Finding the column name of row which has highest value while comprehending the row based on max column value

Question

I'm pretty new to Python. Im trying to define a function for the below set of data. Sample data

I'm first looking for the max value in cell 3 . Based on this max value I'm checking which column given in the last 8 columns of the data has the highest value against it.

For example, As per the given data, Max value in cell 3 is 1470758 . Now I'm checking which columns from column cell 9 to cell 16 has the highest value against this max value. In the case of this sample data the answer should be cell 10 with a value of 7201. So the output should be cell 10.

Here's my code:

def winner(filename):
    data=pd.read_csv(filename, sep=',')
    maxC=data.npop.max()
    while data.loc[data['npop']]==maxC:
        data3=data.iloc[:,-8:].max()
#missing code
winner("demo.csv")

Please help. I didn't understand what I should be writing in the missing code section.

Answer 1

Line by line explanation of code is given with comments.

Try this :

def winner(filename):
    df=pd.read_csv(filename, sep=',')  # Read the csv into dataframe.
    column_names = list(df.columns.values) # Get list of column names

    max_col3_index = df['col3'].idmax()  # this will return the index of max value in `col3` column.

    row_data = df.loc[max_col3_index, column_names[-8:]]  # get series of data present in last 8 columns at above index.

    final_column_name = row_data.idxmax()  # Get the name of column having max value in above series.

    print(final_column_name)