By default, at the end of a groupby operation, pandas puts all of the grouping columns in the index. The as_index parameter in the groupby method can be set to False to avoid this behavior. You can chain the reset_index method after grouping to get the same effect as done in step 3. Let's see an example of this by finding the average distance traveled per flight from each airline:

>>> flights.groupby(['AIRLINE'], as_index=False)['DIST'].agg('mean') 
                                                        .round(0)

Take a look at the order of the airlines in the previous result. By default, pandas sorts the grouping columns. The sort parameter exists within the groupby method and is defaulted to True. You may set it to False to keep the order of the grouping columns the same as how they are encountered in the dataset. You also get a small performance improvement by not sorting your data.