Pandas 根据列名不一致的列合并dataframe

如果两个DataFrame的列名不一致，但你希望根据列名不一致的列进行合并，可以通过明确指定left_on和right_on参数来实现这一点。left_on参数接受一个列表，表示左侧DataFrame中用于合并的列名，而right_on参数同样接受一个列表，表示右侧DataFrame中用于合并的列名。

import pandas as pd  
  
df1 = pd.DataFrame({  
    'name': ['Alice', 'Bob', 'Charlie'],  
    'age': [20, 22, 21], 
    'hobby': ['swim', 'shopping', 'sleep']  
})  
   
df2 = pd.DataFrame({  
    'last_name': ['Alice', 'Bob', 'Charlie'],  # 假设X列的值与A列相匹配  
    'how_old': [20, 34, 21],  
    'food': ['beef', 'cake', 'milk']  
})  
  
# 根据列名不一致的列进行合并  
merged_df = pd.merge(df1, df2, left_on=['name', 'age'], right_on=['last_name', 'how_old'])
  
print(merged_df)

      name  age  hobby last_name  how_old  food
0    Alice   20   swim     Alice       20  beef
1  Charlie   21  sleep   Charlie       21  milk

ps：如果使用外连接（outer）merge的话

merged_df = pd.merge(df1, df2, how='outer',left_on=['name', 'age'], right_on=['last_name', 'how_old'])

输出：

      name   age     hobby last_name  how_old  food
0    Alice  20.0      swim     Alice     20.0  beef
1      Bob  22.0  shopping       NaN      NaN   NaN
2      NaN   NaN       NaN       Bob     34.0  cake
3  Charlie  21.0     sleep   Charlie     21.0  milk

如果用--而不是NaN来填充缺失的列：

merged_df.fillna('--', inplace=True)
# 这里会有 FutureWarning: Setting an item of incompatible dtype is deprecated and will raise an error in a future version of pandas. Value '--' has dtype incompatible with float64, please explicitly cast to a compatible dtype first.

输出：

      name   age     hobby last_name how_old  food
0    Alice  20.0      swim     Alice    20.0  beef
1      Bob  22.0  shopping        --      --    --
2       --    --        --       Bob    34.0  cake
3  Charlie  21.0     sleep   Charlie    21.0  milk