value_counts() for frequency in Pandas - Practice Problems & Coding Challenges

Challenge - 5 Problems

❓ Predict Output

intermediate

Output of value_counts() on a simple Series

What is the output of the following code?

import pandas as pd
s = pd.Series(['apple', 'banana', 'apple', 'orange', 'banana', 'banana'])
result = s.value_counts()
print(result)

Pandas

import pandas as pd
s = pd.Series(['apple', 'banana', 'apple', 'orange', 'banana', 'banana'])
result = s.value_counts()
print(result)

A

orange    3
banana    2
apple     1
dtype: int64

B

apple     2
banana    2
orange    2
dtype: int64

C

banana    3
apple     2
orange    1
dtype: int64

D

apple     3
banana    2
orange    1
dtype: int64

💡 Hint

Count how many times each fruit appears in the list.

The value_counts() method counts how many times each unique value appears. 'banana' appears 3 times, 'apple' 2 times, and 'orange' once.

❓ data_output

intermediate

Number of unique values counted by value_counts()

Given the Series below, how many unique values does value_counts() return?

import pandas as pd
s = pd.Series([10, 20, 10, 30, 20, 20, 40])
counts = s.value_counts()

Pandas

import pandas as pd
s = pd.Series([10, 20, 10, 30, 20, 20, 40])
counts = s.value_counts()
print(len(counts))

A 4

B 7

C 5

D 3

💡 Hint

Count how many different numbers appear in the Series.

The unique values are 10, 20, 30, and 40, so value_counts() returns 4 counts.

🔧 Debug

advanced

Identify the error in using value_counts() on a DataFrame

What error will this code produce?

import pandas as pd
df = pd.DataFrame({'A': [1, 2, 2, 3], 'B': ['x', 'y', 'x', 'z']})
result = df.value_counts()
print(result)

Pandas

import pandas as pd
df = pd.DataFrame({'A': [1, 2, 2, 3], 'B': ['x', 'y', 'x', 'z']})
result = df.value_counts()
print(result)

A KeyError: 'value_counts'

B AttributeError: 'DataFrame' object has no attribute 'value_counts'

C TypeError: unhashable type: 'Series'

D No error, prints counts of unique rows

💡 Hint

Check if DataFrame has value_counts method and what it returns.

Since pandas 1.1.0, DataFrame has a value_counts() method that counts unique rows. So no error occurs and it prints counts of unique rows.

🚀 Application

advanced

Using value_counts() with normalization

What does the normalize=True option do in value_counts()?
Given:

import pandas as pd
s = pd.Series(['cat', 'dog', 'cat', 'bird', 'dog', 'dog'])
result = s.value_counts(normalize=True)
print(result)

Pandas

import pandas as pd
s = pd.Series(['cat', 'dog', 'cat', 'bird', 'dog', 'dog'])
result = s.value_counts(normalize=True)
print(result)

A Shows the counts as percentages of total values

B Sorts the counts in ascending order

C Filters out values with counts less than 2

D Returns the counts as cumulative sums

💡 Hint

Think about what 'normalize' means in statistics.

normalize=True returns the relative frequencies (proportions) instead of raw counts.

🧠 Conceptual

expert

Interpreting value_counts() output with missing values

Consider this Series with missing values:

import pandas as pd
s = pd.Series(['red', 'blue', None, 'red', 'green', None, 'blue'])
result = s.value_counts(dropna=False)
print(result)

What does the output show compared to dropna=True?

Pandas

import pandas as pd
s = pd.Series(['red', 'blue', None, 'red', 'green', None, 'blue'])
result = s.value_counts(dropna=False)
print(result)

A Raises an error because None cannot be counted

B Includes counts of missing values (NaN) in the output

C Excludes missing values and only counts non-null entries

D Replaces missing values with zero before counting

💡 Hint

Check what dropna=False means for counting nulls.

dropna=False includes NaN values in the counts, so missing values are counted and shown in the output.