How to get the Cardinality of Columns in Pandas

A quick and efficient way to get the cardinality of columns, i.e. the number of unique values, is to run this line of code.

Let’s create a data frame:

import pandas as pd
import numpy as np
 
df = pd.DataFrame({'id':list(range(10)),
                   'A':[10,9,8,7,np.nan,np.nan,20,15,12,np.nan], 
                   'B':["A","B","A","A",np.nan,"B","A","B",np.nan,"A"],
                   'C':[np.nan,"BB","CC","BB","BB","CC","AA","BB",np.nan,"AA"],
                   'D':[np.nan,20,18,22,18,17,19,np.nan,17,23]})
df
 

Let’s get the number of unique values by column.

# if you want to ignore NAs
df.nunique(dropna=True)

id    10
A      7
B      2
C      3
D      6
dtype: int64

# if you want to count the as a different unique value the NA
df.nunique(dropna=False)
 

id    10
A      8
B      3
C      4
D      7
dtype: int64