CODEWITHSUNDEEP
pythonpandasnumpydataframe
kodachrome
kodachrome

Reputation: 350

Multiplying a dataframe by a column... but values are strings?

I have a Dataframe that includes binary variables about respondents' behavior and the weight associated with each respondent. I'd like to multiply the scores by each respondents' weight so I can easily get a weighted average for the total behavior.

The easiest thing would be to multiply the weight column against another column in a loop, as in df.columns[761]*df.columns[i]. However, when I try to do, it throws an error of:

'can't multiply sequence by non-int of type 'str'.'

I shouldn't have any strings, but in the off-chance there are, I tried to convert the df to numeric, like so df.apply(pd.to_numeric, errors='coerce').

But the problem still remains. I'm at my wits' end. Is there a workaround? Should I go row by row (and if so, do I need to loop through every column, or is there a nice clean way?).

Upvotes: 0

Views: 1050

Answers (1)

Quentin
Quentin

Reputation: 700

You could always break apart your dataframe.

for col in df.columns:
    for index, k in enumerate(df[col]):
        try:
            float(k)
        except:
            # Print out the row number, col and row value that's failing
            print(index, col, k)

It's entirely possible you've got strings/none-types that are causing your multiplication.

There's also df[col].apply(float) but it won't catch those errant rows.

Upvotes: 2

Related Questions