Pandas: how to group rows with consecutively repeating values in columns?-CodePudding

I have a data frame df `

df=pd.DataFrame([['1001',34.3],['1009',34.3],['1003',776],['1015',18.95],['1023',18.95],['1007',18.95],['1009',18.95],['1037',321.2],['1001',344.2],['1016',3.2],['1017',3.2],['1027',344.2]],columns=['id','amount'])

    id      amount  
0   1001    34.30  
1   1009    34.30  
2   1003    776.00  
3   1015    18.95   
4   1023    18.95
5   1007    18.95
6   1009    18.95
7   1037    321.20
8   1001    344.20 
9   1016    3.20
10   1017    3.20
11   1027    344.20

I would likw to have df_new grouped by consecutively repeating values in column 'amount' by first value:

    id      amount  
0   1001    34.30  
2   1003    776.00  
3   1015    18.95   
7   1037    321.20
8   1001    344.20 
9   1016    3.20
11   1027    344.20

CodePudding user response：

here is one way to do it


# take a difference b/w the amount of two consecutive rows and then
# choose rows where the difference is not zero

out= df[df['amount'].diff().ne(0) ]

out

id  amount
0   1001    34.30
2   1003    776.00
3   1015    18.95
7   1037    321.20
8   1001    344.20
9   1016    3.20
11  1027    344.20