Панды меняют каждый ряд на DF

for i, row in df.iterrows():
    ifor_val = something
    if <condition>:
        ifor_val = something_else
    df.at[i,'ifor'] = ifor_val
Arex