python - 将函数应用于 pandas 数据框列中每行的每个单词

Question

我有一个示例数据框，如下所示:

df = pd.DataFrame({
'notes': pd.Series(['speling', 'korrecter']), 
'name': pd.Series(['Walter White', 'Walter White']), 
})

  name                notes
0  Walter White     This speling is incorrect
1  Walter White     Corrector should correct korrecter

我想调整 Peter Norvig 提供的拼写检查器 here 。然后，我想通过检查行中的每个单词来将此函数应用于每一行。我想知道如何在 Python Pandas 上下文中完成此操作？

我希望输出为:

    name                notes
0  Walter White     This spelling is incorrect
1  Walter White     Corrector should correct corrector 

感谢任何意见。谢谢!

Answer 1

您可以使用 str.split 尝试此解决方案，但我认为大型 df 中的性能可能存在问题:

import pandas as pd
import numpy as np

df = pd.DataFrame({
'notes': pd.Series(['This speling is incorrect', 'Corrector should correct korrecter one']), 
'name': pd.Series(['Walter White', 'Walter White']), 
})
print df
           name                                   notes
0  Walter White               This speling is incorrect
1  Walter White  Corrector should correct korrecter one    

#simulate function correct
def correct(x):
    return x + '888'

#split column notes and apply correct
df1 = df.notes.str.split(expand=True).apply(correct)
print df1
              0           1           2             3       4
0       This888  speling888       is888  incorrect888     NaN
1  Corrector888   should888  correct888  korrecter888  one888

#remove NaN and concanecate all words together
df['notes'] = df1.fillna('').apply(lambda row: ' '.join(row), axis=1)
print df
           name                                              notes
0  Walter White             This888 speling888 is888 incorrect888 
1  Walter White  Corrector888 should888 correct888 korrecter888...