复制链接

克隆策略

In [2]:

# 本代码由可视化策略环境自动生成 2021年12月22日 12:53
# 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 m6_run_bigquant_run(input_1, input_2, input_3):
    # 示例代码如下。在这里编写您的代码
    df = input_1.read_df()
    ins = m1.data.read_pickle()['instruments']
    start = m1.data.read_pickle()['start_date']
    end = m1.data.read_pickle()['end_date']
    
    df1 = D.features(ins,start,end,fields=['rank_turn_0','rank_amount_0','st_status_0'])
    df_final = pd.merge(df,df1,on=['date','instrument'])
    df_final = df_final[df_final['instrument'].str.startswith("688") == False]
    df_final = df_final[df_final['instrument'].str.startswith("3") == False]

    df_final = df_final[df_final["st_status_0"] == 0]
    df_final = df_final[df_final['rank_turn_0'] >= 0.9]
    df_final = df_final[df_final['rank_amount_0'] >= 0.85]
    print("用于训练的样本总个数为",len(df_final))
    print(df_final.iloc[0])
    data_1 = DataSource.write_df(df_final)
    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 m6_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 m8_run_bigquant_run(input_1, input_2, input_3):
    # 示例代码如下。在这里编写您的代码
    df = input_1.read_df()
    ins = m9.data.read_pickle()['instruments']
    start = m9.data.read_pickle()['start_date']
    end = m9.data.read_pickle()['end_date']
    
    df1 = D.features(ins,start,end,fields=['rank_turn_0','rank_amount_0','st_status_0'])
    df_final = pd.merge(df,df1,on=['date','instrument'])
    df_final = df_final[df_final['instrument'].str.startswith("688") == False]
    df_final = df_final[df_final['instrument'].str.startswith("3") == False]

    df_final = df_final[df_final["st_status_0"] == 0]
    df_final = df_final[df_final['rank_turn_0'] >= 0.9]
    df_final = df_final[df_final['rank_amount_0'] >= 0.85]
    print("用于回测的样本总个数为",len(df_final))
    data_1 = DataSource.write_df(df_final)
    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 m8_post_run_bigquant_run(outputs):
    return outputs

# 回测引擎：初始化函数，只执行一次
def m19_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=0.0003, sell_cost=0.0013, 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 = 2
    # 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 = [1 / stock_count 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 = 1
    context.options['hold_days'] = 1

# 回测引擎：每日数据处理函数，每天执行一次
def m19_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()}
        tmp = list(reversed(list(ranker_prediction.instrument[ranker_prediction.instrument.apply(
                lambda x: x in equities and not context.has_unfinished_sell_order(equities[x]))])))
        instruments = equities
#         # print('rank order for sell %s' % instruments)
        for instrument in instruments:
#             if instrument in tmp:
#                 print("涨停,不卖出",instrument)
#                 continue
            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 m19_prepare_bigquant_run(context):
    pass


m1 = M.instruments.v2(
    start_date='2015-01-01',
    end_date='2019-1-01',
    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, -2) / shift(open, -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))

# 将分数映射到分类，这里使用20个分类
all_wbins(label, 20)

# 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
)

m3 = M.input_features.v1(
    features='status = shift(stock_status_CN_STOCK_A__price_limit_status, -1)'
)

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=False,
    remove_extra_columns=False
)

m7 = M.join.v3(
    data1=m2.data,
    data2=m16.data,
    on='date,instrument',
    how='inner',
    sort=False
)

m6 = M.cached.v3(
    input_1=m7.data,
    run=m6_run_bigquant_run,
    post_run=m6_post_run_bigquant_run,
    input_ports='',
    params='{}',
    output_ports=''
)

m4 = M.dropnan.v2(
    input_data=m6.data_1
)

m9 = M.instruments.v2(
    start_date='2019-01-01',
    end_date='2021-12-22',
    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=30
)

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

m8 = M.cached.v3(
    input_1=m18.data,
    run=m8_run_bigquant_run,
    post_run=m8_post_run_bigquant_run,
    input_ports='',
    params='{}',
    output_ports=''
)

m5 = M.dropnan.v2(
    input_data=m8.data_1
)

m10 = M.lightgbm.v2(
    training_ds=m4.data,
    features=m3.data,
    predict_ds=m5.data,
    num_boost_round=79,
    objective='排序(ndcg)',
    num_class=1,
    num_leaves=60,
    learning_rate=0.1,
    min_data_in_leaf=900,
    max_bin=1023,
    key_cols='date,instrument',
    group_col='date',
    random_seed=101,
    other_train_parameters={'n_jobs':4,'label_gain':','.join([str(x) for x in range(20)]),"max_position":29,"eval_at":"1,3,5,10"}
)

m19 = M.trade.v4(
    instruments=m9.data,
    options_data=m10.predictions,
    start_date='',
    end_date='',
    initialize=m19_initialize_bigquant_run,
    handle_data=m19_handle_data_bigquant_run,
    prepare=m19_prepare_bigquant_run,
    volume_limit=0.25,
    order_price_field_buy='open',
    order_price_field_sell='vwap_11',
    capital_base=100000,
    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='000300.SHA'
)

m12 = M.strategy_evaluate_risk_overview.v1(
    raw_perf=m19.raw_perf
)

[2021-12-22 12:52:29.663659] INFO: moduleinvoker: instruments.v2 开始运行..

[2021-12-22 12:52:29.669673] INFO: moduleinvoker: 命中缓存

[2021-12-22 12:52:29.671039] INFO: moduleinvoker: instruments.v2 运行完成[0.007386s].

[2021-12-22 12:52:29.677855] INFO: moduleinvoker: advanced_auto_labeler.v2 开始运行..

[2021-12-22 12:52:29.683866] INFO: moduleinvoker: 命中缓存

[2021-12-22 12:52:29.685098] INFO: moduleinvoker: advanced_auto_labeler.v2 运行完成[0.007241s].

[2021-12-22 12:52:29.688599] INFO: moduleinvoker: input_features.v1 开始运行..

[2021-12-22 12:52:29.694893] INFO: moduleinvoker: 命中缓存

[2021-12-22 12:52:29.696084] INFO: moduleinvoker: input_features.v1 运行完成[0.007486s].

[2021-12-22 12:52:29.711106] INFO: moduleinvoker: general_feature_extractor.v7 开始运行..

[2021-12-22 12:52:29.716653] INFO: moduleinvoker: 命中缓存

[2021-12-22 12:52:29.717881] INFO: moduleinvoker: general_feature_extractor.v7 运行完成[0.006778s].

[2021-12-22 12:52:29.723669] INFO: moduleinvoker: derived_feature_extractor.v3 开始运行..

[2021-12-22 12:52:29.729301] INFO: moduleinvoker: 命中缓存

[2021-12-22 12:52:29.730489] INFO: moduleinvoker: derived_feature_extractor.v3 运行完成[0.00682s].

[2021-12-22 12:52:29.737485] INFO: moduleinvoker: join.v3 开始运行..

[2021-12-22 12:52:29.744553] INFO: moduleinvoker: 命中缓存

[2021-12-22 12:52:29.745802] INFO: moduleinvoker: join.v3 运行完成[0.008316s].

[2021-12-22 12:52:29.756086] INFO: moduleinvoker: cached.v3 开始运行..

[2021-12-22 12:52:29.761795] INFO: moduleinvoker: 命中缓存

[2021-12-22 12:52:29.763067] INFO: moduleinvoker: cached.v3 运行完成[0.006982s].

[2021-12-22 12:52:29.770412] INFO: moduleinvoker: dropnan.v2 开始运行..

[2021-12-22 12:52:29.776961] INFO: moduleinvoker: 命中缓存

[2021-12-22 12:52:29.778176] INFO: moduleinvoker: dropnan.v2 运行完成[0.007764s].

