如何根据另一列值填充空索引或空行？

Question

I have a data frame:

             Date                Cities       Random_Number
Country
US           2020-01-01          LA           100
             2020-01-03          LA           150
UK           2020-01-01          Ldn          125
             2020-01-03          Birmingham   135

My desired data frame:

             Date                Cities       Random_Number
Country
US           2020-01-01          LA           100
US           2020-01-03          LA           150
UK           2020-01-01          Ldn          125
UK           2020-01-03          Birmingham   135

My aim is to have empty index row to be filled. Many thanks.

Answer 1

Because there are empty strings first convert them to missing values by Series.mask and then forward filling missing values by ffill :

df = df.reset_index()
print (df)
  Country        Date      Cities  Random_Number
0      US  2020-01-01          LA            100
1          2020-01-03          LA            150
2      UK  2020-01-01         Ldn            125
3          2020-01-03  Birmingham            135

df['Country'] = df['Country'].mask(df['Country'] == '').ffill()
print (df)
  Country        Date      Cities  Random_Number
0      US  2020-01-01          LA            100
1      US  2020-01-03          LA            150
2      UK  2020-01-01         Ldn            125
3      UK  2020-01-03  Birmingham            135

Answer 2

can you try this

data.fillna(method='ffill')

Got your desired output.

Answer 3

You can try df.head(4) to 'ungroup' the DataFrame.

df = pd.DataFrame([['US', '2020-01-01', 'LA', 100],
                   ['US', '2020-01-03', 'LA', 150],
                   ['UK', '2020-01-01', 'Ldn', 125],
                   ['UK', '2020-01-03', 'Birmingham', 135]],
                  columns=['Country', 'Date', 'Cities', 'Random_Number']).groupby('Country')
print(df)

Result:

             Date                Cities       Random_Number
Country
US           2020-01-01          LA           100
             2020-01-03          LA           150
UK           2020-01-01          Ldn          125
             2020-01-03          Birmingham   135  

Ungroup:

print(df.head(4))

Result:

  Country        Date      Cities  Random_Number
0      US  2020-01-01          LA            100
1      US  2020-01-03          LA            150
2      UK  2020-01-01         Ldn            125
3      UK  2020-01-03  Birmingham            135