transform模块出错

用户成长系列
新手专区
标签: #<Tag:0x00007f9d712f2978> #<Tag:0x00007f9d712f2838>

(cph315) #1

我在给特征转换时,出现如下错误:
Exception Traceback (most recent call last)
in ()
228 # clip最后的数据,保证输入落到如下区间
229 clip_lower=0,
–> 230 clip_upper=200000000
231 )
232

Exception: 没有为特征 (fs_current_liabilities_0+fs_non_current_liabilities_0)/(fs_current_assets_0+fs_non_current_assets_0) 找到匹配的 transform。请添加相应的transform,或者在最后使用 (’.*’, None) 来跳过所有不用transform的特征

我的代码如下:
m5 = M.derived_feature_extractor.v2(
input_data=m4.data,
features=m3.data,
date_col=‘date’,
instrument_col=‘instrument’
)

m15 = M.transform.v2(
data=m5.data,
# stockranker 默认的转换函数,主要是将特征映射到非负整数区间,因为stockranker要求输入特征数据为非负整数
transforms=T.get_stock_ranker_default_transforms(),
drop_null=True, # 缺失数据处理,如果某一行有空列,则删除
astype=‘int32’, # 数据类型转换
except_columns=[‘date’, ‘instrument’], # 跳过的列,不需要处理
# clip最后的数据,保证输入落到如下区间
clip_lower=0,
clip_upper=200000000
)


(upndown) #2

其实,你可以直接在学院和社区克隆策略,或者通过平台新建策略时默认的模板,这样代码就不会出错了。
你要是检验某个因子的有效性的话,直接在可视化模板里修改因子即可。


(cph315) #3

我这个因子就是通过可视化策略生成器选择的,它出来就是这样。


(cph315) #4

这个是我的整个代码:

克隆策略
In [10]:
class conf:
    # 定义一个conf类,存储需要使用的变量
    start_date='2014-01-01'  # 日期,作为训练集的起始日期
    end_date='2017-02-17'  # 日期,作为测试集的结束日期
    df = D.history_data(D.instruments(),end_date,end_date,['in_csi800'])
    instruments = list(set(df[df['in_csi800']==1]['instrument']))
    print('instruments len: ', len(instruments))
    # 股票池
   
    # 10日收益率
    hold_days = 5
    # 以沪深300为基准的相对收益
    benchmark = '000300.SHA'
    
# 本代码由可视化策略环境自动生成 2018年1月20日 12:39
# 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。


m1 = M.instruments.v2(
    start_date='2012-01-01',
    end_date='2016-01-01',
    market='CN_STOCK_A',
    instrument_list = conf.instruments,
    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, 10)

