoker: 命中缓存

[2023-02-14 14:07:57.558269] INFO: moduleinvoker: chinaa_stock_filter.v1 运行完成[0.014494s].

[2023-02-14 14:07:57.566387] INFO: moduleinvoker: derived_feature_extractor.v3 开始运行..

[2023-02-14 14:07:57.577232] INFO: moduleinvoker: 命中缓存

[2023-02-14 14:07:57.578911] INFO: moduleinvoker: derived_feature_extractor.v3 运行完成[0.012522s].

[2023-02-14 14:07:57.585003] INFO: moduleinvoker: standardlize.v8 开始运行..

[2023-02-14 14:07:57.598595] INFO: moduleinvoker: 命中缓存

[2023-02-14 14:07:57.601206] INFO: moduleinvoker: standardlize.v8 运行完成[0.016204s].

[2023-02-14 14:07:57.611753] INFO: moduleinvoker: fillnan.v1 开始运行..

[2023-02-14 14:07:57.620322] INFO: moduleinvoker: 命中缓存

[2023-02-14 14:07:57.622162] INFO: moduleinvoker: fillnan.v1 运行完成[0.010417s].

[2023-02-14 14:07:57.636555] INFO: moduleinvoker: dl_convert_to_bin.v2 开始运行..

[2023-02-14 14:07:57.645017] INFO: moduleinvoker: 命中缓存

[2023-02-14 14:07:57.647057] INFO: moduleinvoker: dl_convert_to_bin.v2 运行完成[0.010519s].

[2023-02-14 14:07:57.659893] INFO: moduleinvoker: cached.v3 开始运行..

[2023-02-14 14:08:00.332174] ERROR: moduleinvoker: module name: cached, module version: v3, trackeback: ValueError: optimizer function should't be none.

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
<ipython-input-3-d63e108d3e66> in <module>
    367 )
    368 
--> 369 m33 = M.cached.v3(
    370     input_1=m2.data,
    371     input_2=m32.data,

<ipython-input-3-d63e108d3e66> in m33_run_bigquant_run(input_1, input_2, input_3)
     15 
     16     model = GATModel(d_feat=98, hidden_size=64, num_layers=2, dropout=0.1, base_model="GRU")
---> 17     model.compile(device="cuda:0")
     18     model.fit(x_train, y_train, val_data=(x_val, y_val), batch_size=1024, epochs=10, verbose=1, num_workers=2)
     19 

ValueError: optimizer function should't be none.

In [ ]: