selecting values from different columns, checking for nan and forming new column based on conditions

Question

I have a df

I have to form a new column 'AssistanceFactor' such that its value is 1 if Doc2 is NaN and DayType is 'Holiday'

1 if Doc2 is not Nan and DayType is 'Holiday'

0.75 if Doc2 is not Nan and DayType is not 'Holiday'

1 if Doc2 is not NaN and DayType is not 'Holiday'

What would be the best method for this? I tried iterrows and conditions as well as np.select /where. But I am not getting the hang of it.

Thanks in advance

So basically you want 0.75 if Doc2 is not NaN and DayType not Holiday and 1 otherwise — mozway, Apr 04 '22 at 09:10

score 1 · Accepted Answer · answered Apr 04 '22 at 09:09

1

Because 3 conditions return 1 and one return 0.75 is possible test only 3rd condition:

df['AssistanceFactor'] = np.where(df.Doc2.notna() & df.DayType.ne('Holiday'), 0.75, 1)

answered Apr 04 '22 at 09:09

jezrael

So, do you consider it a duplicate or not? ;) – mozway Apr 04 '22 at 09:13
@mozway - for 90%, there is not test `notna` ;) – jezrael Apr 04 '22 at 09:14
I tried this, but I get an error saying length of values not matching length of index. Any solutions? – dratoms Apr 04 '22 at 09:31
@dratoms - there is compared `df` and assign to `df` ? all same Dataframe? – jezrael Apr 04 '22 at 09:36

1 Answers1