np.select -超过最大递归深度

Question

当值介于两个数字之间时，我想为dataframe行分配一个特定值，如下所示。

我得到一个错误“最大递归深度超过”。

你能帮我修一下吗？

dfmerged['OptionType'] = np.select(
[
    np.logical_and(dfmerged['right'] == 'C', dfmerged['delta'] >= 0.75),
    np.logical_and(dfmerged['right'] == 'C', dfmerged['delta'] < 0.75 , dfmerged['delta'] > 0.55 ),
    np.logical_and(dfmerged['right'] == 'C', dfmerged['delta'] <= 0.55, dfmerged['delta'] > 0.45),
    np.logical_and(dfmerged['right'] == 'C', dfmerged['delta'] <= 0.45, dfmerged['delta'] >= 0.25),
    np.logical_and(dfmerged['right'] == 'C', dfmerged['delta'] < 0.25),
    np.logical_and(dfmerged['right'] == 'P', dfmerged['delta'] <= -0.75),
    np.logical_and(dfmerged['right'] == 'P', dfmerged['delta'] > -0.75, dfmerged['delta'] < -0.55),
    np.logical_and(dfmerged['right'] == 'P', dfmerged['delta'] >= -0.55, dfmerged['delta'] < -0.45),
    np.logical_and(dfmerged['right'] == 'P', dfmerged['delta'] >= -0.45, dfmerged['delta'] <= -0.25),
    np.logical_and(dfmerged['right'] == 'P', dfmerged['delta'] < -0.25),
],
[
    'Deep ITM',
    'ITM',
    'ATM',
    'OTM',
    'Deep OTM',
    'Deep ITM',
    'ITM',
    'ATM',
    'OTM',
    'Deep OTM'

编辑#1

我尝试了下面的错误:值的长度(1)不匹配索引的长度(8344).你有什么办法解决这个问题吗？

right_c = dfmerged['right'] == 'C'
right_p = dfmerged['right'] == 'P'
above_075 = dfmerged['delta'] >= 0.75
tween_055_075 = np.logical_and( dfmerged['delta'] < 0.75, dfmerged['delta'] > 0.55 )
tween_045_055 = np.logical_and( dfmerged['delta'] <= 0.55, dfmerged['delta'] > 0.45 )
tween_025_045 = np.logical_and( dfmerged['delta'] <= 0.45, dfmerged['delta'] >= 0.25 )
below_025 = dfmerged['delta'] < 0.25
p_below_075 = dfmerged['delta'] <= -0.75,
p_tween_075_055 = np.logical_and(dfmerged['delta'] > -0.75, dfmerged['delta'] < -0.55),
p_tween_055_045 = np.logical_and(dfmerged['delta'] >= -0.55, dfmerged['delta'] < -0.45),
p_tween_045_025 = np.logical_and(dfmerged['delta'] >= -0.45, dfmerged['delta'] <= -0.25),
p_below_075_025 = np.logical_and(dfmerged['right'] == 'P', dfmerged['delta'] < -0.25)

dfmerged['OptionType'] = np.select(
     [
        np.logical_and(right_c, above_075),
        np.logical_and(right_c, tween_055_075),
        np.logical_and(right_c, tween_045_055),
        np.logical_and(right_c, tween_025_045),
        np.logical_and(right_c, below_025),
        np.logical_and(right_p, p_below_075),
        np.logical_and(right_p, p_tween_075_055),
        np.logical_and(right_p,p_tween_055_045),
        np.logical_and(right_p,p_tween_045_025),
        np.logical_and(right_p,p_below_075_025)
    ],
      [
          'Deep ITM',
          'ITM',
          'ATM',
          'OTM',
          'Deep OTM',
          'Deep ITM',
          'ITM',
          'ATM',
          'OTM',
          'Deep OTM'
      ],
      default='Unknown')

Answer 1

我只是在这里集思广益，但是Python没有全局优化器，所以它将解析dfmerged‘’right‘10次和dfmerged’‘delta’16次。我建议你做你自己的全局优化者：

right_c = dfmerged['right'] == 'C'
right_p = dfmerged['right'] == 'P'
above_075 = dfmerged['delta'] >= 0.75
tween_055_075 = np.and( dfmerged['delta'] < 0.75, dfmerged['delta'] > 0.55 )
tween_045_055 = np.and( dfmerged['delta'] <= 0.55, dfmerged['delta'] > 0.45 )
tween_025_045 = np.and( dfmerged['delta'] <= 0.45, dfmerged['delta'] > 0.25 )
below_025 = dfmerged['delta'] < 0.25
dfmerged['OptionType'] = np.select(
[
    np.logical_and[ right_c, above_075 ),
    np.logibal_and[ right_c, tween_055_075 ),
... etc ...
]

虽然我现在看到我对'P‘实体有错误的迹象，但我认为你明白了。