WorldQuant 101 Alpha因子测试出错？？？

189 · 2019-11-21 03:27:33 UTC

照着例子只修改了特征列表，跑一遍，就出错

克隆策略

In [1]:

# 本代码由可视化策略环境自动生成 2019年11月21日 11:26
# 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。


# 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 = 5
    # 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 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()}
        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 m19_prepare_bigquant_run(context):
    pass


m1 = M.instruments.v2(
    start_date='2016-01-01',
    end_date='2018-01-01',
    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, -5) / 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="""# #号开始的表示注释
# 多个特征，每行一个，可以包含基础特征和衍生特征

factor=rank(close_0/max(mean(close_0, 10), mean(close_0, 20), mean(close_0, 30), mean(close_0, 60),mean(close_0, 120),mean(close_0, 250))-1)*100"""
)

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=True,
    remove_extra_columns=False
)

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

m6 = M.features_short.v1(
    input_1=m3.data
)

m9 = M.instruments.v2(
    start_date=T.live_run_param('trading_date', '2018-01-01'),
    end_date=T.live_run_param('trading_date', '2019-11-21'),
    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=0
)

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

m4 = M.stock_ranker.v2(
    training_ds=m7.data,
    features=m6.data_1,
    predict_ds=m18.data,
    learning_algorithm='排序',
    number_of_leaves=30,
    minimum_docs_per_leaf=1000,
    number_of_trees=20,
    learning_rate=0.1,
    max_bins=1023,
    feature_fraction=1,
    slim_data=True
)

m19 = M.trade.v4(
    instruments=m9.data,
    options_data=m4.predictions,
    start_date='',
    end_date='',
    initialize=m19_initialize_bigquant_run,
    handle_data=m19_handle_data_bigquant_run,
    prepare=m19_prepare_bigquant_run,
    volume_limit=0.025,
    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'
)

[2019-11-21 11:05:02.714667] INFO: bigquant: instruments.v2 开始运行..

[2019-11-21 11:05:03.885044] INFO: bigquant: 命中缓存

[2019-11-21 11:05:03.887591] INFO: bigquant: instruments.v2 运行完成[1.172932s].

[2019-11-21 11:05:03.970492] INFO: bigquant: advanced_auto_labeler.v2 开始运行..

[2019-11-21 11:05:04.885032] INFO: bigquant: 命中缓存

[2019-11-21 11:05:04.938268] INFO: bigquant: advanced_auto_labeler.v2 运行完成[0.967769s].

[2019-11-21 11:05:04.978188] INFO: bigquant: input_features.v1 开始运行..

[2019-11-21 11:05:05.810738] INFO: bigquant: input_features.v1 运行完成[0.832543s].

[2019-11-21 11:05:07.185032] INFO: bigquant: general_feature_extractor.v7 开始运行..

[2019-11-21 11:05:11.666996] INFO: 基础特征抽取: 年份 2016, 特征行数=641546

[2019-11-21 11:05:16.712803] INFO: 基础特征抽取: 年份 2017, 特征行数=743233

[2019-11-21 11:05:18.867235] INFO: 基础特征抽取: 年份 2018, 特征行数=0

[2019-11-21 11:05:23.740737] INFO: 基础特征抽取: 总行数: 1384779

[2019-11-21 11:05:23.834418] INFO: bigquant: general_feature_extractor.v7 运行完成[16.649377s].

[2019-11-21 11:05:23.862756] INFO: bigquant: derived_feature_extractor.v3 开始运行..

[2019-11-21 11:05:49.864371] INFO: derived_feature_extractor: 提取完成 factor=rank(close_0/max(mean(close_0, 10), mean(close_0, 20), mean(close_0, 30), mean(close_0, 60),mean(close_0, 120),mean(close_0, 250))-1)*100, 23.216s

[2019-11-21 11:05:51.013384] INFO: derived_feature_extractor: /y_2016, 641546

[2019-11-21 11:05:51.484350] INFO: derived_feature_extractor: /y_2017, 743233

[2019-11-21 11:05:54.357678] INFO: bigquant: derived_feature_extractor.v3 运行完成[30.494911s].

[2019-11-21 11:05:54.385794] INFO: bigquant: join.v3 开始运行..

[2019-11-21 11:05:57.936201] INFO: join: /y_2016, 行数=0/0, 耗时=0.747948s

[2019-11-21 11:06:00.313135] INFO: join: /y_2017, 行数=559525/574928, 耗时=2.374498s

[2019-11-21 11:06:04.263222] INFO: join: 最终行数: 559525

[2019-11-21 11:06:04.272224] INFO: bigquant: join.v3 运行完成[9.886433s].

[2019-11-21 11:06:04.338453] INFO: bigquant: features_short.v1 开始运行..

[2019-11-21 11:06:05.664581] INFO: bigquant: features_short.v1 运行完成[1.326119s].

[2019-11-21 11:06:05.669990] INFO: bigquant: instruments.v2 开始运行..

[2019-11-21 11:06:06.655175] INFO: bigquant: instruments.v2 运行完成[0.985143s].

[2019-11-21 11:06:07.953190] INFO: bigquant: general_feature_extractor.v7 开始运行..

[2019-11-21 11:06:23.167041] INFO: 基础特征抽取: 年份 2018, 特征行数=816987

[2019-11-21 11:06:26.276528] INFO: 基础特征抽取: 年份 2019, 特征行数=776864

[2019-11-21 11:06:35.363729] INFO: 基础特征抽取: 总行数: 1593851

[2019-11-21 11:06:35.379784] INFO: bigquant: general_feature_extractor.v7 运行完成[27.426579s].

[2019-11-21 11:06:35.383175] INFO: bigquant: derived_feature_extractor.v3 开始运行..

[2019-11-21 11:07:04.176374] INFO: derived_feature_extractor: 提取完成 factor=rank(close_0/max(mean(close_0, 10), mean(close_0, 20), mean(close_0, 30), mean(close_0, 60),mean(close_0, 120),mean(close_0, 250))-1)*100, 26.953s

[2019-11-21 11:07:04.775216] INFO: derived_feature_extractor: /y_2018, 816987

[2019-11-21 11:07:05.257006] INFO: derived_feature_extractor: /y_2019, 776864

[2019-11-21 11:07:07.098595] INFO: bigquant: derived_feature_extractor.v3 运行完成[31.715391s].

[2019-11-21 11:07:07.123358] INFO: bigquant: stock_ranker.v2 开始运行..

[2019-11-21 11:07:07.476388] INFO: bigquant: stock_ranker_train.v5 开始运行..

[2019-11-21 11:07:08.520137] INFO: StockRanker: 特征预处理 ..

[2019-11-21 11:07:08.937044] INFO: StockRanker: prepare data: training ..

[2019-11-21 11:07:18.175225] INFO: StockRanker训练: 00785dbc 准备训练: 559525 行数

[2019-11-21 11:07:18.993100] INFO: StockRanker训练: 正在训练 ..

[2019-11-21 11:09:06.426953] INFO: bigquant: stock_ranker_train.v5 运行完成[118.950596s].

[2019-11-21 11:09:06.462312] INFO: bigquant: stock_ranker_predict.v5 开始运行..

[2019-11-21 11:09:10.731589] INFO: StockRanker预测: /y_2018 ..

[2019-11-21 11:09:10.747457] ERROR: bigquant: module name: stock_ranker_predict, module version: v5, trackeback: Traceback (most recent call last):
ValueError: Input data must be 2 dimensional and non empty.

[2019-11-21 11:09:11.106816] ERROR: bigquant: module name: stock_ranker, module version: v2, trackeback: Traceback (most recent call last):
ValueError: Input data must be 2 dimensional and non empty.

StockRanker
特征重要性
模型
NDCG

设置测试数据集，查看训练迭代过程的NDCG

bigcharts-data-start/{"__type":"tabs","__id":"bigchart-de8c32b0b98b48038cff21db0c368cc4"}/bigcharts-data-end

StockRanker预测(stock_ranker_predict)使用错误，你可以：

1.一键查看文档

2.一键搜索答案

StockRanker(stock_ranker)使用错误，你可以：

1.一键查看文档

2.一键搜索答案

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
<ipython-input-1-e7bccf635856> in <module>()
    171     max_bins=1023,
    172     feature_fraction=1,
--> 173     slim_data=True
    174 )
    175 

ValueError: Input data must be 2 dimensional and non empty.

iQuant · 2019-11-21 03:42:42 UTC

应该是因子名称太长，可以重命名后再尝试一下。

189 · 2019-11-21 06:40:08 UTC

我之前是分拆了，最后才再合并成一个衍生因子，是出错了，才改成全部直接合并成为一个因子,出错信息不太明白：Input data must be 2 dimensional and non empty.