I have attached sample data frame in which I need top 3 companies based on count and discard the rest.
kindly suggest a robust way to do that.so far I have pivoted my entire actual data and sorted count.