如何将数据帧堆叠在一起(Pandas,Python3)

时间:2014-10-01 03:02:58

标签: python python-3.x pandas

假设我有3只熊猫DF

DF1

 Words      Score
 The Man     2
 The Girl    4

Df2

 Words2      Score2
The Boy       6
The Mother    7

Df3

Words3       Score3
The Son        3
The Daughter   4

现在,我将它们连接在一起,使其在一个DF中变为6列。这一切都很好,但我想知道,是否有一个pandas函数将它们垂直堆叠到两列中并更改标题?

那要做这样的事情吗?

Family Members     Score
The Man             2
The Girl            4
The Boy             6
The Mother          7
The Son             3
The Daughter        4

我在这里阅读的所有内容http://pandas.pydata.org/pandas-docs/stable/merging.html似乎只有"水平"加入DF的方法!

1 个答案:

答案 0 :(得分:6)

只要您重命名列以使它们在每个数据框中都相同,pd.concat()就可以正常工作:

# I read in your data as df1, df2 and df3 using:
# df1 = pd.read_clipboard(sep='\s\s+')
# Example dataframe:

Out[8]: 
      Words  Score
0   The Man      2
1  The Girl      4


all_dfs = [df1, df2, df3]

# Give all df's common column names
for df in all_dfs:
    df.columns = ['Family_Members', 'Score']

pd.concat(all_dfs).reset_index(drop=True)

Out[16]: 
  Family_Members  Score
0        The Man      2
1       The Girl      4
2        The Boy      6
3     The Mother      7
4        The Son      3
5   The Daughter      4