How to get first value in groupby , TypeError: first() missing 1 required positional argument: 'offset'

Question

I have df, like this

States Counts
AK     one
AK     two
AK     one
LO     one
LO     three
LO     three

trying to get most occured counts for each status

my code:

 df.groupby('States')['Counts'].value_counts().first(), gives

TypeError: first() missing 1 required positional argument: 'offset'

expected output:

  States Counts
  AK     one
  LO     three

score 5 · Accepted Answer · answered Sep 02 '18 at 07:54

5

Use lambda function:

df = df.groupby('States')['Counts'].apply(lambda x: x.value_counts().index[0]).reset_index(name='val')
print (df)
  States    val
0     AK    one
1     LO  three

answered Sep 02 '18 at 07:54

jezrael

822,522
95
1,334
1,252

thank you @jezrael, any other shorter way ? – Pyd Sep 02 '18 at 07:56
@pyd - hmmm, I see dupe with `mode`, not sure how working it and maybe `x.mode.iat[0]` is necessary – jezrael Sep 02 '18 at 08:03
4

`df.groupby('States')['Counts'].apply(lambda x : x.mode())` – Pyd Jan 20 '19 at 16:04

How to get first value in groupby , TypeError: first() missing 1 required positional argument: 'offset'

1 Answers1