Python program to filter perfect squares in a given series

A perfect square is a number that can be expressed as the product of an integer with itself. In this tutorial, we'll learn how to filter perfect squares from a Pandas Series using different approaches.

Given a series with numbers, we want to identify and filter only the perfect squares ?

Input and Expected Output

Input −

0    14
1    16
2    30
3    49
4    80

Output −

1    16
3    49

Method 1: Using Mathematical Check with sqrt()

We can check if a number is a perfect square by taking its square root and verifying if it's an integer ?

import pandas as pd
import math

# Create a series
data = pd.Series([14, 16, 30, 49, 80])
print("Given series:")
print(data)

# Find perfect squares using mathematical approach
perfect_squares = []
for num in data:
    sqrt_val = math.sqrt(num)
    if sqrt_val == int(sqrt_val):
        perfect_squares.append(num)

# Filter the series
result = data[data.isin(perfect_squares)]
print("\nPerfect squares in the series:")
print(result)

Given series:
0    14
1    16
2    30
3    49
4    80
dtype: int64

Perfect squares in the series:
1    16
3    49
dtype: int64

Method 2: Using Lambda Function with Filter

A more concise approach using lambda function and the mathematical check ?

import pandas as pd
import math

# Create a series
data = pd.Series([14, 16, 30, 49, 80])
print("Given series:")
print(data)

# Use lambda function to filter perfect squares
perfect_squares = list(filter(lambda x: x == int(math.sqrt(x) + 0.5)**2, data))
result = data[data.isin(perfect_squares)]

print("\nPerfect squares in the series:")
print(result)

Given series:
0    14
1    16
2    30
3    49
4    80
dtype: int64

Perfect squares in the series:
1    16
3    49
dtype: int64

Method 3: Using Boolean Indexing

A direct approach using boolean indexing with Pandas ?

import pandas as pd
import math

# Create a series
data = pd.Series([14, 16, 30, 49, 80])
print("Given series:")
print(data)

# Create boolean mask for perfect squares
is_perfect_square = data.apply(lambda x: math.sqrt(x) == int(math.sqrt(x)))
result = data[is_perfect_square]

print("\nPerfect squares in the series:")
print(result)

Given series:
0    14
1    16
2    30
3    49
4    80
dtype: int64

Perfect squares in the series:
1    16
3    49
dtype: int64

How It Works

The key concept is checking if the square root of a number is an integer:

• Mathematical approach: Calculate sqrt(x) and check if it equals int(sqrt(x))

• Alternative check: Verify if x == int(sqrt(x) + 0.5)**2 (handles floating point precision)

• Boolean indexing: Use apply() with a lambda function to create a boolean mask

Comparison

Conclusion

Use boolean indexing with apply() for the most Pandas-native approach. The mathematical check using sqrt() is the most straightforward method for filtering perfect squares from a series.