[2021-12-22 12:52:29.782444] INFO: moduleinvoker: instruments.v2 开始运行..

[2021-12-22 12:52:29.839976] INFO: moduleinvoker: instruments.v2 运行完成[0.057519s].

[2021-12-22 12:52:29.850582] INFO: moduleinvoker: general_feature_extractor.v7 开始运行..

[2021-12-22 12:52:30.050426] INFO: 基础特征抽取: 年份 2018, 特征行数=71261

[2021-12-22 12:52:30.787800] INFO: 基础特征抽取: 年份 2019, 特征行数=890398

[2021-12-22 12:52:31.507288] INFO: 基础特征抽取: 年份 2020, 特征行数=951963

[2021-12-22 12:52:32.440062] INFO: 基础特征抽取: 年份 2021, 特征行数=1027907

[2021-12-22 12:52:32.517696] INFO: 基础特征抽取: 总行数: 2941529

[2021-12-22 12:52:32.525231] INFO: moduleinvoker: general_feature_extractor.v7 运行完成[2.674648s].

[2021-12-22 12:52:32.532086] INFO: moduleinvoker: derived_feature_extractor.v3 开始运行..

[2021-12-22 12:52:37.157872] INFO: derived_feature_extractor: 提取完成 status = shift(stock_status_CN_STOCK_A__price_limit_status, -1), 0.405s

[2021-12-22 12:52:37.352223] INFO: derived_feature_extractor: /y_2018, 71261

[2021-12-22 12:52:38.600459] INFO: derived_feature_extractor: /y_2019, 890398

[2021-12-22 12:52:40.078180] INFO: derived_feature_extractor: /y_2020, 951963

[2021-12-22 12:52:41.571827] INFO: derived_feature_extractor: /y_2021, 1027907

[2021-12-22 12:52:41.752886] INFO: moduleinvoker: derived_feature_extractor.v3 运行完成[9.220787s].

[2021-12-22 12:52:41.763660] INFO: moduleinvoker: cached.v3 开始运行..

[2021-12-22 12:53:10.803793] INFO: moduleinvoker: cached.v3 运行完成[29.040131s].

[2021-12-22 12:53:10.841413] INFO: moduleinvoker: dropnan.v2 开始运行..

[2021-12-22 12:53:10.959067] INFO: dropnan: /data, 75018/75130

[2021-12-22 12:53:10.990950] INFO: dropnan: 行数: 75018/75130

[2021-12-22 12:53:10.996272] INFO: moduleinvoker: dropnan.v2 运行完成[0.154863s].

[2021-12-22 12:53:11.005926] INFO: moduleinvoker: lightgbm.v2 开始运行..

[2021-12-22 12:53:11.153552] ERROR: LightGBM: 部分特征没有在数据中，执行失败

[2021-12-22 12:53:11.155669] ERROR: moduleinvoker: module name: cached, module version: v2, trackeback: KeyError: "None of [Index(['status = shift(stock_status_CN_STOCK_A__price_limit_status, -1)'], dtype='object')] are in the [columns]"

[2021-12-22 12:53:11.165320] ERROR: moduleinvoker: module name: lightgbm, module version: v2, trackeback: KeyError: "None of [Index(['status = shift(stock_status_CN_STOCK_A__price_limit_status, -1)'], dtype='object')] are in the [columns]"

用于回测的样本总个数为 75130

---------------------------------------------------------------------------
KeyError                                  Traceback (most recent call last)
<ipython-input-2-d9fce71b71aa> in <module>
    233 )
    234 
--> 235 m10 = M.lightgbm.v2(
    236     training_ds=m4.data,
    237     features=m3.data,

KeyError: "None of [Index(['status = shift(stock_status_CN_STOCK_A__price_limit_status, -1)'], dtype='object')] are in the [columns]"

In [ ]:

m10.feature_gains.read().sort_values('gain',ascending=False)

In [ ]: