熊猫删除双重条件的重复项

Question

consider the following DF

    import pandas as pd
    df = pd.DataFrame({'ID': [1,1,1,1,2,2,2,2], 
    'Course': 
    ['English','English','English','History','Science', 'Science', 'Science','Math'],
    'Status':
    ['Attended', 'Requested', 'Partially Attended', 'No show',
    'Requested','Attended','Partially Attended','No show']})
    df.set_index(['ID'])
    print(df)

Course  Status
ID      
1   English Attended
1   English Requested
1   English Partially Attended
1   History No show
2   Science Requested
2   Science Attended
2   Science Partially Attended
2   Math    No show

I'm trying to work out a way to remove duplicates based on the following 3 assumptions.

1. ID occurs more than once.
2. Where the ID occurs more than once the course has to be the same. (so 1, History and 2, Math are fine to stay)
3. if a match is found then I want to only drop in the instance where a course is completed and there is a request, dropping the row with the request. A no-show and partially attended are fine.

I'm currently studying and taking DataCamps Python and pandas courses so I'm familiar with groupby, aggregate, sort functions where I can drop the later or earlier duplicate with time-series data. I have no idea how to apply conditions or logic to the drop functions. I've searched this forum for similar functions but I've not been apply anything to my own DF.

my desired result is as follows:

Course  Status
ID      
1   English Attended
1   English Partially Attended
1   History No show
2   Science Attended
2   Science Partially Attended
2   Math    No show

Answer 1

not duplicated or not Requested

df[~df.duplicated(['ID', 'Course'], keep=False) | df.Status.ne('Requested')]

    Course  ID              Status
0  English   1            Attended
2  English   1  Partially Attended
3  History   1             No show
5  Science   2            Attended
6  Science   2  Partially Attended
7     Math   2             No show

---

pandas.DataFrame.duplicated

Identifies if things are duplicates. I pass a list of column names to use to determine duplicity. By using keep=False I specify that I want to count the first or last occurrence as a duplicate as well.

df.duplicated(['ID', 'Course'], keep=False)

0     True
1     True
2     True
3    False
4     True
5     True
6     True
7    False
dtype: bool

However, if it is a duplicate, also check if it is Requested

df.Status.ne('Requested')

0     True
1    False
2     True
3     True
4    False
5     True
6     True
7     True
Name: Status, dtype: bool

So we want rows that are either not duplicates, and if they are at least don't have Status equal to Requested