# 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
return_5
return_10
return_20
return_40
rank_return_5
rank_return_10
rank_return_20
rank_return_40
turn_0
rank_turn_0
avg_turn_5
avg_turn_10
avg_turn_20
avg_turn_40
rank_avg_turn_5
avg_amount_5
avg_amount_10
avg_amount_20
avg_amount_40
#(volume_0+volume_1+volume_2+volume_3+volume_4)/5
#(volume_0+volume_1+volume_2+volume_3+volume_4+volume_5+volume_6+volume_7+volume_8+volume_9)/10
#(volume_0+volume_1+volume_2+volume_3+volume_4+volume_5+volume_6+volume_7+volume_8+volume_9+volume_10+volume_11+volume_12+volume_13+volume_14+volume_15+volume_16+volume_17+volume_18+volume_19)/20
#(high_0-low_0+high_1-low_1+high_2-low_2+high_3-low_3+high_4-low_4)/5
fs_common_equity_0
market_cap_0
rank_market_cap_0
market_cap_float_0
rank_market_cap_float_0
pe_ttm_0
rank_pe_ttm_0
pe_lyr_0
rank_pe_lyr_0
pb_lf_0
rank_pb_lf_0
ps_ttm_0
rank_ps_ttm_0
mf_net_amount_5
mf_net_amount_10
mf_net_amount_20
avg_mf_net_amount_5
avg_mf_net_amount_10
avg_mf_net_amount_20
rank_avg_mf_net_amount_5
rank_avg_mf_net_amount_10
rank_avg_mf_net_amount_20
fs_net_profit_0
fs_net_profit_ttm_0
fs_deducted_profit_0
fs_deducted_profit_ttm_0
fs_gross_profit_margin_0
fs_gross_profit_margin_ttm_0
fs_net_profit_margin_0
fs_net_profit_margin_ttm_0
fs_operating_revenue_0
fs_operating_revenue_ttm_0
fs_net_profit_yoy_0
rank_fs_net_profit_yoy_0
fs_net_profit_qoq_0
rank_fs_net_profit_qoq_0
fs_operating_revenue_yoy_0
rank_fs_operating_revenue_yoy_0
fs_operating_revenue_qoq_0
rank_fs_operating_revenue_qoq_0
fs_eps_yoy_0
fs_roe_0
fs_roe_ttm_0
fs_roa_0
fs_roa_ttm_0
fs_eps_0
fs_bps_0
fs_cash_ratio_0
#fs_operating_revenue_ttm_0/(fs_current_assets_0+fs_non_current_assets_0)
#fs_current_assets_0/fs_current_liabilities_0
#(fs_current_liabilities_0+fs_non_current_liabilities_0)/ fs_common_equity_0
#(fs_current_liabilities_0+fs_non_current_liabilities_0)/(fs_current_assets_0+fs_non_current_assets_0)
fs_free_cash_flow_0
fs_net_cash_flow_0
fs_net_cash_flow_ttm_0
fs_current_assets_0
fs_non_current_assets_0
fs_current_liabilities_0
fs_non_current_liabilities_0
sh_holder_avg_pct_0
rank_sh_holder_avg_pct_0
sh_holder_avg_pct_3m_chng_0
rank_sh_holder_avg_pct_3m_chng_0
sh_holder_avg_pct_6m_chng_0
rank_sh_holder_avg_pct_6m_chng_0
ta_sma_5_0
ta_sma_10_0
ta_sma_20_0
ta_atr_14_0
ta_atr_28_0
ta_mfi_14_0
ta_mfi_28_0
ta_rsi_14_0
ta_rsi_28_0
ta_trix_14_0
ta_trix_28_0
ta_sar_0
ta_mom_10_0
ta_mom_20_0
ta_mom_30_0
ta_mom_60_0
ta_willr_14_0
ta_willr_28_0
ta_ad_0
ta_aroon_down_14_0
ta_aroon_down_28_0
ta_aroon_up_14_0
ta_aroon_up_28_0
ta_aroonosc_14_0
ta_aroonosc_28_0
ta_bbands_upperband_14_0
ta_bbands_upperband_28_0
ta_bbands_middleband_14_0
ta_bbands_middleband_28_0
ta_bbands_lowerband_14_0
ta_bbands_lowerband_28_0
ta_adx_14_0
ta_adx_28_0
ta_cci_14_0
ta_cci_28_0
ta_macd_macd_12_26_9_0
ta_macd_macdsignal_12_26_9_0
ta_macd_macdhist_12_26_9_0
ta_obv_0
ta_stoch_slowk_5_3_0_3_0_0
ta_stoch_slowd_5_3_0_3_0_0
swing_volatility_5_0
swing_volatility_10_0
swing_volatility_30_0
rank_swing_volatility_5_0
rank_swing_volatility_10_0
rank_swing_volatility_30_0
volatility_5_0
volatility_10_0
volatility_30_0
rank_volatility_5_0
rank_volatility_10_0
rank_volatility_30_0
beta_csi800_5_0
beta_csi800_10_0
beta_csi800_30_0
rank_beta_csi800_5_0
rank_beta_csi800_10_0
rank_beta_csi800_30_0
"""
)

