计算组合列的唯一值，并将它们放入 dataframe

Question

I am trying to count the unique values for the combined columns, and put them in a dataframe, I have two columns. One column called 'Municipality' and the other 'Date'. The Municipality has 27 different names while Date has 151 dates for each Municipality, 4,077 rows all together. I can put these two in a data frame but I can not get the count. ie

days1 = (df['Municipality'])
days = days1[5247:9324].reset_index(drop=True)
ddate1 = (df['Date'])
ddate = ddate1[5247:9324].reset_index(drop=True)

frames = [days, ddate]

result = pd.concat(frames, axis = 1)
result

    Municipality    Date
0   Alta Floresta D'Oeste   2020-03-27
1   Alta Floresta D'Oeste   2020-03-28
2   Alta Floresta D'Oeste   2020-03-29
3   Alta Floresta D'Oeste   2020-03-30
4   Alta Floresta D'Oeste   2020-03-31
... ... ...
4072    Alto Paraíso    2020-08-20
4073    Alto Paraíso    2020-08-21
4074    Alto Paraíso    2020-08-22
4075    Alto Paraíso    2020-08-23
4076    Alto Paraíso    2020-08-24
4077 rows × 2 columns

the goal is to have each Municipality (27) and count the dates for each Municipality which should be 151 for each. New to this so thanks for any help.

Answer 1

I have a feeling you're looking for groupby.transform . With this, you will add a column that will count the dates, for each municipality.

import pandas as pd
result['date_count'] = result.groupby('Municipality')['Date'].transform('count')

---

result

            Municipality       Date  date_count
0  Alta Floresta D'Oeste 2020-03-27           5
1  Alta Floresta D'Oeste 2020-03-28           5
2  Alta Floresta D'Oeste 2020-03-29           5
3  Alta Floresta D'Oeste 2020-03-30           5
4  Alta Floresta D'Oeste 2020-03-31           5
5           Alto Paraíso 2020-08-20           5
6           Alto Paraíso 2020-08-21           5
7           Alto Paraíso 2020-08-22           5
8           Alto Paraíso 2020-08-23           5
9           Alto Paraíso 2020-08-24           5

In your own dataset, the 'date_count' column should say 151. You can have a further ready about groupby.transform here .