GroupBy Two Columns and then find Min of Third Column in Pandas

Question

I have a dataset that looks like this:

SUBJECTID   session    value  attempt
  1          home:1:1    5      1
  1          home:2:1    2      1
  1          home:2:2    4      2

For every SUBJECTID and every session I want to only include the data for the lowest number attempt (note this isn't always attempt 1)

I've tried the following in pandas with no luck

group_a = totaltimes_a.groupby('SUBJECT')['session']
min_value = group_a.attempt.min()
totaltimes_a = totaltimes_a.merge(min_value, on='session',suffixes=('', '_min'))
totaltimes_a = totaltimes_a[totaltimes_a.attempt==totaltimes_a.attempt_min].drop('attempt_min', axis=1)

Answer 1

I hope I've understood your question right:

df["attempt_min"] = df.groupby(["SUBJECTID", "session"])["attempt"].transform(
    "min"
)
print(df)

Prints:

   SUBJECTID   session  value  attempt  attempt_min
0          1  home:1:1      5        1            1
1          1  home:2:1      2        1            1
2          1  home:2:2      4        2            2