m4 = M.general_feature_extractor.v6(
    instruments=m1.data,
    features=m3.data,
    start_date='',
    end_date='',
    before_start_days=0
)

m5 = M.derived_feature_extractor.v2(
    input_data=m4.data,
    features=m3.data,
    date_col='date',
    instrument_col='instrument'
)

m15 = M.transform.v2(
        data=m5.data,
        # stockranker 默认的转换函数,主要是将特征映射到非负整数区间,因为stockranker要求输入特征数据为非负整数
        transforms=T.get_stock_ranker_default_transforms(),
        drop_null=True, # 缺失数据处理,如果某一行有空列,则删除
        astype='int32', # 数据类型转换
        except_columns=['date', 'instrument'], # 跳过的列,不需要处理
        # clip最后的数据,保证输入落到如下区间
        clip_lower=0, 
        clip_upper=200000000
)

m7 = M.join.v3(
    data1=m2.data,
    data2=m15.data,
    on='date,instrument',
    how='inner',
    sort=False
)

m13 = M.dropnan.v1(
    input_data=m7.data
)

m9 = M.instruments.v2(
    start_date=T.live_run_param('trading_date', '2016-01-01'),
    end_date=T.live_run_param('trading_date', '2017-01-01'),
    market='CN_STOCK_A',
    instrument_list = conf.instruments,
    max_count=0
)

m10 = M.general_feature_extractor.v6(
    instruments=m9.data,
    features=m3.data,
    start_date='',
    end_date='',
    before_start_days=0
)

m11 = M.derived_feature_extractor.v2(
    input_data=m10.data,
    features=m3.data,
    date_col='date',
    instrument_col='instrument'
)

m14 = M.transform.v2(data=m11.data, 
        # stockranker 默认的转换函数,主要是将特征映射到非负整数区间,因为stockranker要求输入特征数据为非负整数
        transforms=T.get_stock_ranker_default_transforms(),
        drop_null=True, # 缺失数据处理,如果某一行有空列,则删除
        astype='int32', # 数据类型转换
        except_columns=['date', 'instrument'], # 跳过的列,不需要处理
        # clip最后的数据,保证输入落到如下区间
        clip_lower=0, clip_upper=200000000
)

m6 = M.stock_ranker_train.v5(
    training_ds=m15.data,
    features=m3.data,
    learning_algorithm='排序',
    number_of_leaves=30,
    minimum_docs_per_leaf=1000,
    number_of_trees=1000,
    learning_rate=0.1,
    max_bins=1023,
    feature_fraction=1,
    #m_lazy_run=False
)

print(m6.feature_gains.read_hdf())

m8 = M.stock_ranker_predict.v5(
    model=m6.model,
    data=m14.data,
    #m_lazy_run=False
)

# 回测引擎:每日数据处理函数,每天执行一次
def m12_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 m12_prepare_bigquant_run(context):
    pass

