根据行合并两个数据框

Question

我知道有很多关于合并两个 Pandas df 的资源，但我正在尝试根据第二个 df 的 ID 合并一个 df 但我需要从行中创建新列第二个df。这有点令人困惑，但我在这里有一个例子可以阐明我正在尝试做的事情。

我有：

dfa = pd.DataFrame({"ID": ["1", "2", "3"],"Color":["Red", "White", "Blue"],"Length":["16", "14.97", "22.75"]})

dfb = pd.DataFrame({"ID": ["1", "1", "2","3"],"Col1":["Color", "Width", "Length","Color"],"Value":["Blue", "14.97", "22.75","Green"]})

我想要的：

dfc = pd.DataFrame({"ID": ["1", "2", "3"],"Color":["Blue", "White", "Green"],"Length":["16", "14.97", "22.75"],"c:Color":["Blue","NaN","Green"],"c:Width":["14.97","NaN","NaN"],"c:Length":["NaN","22.75","NaN"]})

如有任何帮助，我们将不胜感激！

Answer 1

在merge之前使用pivot:

>>> dfa.merge(dfb.pivot('ID', 'Col1', 'Value').add_prefix('c:'), on='ID')

  ID  Color Length c:Color c:Length c:Width
0  1    Red     16    Blue      NaN   14.97
1  2  White  14.97     NaN    22.75     NaN
2  3   Blue  22.75   Green      NaN     NaN

要获得 'exactly' 你的输出：

>>> dfa.merge(dfb.pivot('ID', 'Col1', 'Value')[dfb['Col1'].unique()].add_prefix('c:'), on='ID')
  ID  Color Length c:Color c:Width c:Length
0  1    Red     16    Blue   14.97      NaN
1  2  White  14.97     NaN     NaN    22.75
2  3   Blue  22.75   Green     NaN      NaN

Answer 2

加入前需要转换为宽屏：

dfa.merge(
    dfb.pivot(
        index='ID', 
        columns='Col1', 
        values='Value'
        ).add_prefix('c:'),
    on = 'ID'
    )

根据行合并两个数据框

Merge two dataframes based on rows

python

merge

dataframe

pandas