dataframe: count multiple occurrences across all columns and output dataframe with same columns, single occcurrances as indexes I have a Pandas dataframe like this: