在 python 中出错为“**系列的真值不明确”

Question

I am getting the below error when I am trying to perform an if statement and assigning it back to a dataframe column

The truth value of a Series is ambiguous. Use a.empty, a.bool(), a.item(), a.any() or a.all().

def cal_test():
    if IP['Type'] == 'ABC':
        (IP['tt'] + ' ' + IP[abc])
    else:
        IP[tt]

IP['column_name']  =   cal_test()

Answer 1

IP['column_name'] = IP['tt']
IP.loc[IP['Type'] == 'ABC', 'column_name'] = IP['tt'] + ' ' + IP['abc']

Answer 2

The expression:

IP['Type'] == 'ABC'

means to compare a column "Type" of dataframe IP against a string "ABC" , which of course ambiguous -- do you want it to fail because you are comparing apple to orange or do you mean if any of the item in the column match the string?

The right way to compare itemize is IP['Type'].eq('ABC') which returns you a column (ie, series) of boolean value, or you can use any or all to mean any value in the column or all values in the column do the match.

And, your function returns nothing. So you cannot possibly put anything "back" to the dataframe. Try do this instead:

IP['column_name'] = IP.apply(
    lambda row: (row['tt']+' '+row['abc']) if row['Type'] == 'ABC' else
                row['tt'],
    axis=1
)