I want to check if a column value changes for each unique value of "Name".

The dataframe I have looks like this

My desired output is an extra column which either says "True" if a value has changed, and "False" if it didn't change. Desired output would look like this:

| Name | Is finished? | Change|
| -------- | -------------- |---|
| Paul    |      No|  False|
| Paul   | No|False|
| Paul    | No|False|
| John   | No|False|
| John    | No|False|
| John   | Yes|True|
| John| Yes|False|
| Ringo   | No|False|
| Ringo   | No|False|
| Ringo   | Yes|True|
| Ringo   | Yes|False|
| Ringo   | Yes|False|

CodePudding user response：

Let us do groupby and shift to compare the current and previous row in Is finished column

c = 'Is finished?'
df['Change'] = df[c].eq('Yes') & df.groupby('Name')[c].shift().eq('No')

     Name Is finished?  Change
0    Paul           No   False
1    Paul           No   False
2    Paul           No   False
3    John           No   False
4    John           No   False
5    John          Yes    True
6    John          Yes   False
7   Ringo           No   False
8   Ringo           No   False
9   Ringo          Yes    True
10  Ringo          Yes   False
11  Ringo          Yes   False

CodePudding user response：

import pandas as pd
df = pd.DataFrame({'Name': {0: 'Paul',1: 'Paul',2: 'Paul',3: 'John', 4: 'John',5: 'John',6: 'John',7: 'Ringo',8: 'Ringo',9: 'Ringo',10: 'Ringo',11: 'Ringo'},'Is finished?': {0: 'No',1: 'No',2: 'No',3: 'No',4: 'No',5: 'Yes',6: 'Yes',7: 'No',8: 'No',9: 'Yes',10: 'Yes',11: 'Yes'}})

result = df.drop(df.index)
for _, g in df.groupby('Name'):
    g['Shift'] = g['Is finished?'].shift(1)
    g['Changed?'] = g['Is finished?'] != g['Shift']
    result = pd.concat([result, g.drop(columns=['Shift'])], axis=0)
    
print(result)

Output:

     Name Is finished? Changed?
3    John           No     True
4    John           No    False
5    John          Yes     True
6    John          Yes    False
0    Paul           No     True
1    Paul           No    False
2    Paul           No    False
7   Ringo           No     True
8   Ringo           No    False
9   Ringo          Yes     True
10  Ringo          Yes    False
11  Ringo          Yes    False