假設有一群人可以選擇英語和/或西班牙語單詞。假設他們這樣選擇：
>>> pandas.DataFrame(dict(person=['mary','james','patricia','robert','jennifer','michael'],english=['water',None,'hello','thanks',None,'green'],spanish=[None,'agua',None,None,'bienvenido','verde']))
person english spanish
0 mary water None
1 james None agua
2 patricia hello None
3 robert thanks None
4 jennifer None bienvenido
5 michael green verde
假設我也有一本英西詞典（假設沒有重復，即一對一的關系）：
>>> pandas.DataFrame(dict(english=['hello','bad','green','thanks','welcome','water'],spanish=['hola','malo','verde','gracias','bienvenido','agua']))
english spanish
0 hello hola
1 bad malo
2 green verde
3 thanks gracias
4 welcome bienvenido
5 water agua
如何填寫任何缺失的單詞，即DataFrame使用第二個DataFramewhereenglish或spanishis更新第一個單詞，None以達到此目的：
>>> pandas.DataFrame(dict(person=['mary','james','patricia','robert','jennifer','michael'],english=['water','water','hello','thanks','welcome','green'],spanish=['agua','agua','hola','gracias','bienvenido','verde']))
person english spanish
0 mary water agua
1 james water agua
2 patricia hello hola
3 robert thanks gracias
4 jennifer welcome bienvenido
5 michael green verde
uj5u.com熱心網友回復：
你可以檢查map一下fillna
df['english'] = df['english'].fillna(df['spanish'].map(df2.set_index('spanish')['english']))
df['spanish'] = df['spanish'].fillna(df['english'].map(df2.set_index('english')['spanish']))
df
Out[200]:
person english spanish
0 mary water agua
1 james water agua
2 patricia hello hola
3 robert thanks gracias
4 jennifer welcome bienvenido
5 michael green verde
轉載請註明出處，本文鏈接：https://www.uj5u.com/qiye/461332.html