How do I create a Pivot table in Pandas using the default index as the pivot index?

Question

I have the below data frame:

   Dataset Scores
0   times_pregnant  6.000
1   plasma_glucose_concentration    148.000
2   diastolic_blood_pressure    72.000
3   triceps_thickness   35.000
4   2_hour_serum_insulin    0.000
... ... ...
7828    BMI 30.400
7829    diabetes_pedigreen  0.315
7830    age 23.000
7831    diabetes    0.000
7832    times_pregnant  1.000

I want to turn the "Dataset" column contents into column headers and have the "Scores" column as values. When I try to pivot without specifying an index it just gives me the average of all the scores under each column header, but I want each score listed under each column header.

Current result:

Dataset 2_hour_serum_insulin    BMI age diabetes    diabetes_pedigreen  diastolic_blood_pressure    plasma_glucose_concentration    times_pregnant  triceps_thickness
Scores  79.799479   31.992578   33.240885   0.348958    0.471876    69.105469   120.894531  3.845052    20.536458

Desired result:

times_pregnant  BMI    Age
2               23.50  45
4               30.40  23
3               41.50  23
5               25.40  42

Pandas gives you a default unique index but I don't understand how to set this as my index to allow all my scores to be listed out.

Is pivoting the best way to do this?

Answer 1

Assuming each woman has the same amount of scores, you could add an index for each woman and pivot on it (df is your dataframe):

 u = df['Dataset'].nunique()
 df['women_idx'] = [int(x/u) for x in range(df.shape[0])]

 df.pivot(columns='Dataset', index='women_idx')