value_count() to get numbers of unique values and its frequencies as a series

Pandas

We will get unique values and its frequency as series.

Parameters

normalize : Default is False, If set to True then give realtive frequency ( instead of numbers ) of unique values.
sort : Default is True, sort by frequency of unique values
ascending :Default is False, Sort in ascending order
bins : integer or non-uniform width or interval index
dropna: (default True) Not to include counts for NaN

Examples with Parameters

count_values() will return all unique values with number of occurrence. We have different classes in our data column CLASS1.

import pandas as pd 
my_dict={'NAME':['Ravi','Raju','Alex','Ron','King','Jack'],
         'ID':[1,2,3,4,5,6],
         'MATH':[80,40,70,70,60,30],
         'CLASS1':['Four','Three','Three','Four','Five','Three']}
my_data = pd.DataFrame(data=my_dict)
print(my_data['CLASS1'].value_counts())

Output

Three    3
Four     2
Five     1

normalize

By default it is False. By making it True ( normalize=True ) instead of number of occurrence we will display relative frequencies of unique values.

print(my_data['CLASS1'].value_counts(normalize=True))

Output

Three     0.500000
Fourth    0.333333
Five      0.166667

sort

By default it is True ( sort=True ). Sorting by frequency of unique values.

my_data = pd.DataFrame(data=my_dict)
print(my_data['CLASS1'].value_counts(sort=False))

Output

Five     1
Three    3
Four     2

ascending

By default ascending=False , we will set it to ascending=True

print(my_data['CLASS1'].value_counts(ascending=True))

Output

Five     1
Four     2
Three    3

bins

We can create segments using bins. Let us create fixed width bins by using integer.

print(my_data['MATH'].value_counts(bins=3))

Output

(63.333, 80.0]                  3
(29.948999999999998, 46.667]    2
(46.667, 63.333]                1

non-uniform width bins

print(my_data['MATH'].value_counts(bins=[1,50,70,90]))

Output


(50.0, 70.0]     3
(0.999, 50.0]    2
(70.0, 90.0]     1

dropna

Default value is True, Don't include counts of NaN. We will set the value to False ( dropna=False ) to include NaN vlaues.

import pandas as pd 
import numpy as np
my_dict={'NAME':['Ravi','Raju','Alex','Ron','King','Jack'],
         'ID':[1,2,3,4,5,6],
         'MATH':[80,40,70,70,np.nan,30],
         'CLASS1':['Four','Three','Three','Four','Five','Three']}
my_data = pd.DataFrame(data=my_dict)
print(my_data['MATH'].value_counts(dropna=False))

Output

We can set it to True ( dropna=True)

print(my_data['MATH'].value_counts(dropna=True))

Output

Unique data

Pandas groupby cut() segment and sort data values into bins

Numpy arrays Python & MySQL Python- Tutorials

Subscribe to our YouTube Channel here

value_count(): Pandas DataFrame

Parameters

Examples with Parameters

normalize

sort

ascending

bins

dropna

Subscribe