10.2.1 Indexing rules

Similar to the way dictionaries use keys to address a value, pandas dataframes use row labels—the dataframe index—and column labels to access individual values:

AF.loc['R1', 'C2']      # this returns 2.0 

Or to generate a subframe:

AF.loc[['R1','R2'],['C1','C2']]

Resulting in:

   C1   C2
R1  1  2.0
R2  4  5.0

You can also address a complete row by using the index label:

AF.loc['R1']

This returns a pandas Series object:

C1    1.0
C2    2.0
C3    3.0
Name: R1, dtype: float64

If loc or iloc are called with list arguments or slices, the result is a dataframe.

In that way, an individual dataframe element can also be addressed as follows:

AF.loc['R1'].loc['C1']    # returns 1.0

An entire column is addressed directly as follows:

AF['C1']

This again returns a pandas Series object:

R1    1.0
R2    4.0
Name: C1, dtype: float64

Alternatively...