将英尺英寸格式的身高Series转换为厘米时遇AttributeError问题求助

问题场景 #

需要将Pandas Series中存储为英尺英寸格式（如5' 11"）的身高数据转换为厘米数值，执行自定义转换函数时触发AttributeError。

数据样例 #

调用fighter_details.Height.head()返回结果：

INDEX   Data
0       NaN
1       5' 11"
2       6' 3"
3       5' 11"
4       5' 6"

自定义转换函数 #

import numpy as np
import pandas as pd

def inch_to_cm(x):
    if x is np.NaN:
        return x
    else:
        # format: '7\' 11"'
        ht_ = x.split("' ")
        ft_ = float(ht_[0])
        in_ = float(ht_[1].replace('"',""))
        return ((12*ft_) + in_) * 2.54

执行代码与报错 #

执行代码：

fighter_details['Height'] = fighter_details['Height'].apply(inch_to_cm)

报错信息：

---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
Input In [240], in <cell line: 1>()
----> 1 fighter_details['Height'] = fighter_details['Height'].apply(inch_to_cm)

File ~/opt/anaconda3/envs/book_env/lib/python3.8/site-packages/pandas/core/series.py:4108, in Series.apply(self, func, convert_dtype, args, **kwds)
   4106     else:
   4107         values = self.astype(object)._values
-> 4108         mapped = lib.map_infer(values, f, convert=convert_dtype)
   4110 if len(mapped) and isinstance(mapped[0], Series):
   4111     # GH 25959 use pd.array instead of tolist
   4112     # so extension arrays can be used
   4113     return self._constructor_expanddim(pd_array(mapped), index=self.index)

File pandas/_libs/lib.pyx:2467, in pandas._libs.lib.map_infer()

Input In [239], in inch_to_cm(x)
      3     return x
      4 else:
      5     # format: '7\' 11"'
----> 6     ht_ = x.split("' ")
      7     ft_ = float(ht_[0])
      8     in_ = float(ht_[1].replace('"',""))

AttributeError: 'float' object has no attribute 'split'

错误原因 #

x is np.NaN无法正确识别Pandas中的缺失值：

• Pandas里的NaN是float类型的np.nan，但用is判断时，由于内存地址等问题，无法匹配所有NaN实例
• 导致NaN值进入else分支，而NaN是float类型，没有split方法，触发AttributeError

解决方法 #

方法1：修正NaN判断逻辑 #

使用pd.isna()或np.isnan()替代is判断，确保正确识别缺失值：

def inch_to_cm(x):
    if pd.isna(x):
        return x
    else:
        ht_ = x.split("' ")
        ft_ = float(ht_[0])
        in_ = float(ht_[1].strip('"'))
        return ((12 * ft_) + in_) * 2.54

重新执行apply即可：

fighter_details['Height'] = fighter_details['Height'].apply(inch_to_cm)

方法2：用Pandas字符串方法高效处理（推荐） #

避免apply循环，用str.extract直接提取英尺和英寸数值，计算更高效：

# 提取英尺和英寸，生成两列数值
height_extract = fighter_details['Height'].str.extract(r'(\d+)\' (\d+)"').astype(float)
# 计算厘米：(英尺*12 + 英寸)*2.54
fighter_details['Height'] = (height_extract[0]*12 + height_extract[1])*2.54

这种方法自动忽略NaN值，无需额外判断，性能更优。

内容的提问来源于stack exchange，提问作者epg32