Return real parts if input is complex with all imaginary parts close to zero in Python

The numpy.real_if_close() function returns the real parts of a complex array when the imaginary parts are close to zero. "Close to zero" is defined as tol * machine_epsilon, where tol is the tolerance parameter.

Syntax

numpy.real_if_close(a, tol=100)

Parameters

The function accepts the following parameters ?

• a ? Input array (complex or real)
• tol ? Tolerance in machine epsilons (default: 100)

Example

Let's see how to extract real parts when imaginary parts are negligible ?

import numpy as np

# Creating a complex array with very small imaginary parts
arr = np.array([2.1 + 4e-14j, 5.2 + 3e-15j])

# Display the array
print("Our Array...")
print(arr)

# Check array properties
print("\nDimensions:", arr.ndim)
print("Datatype:", arr.dtype)
print("Shape:", arr.shape)

# Extract real parts if imaginary parts are close to zero
result = np.real_if_close(arr, tol=1000)
print("\nResult...")
print(result)
print("Result datatype:", result.dtype)

Our Array...
[2.1+4.e-14j 5.2+3.e-15j]

Dimensions: 1
Datatype: complex128
Shape: (2,)

Result...
[2.1 5.2]
Result datatype: float64

Different Tolerance Values

The tolerance parameter determines how close to zero the imaginary parts must be ?

import numpy as np

# Array with larger imaginary parts
arr = np.array([3.0 + 1e-10j, 4.0 + 1e-12j])

print("Original array:")
print(arr)

# With low tolerance - keeps complex
print("\nWith tol=100:")
print(np.real_if_close(arr, tol=100))

# With high tolerance - returns real
print("\nWith tol=1000000:")
print(np.real_if_close(arr, tol=1000000))

Original array:
[3.+1.e-10j 4.+1.e-12j]

With tol=100:
[3.+1.e-10j 4.+1.e-12j]

With tol=1000000:
[3. 4.]

Real Input Arrays

When the input is already real, the function returns the same array ?

import numpy as np

# Real array
real_arr = np.array([1.5, 2.7, 3.9])

print("Real array:")
print(real_arr)

result = np.real_if_close(real_arr)
print("\nResult:")
print(result)
print("Same object?", real_arr is result)

Real array:
[1.5 2.7 3.9]

Result:
[1.5 2.7 3.9]
Same object? True

Conclusion

Use numpy.real_if_close() to extract real parts from complex arrays when imaginary components are negligible. Adjust the tolerance parameter to control the sensitivity of the "close to zero" condition.