How to convert a result out put in series back to a data frame?

Question

I would like to convert a result output in series back to a dataframe, so that I can do further analysis. This is my data

import random
import pandas as pd

random.seed(100)
datadict = {"AB" : ["A", "B"]*6, "CDE" : ["C", "D", "E"]*4, "1234": 
[random.randint(1,5) for x in range(12)], "value": 
[random.randrange(1,100) for x in range(12)]}
df = pd.DataFrame(datadict)

df.groupby(["AB", "CDE", "1234"]).value.mean()

Now, I get the results in series

AB  CDE  1234
A   C    2       95
         4       27
    D    1       99
         4       85
    E    1       30
         4       34
B   C    2        7
         5       40
    D    4       59
         5       43
    E    1       27
         3       83

Now, I would like to convert this result back to a dataframe, so that the data would looks like this

AB  CDE  1234  mean
A   C    2       95
A   C    4       27
A   D    1       99
A   D    4       85
A   E    1       30
A   E    4       34
B   C    2        7
B   C    5       40
B   D    4       59
B   D    5       43
B   E    1       27
B   E    3       83

In this way, I can do further analysis. Can anyone tell how to do it?

score 1 · Answer 1 · answered Sep 17 '18 at 21:37

You can use df.reset_index(). This will add a new integer index like so:

   AB CDE  1234  value
0   A   C     1   94.0
1   A   C     5   19.0
2   A   D     1   21.0
3   A   D     4   21.0
4   A   E     1   63.0
5   A   E     4   31.0
6   B   C     3   96.0
7   B   C     4   77.0
8   B   D     3   34.5
9   B   E     3   18.0
10  B   E     5   95.0

How to convert a result out put in series back to a data frame?

1 Answers1