Database-style operations with Pandas

You may have noted that the “tabular” data is similar to what is usually stored in a database. A database is usually indexed using a primary key, and the various columns can have different data types, just like in a pd.DataFrame. 

The efficiency of the index operations in Pandas makes it suitable for database style manipulations, such as counting, joining, grouping, and aggregations.