# 回测引擎:初始化函数,只执行一次
def m12_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

m12 = M.trade.v3(
    instruments=m9.data,
    options_data=m8.predictions,
    start_date='',
    end_date='',
    handle_data=m12_handle_data_bigquant_run,
    prepare=m12_prepare_bigquant_run,
    initialize=m12_initialize_bigquant_run,
    volume_limit=0.025,
    order_price_field_buy='open',
    order_price_field_sell='close',
    capital_base=1000000,
    benchmark='000300.SHA',
    auto_cancel_non_tradable_orders=True,
    data_frequency='daily',
    price_type='后复权',
    plot_charts=True,
    backtest_only=False,
    amount_integer=False
)
instruments len:  800
[2018-01-20 15:32:17.553899] INFO: bigquant: instruments.v2 开始运行..
[2018-01-20 15:32:17.557528] INFO: bigquant: 命中缓存
[2018-01-20 15:32:17.559172] INFO: bigquant: instruments.v2 运行完成[0.005322s].
[2018-01-20 15:32:17.570568] INFO: bigquant: advanced_auto_labeler.v2 开始运行..
[2018-01-20 15:32:17.574276] INFO: bigquant: 命中缓存
[2018-01-20 15:32:17.575250] INFO: bigquant: advanced_auto_labeler.v2 运行完成[0.004707s].
[2018-01-20 15:32:17.581191] INFO: bigquant: input_features.v1 开始运行..
[2018-01-20 15:32:17.584257] INFO: bigquant: 命中缓存
[2018-01-20 15:32:17.585979] INFO: bigquant: input_features.v1 运行完成[0.004763s].
[2018-01-20 15:32:17.604852] INFO: bigquant: general_feature_extractor.v6 开始运行..
[2018-01-20 15:32:17.608258] INFO: bigquant: 命中缓存
[2018-01-20 15:32:17.609643] INFO: bigquant: general_feature_extractor.v6 运行完成[0.004475s].
[2018-01-20 15:32:17.617980] INFO: bigquant: derived_feature_extractor.v2 开始运行..
[2018-01-20 15:32:17.620837] INFO: bigquant: 命中缓存
[2018-01-20 15:32:17.622241] INFO: bigquant: derived_feature_extractor.v2 运行完成[0.004277s].
[2018-01-20 15:32:17.635521] INFO: bigquant: transform.v2 开始运行..
[2018-01-20 15:32:17.639179] INFO: bigquant: 命中缓存
[2018-01-20 15:32:17.640535] INFO: bigquant: transform.v2 运行完成[0.005051s].
[2018-01-20 15:32:17.651668] INFO: bigquant: join.v3 开始运行..
[2018-01-20 15:32:17.655906] INFO: bigquant: 命中缓存
[2018-01-20 15:32:17.657262] INFO: bigquant: join.v3 运行完成[0.005606s].
[2018-01-20 15:32:17.668961] INFO: bigquant: dropnan.v1 开始运行..
[2018-01-20 15:32:17.672360] INFO: bigquant: 命中缓存
[2018-01-20 15:32:17.673554] INFO: bigquant: dropnan.v1 运行完成[0.004618s].
[2018-01-20 15:32:17.681585] INFO: bigquant: instruments.v2 开始运行..
[2018-01-20 15:32:17.684830] INFO: bigquant: 命中缓存
[2018-01-20 15:32:17.686214] INFO: bigquant: instruments.v2 运行完成[0.004908s].
[2018-01-20 15:32:17.768635] INFO: bigquant: general_feature_extractor.v6 开始运行..
[2018-01-20 15:32:17.771659] INFO: bigquant: 命中缓存
[2018-01-20 15:32:17.772753] INFO: bigquant: general_feature_extractor.v6 运行完成[0.004099s].
[2018-01-20 15:32:17.782988] INFO: bigquant: derived_feature_extractor.v2 开始运行..
[2018-01-20 15:32:17.786585] INFO: bigquant: 命中缓存
[2018-01-20 15:32:17.788132] INFO: bigquant: derived_feature_extractor.v2 运行完成[0.005208s].
[2018-01-20 15:32:17.803099] INFO: bigquant: transform.v2 开始运行..
[2018-01-20 15:32:17.806672] INFO: bigquant: 命中缓存
[2018-01-20 15:32:17.807978] INFO: bigquant: transform.v2 运行完成[0.004907s].
[2018-01-20 15:32:17.819406] INFO: bigquant: stock_ranker_train.v5 开始运行..
---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
<ipython-input-10-1826398010f1> in <module>()
    286     learning_rate=0.1,
    287     max_bins=1023,
--> 288     feature_fraction=1,
    289     #m_lazy_run=False
    290 )

TypeError: int() argument must be a string, a bytes-like object or a number, not 'NoneType'


(cph315) #5

哦,我忘了打开注释了,只要把注释掉的特征打开就会出现我提的错误。