Python | Numpy np.assert_array_almost_equal() Method

Numpy is the fundamental package for scientific computing with Python. It is a powerful library for numerical computations, and it provides a Python interface to the Lapack and BLAS libraries along with many other useful functions. The np.assert_array_almost_equal() method is one of the many assertion methods available in Numpy.

What is np.assert_array_almost_equal() method?

The np.assert_array_almost_equal() method is used to check whether all elements of two given numpy arrays are equal or not. By default, the method checks for an absolute tolerance of 1e-8. This method can be used to compare two numpy arrays with rounding errors.

Syntax of np.assert_array_almost_equal() method

The syntax for using np.assert_array_almost_equal() method is as follows:

numpy.testing.assert_array_almost_equal(x, y, decimal=6, err_msg='', verbose=True)

Parameters:

• x: It represents the first array that needs to be compared with the second array.
• y: It represents the second array that needs to be compared with the first array.
• decimal: It represents the number of decimal places up to which the values need to be considered equal. By default, it is 6.
• err_msg: It represents the error message that needs to be printed in case of assertion failure.
• verbose: It represents a boolean flag that indicates whether to print diagnostic information in case of assertion failure.

Example

Consider the following example to understand the working of np.assert_array_almost_equal() method:

import numpy as np

# a and b are two numpy arrays
a = np.array([1.0542654, 2.2151215, 3.4563145])
b = np.array([1.0542652, 2.2151211, 3.4563144])

# check if the two numpy arrays are almost equal
np.testing.assert_array_almost_equal(a, b, decimal=6)

In the above example, we have two numpy arrays a and b. We are comparing these two numpy arrays using np.assert_array_almost_equal() method with a tolerance of 6 decimal places. If the elements of the two arrays differ by more than 6 decimal places, then an AssertionError will be raised.

Conclusion

In this article, we have learned about the np.assert_array_almost_equal() method in Numpy. We have seen the syntax of this method along with an example. The np.assert_array_almost_equal() method is a very useful method to compare two numpy arrays with rounding errors. It is a very powerful method to use for numerical computations.