pandas dataframe 列上的条件聚合，将“n”行合并为 1 行

Question

I have the following pandas dataframe:

START   NAME
5.11    name1
9.1     name1
10.86   name1
12.61   name2
14.86   name2
23.11   name2
25.36   name1
26.61   name1
28.36   name2
31.61   name2
32.86   name1
35.61   name1
44.61   name1
46.36   name2

I would this merged by name as follows:

START   END     NAME
5.11    12.61   name1
12.61   25.36   name2
26.61   28.36   name1
28.36   32.86   name2
32.86   46.36   name1
46.36   total   name2

I tried something like this:

df2 = df.copy()
df2 = df2.rename({"name": "temp"}).reset_index()
grp = (df2['name'] != df2['name'].shift()).cumsum().rename('group')
df2 = df2.groupby(['name', grp], sort=False)

But this does not produce the desired output. Any help is appreciated

thanks

Answer 1

1. use shift to compare the row's content is same with the next row
2. keep the NAME that is not the same as the next row's NAME.
3. use shift(-1) to assign the NAME's END.

cond = (df['NAME'] != df['NAME'].shift(1))
dfn = df[cond].copy()
dfn['END'] = dfn['START'].shift(-1).fillna('total')
dfn[['START', 'END', 'NAME']]