数据帧转换和字符串连接

Question

I have a pandas dataframe that looks like this:

index, a, b
0, i, this
1, belong, is 
2, here, right
0, only, one
0, two, another
1, items, example

And I am trying to get the output so that running indexes get folded up and concatenates the strings:

index, a, b
0, i belong here, this is right
1, only, one
2, two items, another example

Any idea how to do this in a clean way?

Answer 1

You need create groups starting with 0 in index and cumulative sum, then aggregate join :

#if index is column
df = df.groupby(df['index'].eq(0).cumsum(), as_index=False).agg(' '.join)
#if index is not column
#df = df.groupby((df.index == 0).cumsum(), as_index=False).agg(' '.join)
print (df)
               a                b
0  i belong here   this is  right
1           only              one
2      two items  another example

Details :

print (df['index'].eq(0).cumsum())
0    1
1    1
2    1
3    2
4    3
5    3
Name: index, dtype: int32

#print ((df.index == 0).cumsum())
#[1 1 1 2 3 3]