How do I create a custom ufunc in NumPy?

How do I create a custom ufunc in NumPy?

To create a custom ufunc in NumPy, you can use the numpy.frompyfunc function. This allows you to define a Python function and convert it into a ufunc.

Can you give me an example of a custom ufunc?

Sure! Here's a simple example:

import numpy as np

def my_func(x):
    return x ** 2

custom_ufunc = np.frompyfunc(my_func, 1, 1)
result = custom_ufunc(np.array([1, 2, 3]))
print(result)

This will output: [1, 4, 9].

What are the benefits of using ufuncs?

Ufuncs are optimized for performance and can operate on entire arrays without the need for explicit loops, making them faster and more efficient.

Can I use ufuncs with multi-dimensional arrays?

Yes, ufuncs can be applied to multi-dimensional arrays. They automatically handle broadcasting, allowing you to perform operations on arrays of different shapes.