CODEWITHSUNDEEP
pythonpandas
hengjuice
hengjuice

Reputation: 112

Changing row values based on previous row values

I have a dataframe which looks like this:

Province              Admissions
Eastern Cape          10
Private               3
Public                7
Free State            20
Private               15 
Public                5

I want to change the 'Private' and 'Public' to reference the Province. I want to achieve the following dataframe:

Province               Admissions
Eastern Cape           10
Eastern Cape-Private   3
Eastern Cape-Public    7
Free State             20
Free State-Private     15 
Free State-Public      5

I've actually already achieved this by the following code:

for row in range(0,len(df)):
    df['Province'] = np.where((df['Province'] == 'Private'), df['Province'].shift(1)+' '+ df['Province'], df['Province'])
    df['Province'] = np.where((df['Province'] == 'Public'), df['Province'].shift(2)+' '+ df['Province'], df['Province'])

However, I would like to do it in a more general approach, in case the order of the Private and Public is swapped. Right now Private comes before Public hence my method workds. Would appreciate any input!

Upvotes: 1

Views: 57

Answers (1)

BENY
BENY

Reputation: 323376

You can do mask and ffill create the adding array

s = df.Province.mask(df.Province.isin(['Private','Public'])).ffill()
df['Province'] = np.where(df.Province.isin(['Private','Public']), s + ' ' + df.Province, df.Province)

Upvotes: 2

Related Questions