Pandas group by filter results

Question

I am using pandas and getting data and using group by. However I am not sure how to filter out unwanted data from getting returned. I only want the count for 1's and not 0's.

print(df.groupby(item)['event'].apply(lambda x: (x==1).sum()).reset_index(name='count'))
hpb   count
0     55
1     36

I've also used the following and it's returning the same info for all columns.

print(patients.query("event==1 and hbp==1").groupby('hbp').count())

       age    cp    eject  ......
hbp
 1      36     36    36

My desired output is below.

hpb   count
1     36

Here is my data I am reading into the dataframe patients. I am trying to get the hbp data where event is 1 and hbp is 1 using query or group by. Currently I am getting both 1 and 0 counts but I only display the hbp 1 counts. see desired output above.

hbp event ..etc
1       1
0       1
1       1
1       1
0       1

Answer 1

Your solution is possible change with specify column after groupby and use sum or count or size, because processing 1 values:

print(patients.query("event==1 and hbp==1")
              .groupby('hbp')['event']
              .sum()
              .reset_index(name='count'))

Or first filter and then count hbp values by Series.value_counts:

print (patients.query("event==1 and hbp==1")['hbp']
               .value_counts()
               .reset_index(name='count'))

   hbp  count
0    1     36