numpy Part 7 – Arbitrary Data, ones and zeros Array

Today we’ll learn how to create an array filled with ones, zeros or arbitrary data.

Sometimes you may want to create an array filled with ones or zeros. There are two methods that you can use to achieve that: ones and zeros. Here’s how they work:

The ones method creates an array filled with ones. It takes the shape of the array as an argument:

>>> a = np.ones((3, 6))
>>> print(a)
[[1. 1. 1. 1. 1. 1.]
 [1. 1. 1. 1. 1. 1.]
 [1. 1. 1. 1. 1. 1.]]

Learn how to make beautiful GUI apps

in Python using the Kivy framework.

Comprehensive, for Kivy beginners, easy to follow.

Get the book here (PDF) or on Amazon:

ebook / paperback (black and white) / paperback (full color)

The zeros method works in a similar way, but it fills the array with zeros:

>>> b = np.zeros((3, 6))
>>> print(b)
[[0. 0. 0. 0. 0. 0.]
 [0. 0. 0. 0. 0. 0.]
 [0. 0. 0. 0. 0. 0.]]

For 1-dimensional arrays you can pass the shape as a 1-tuple:

>>> c = np.zeros((5,))
>>> print(c)
[0. 0. 0. 0. 0.]

or an integer:

>>> d = np.zeros(5)
>>> print(d)
[0. 0. 0. 0. 0.]

By default the ones and zeros are float numbers. If you want them to be a different type, you can pass the dtype argument set to the type of your choice. Suppose we need integers. We can do it this way:

>>> e = np.ones((3, 6), dtype = int)
>>> print(e)
[[1 1 1 1 1 1]
 [1 1 1 1 1 1]
 [1 1 1 1 1 1]]

or this way:

>>> f = np.zeros((2, 4, 3), dtype = np.int8)
>>> print(f)
[[[0 0 0]
  [0 0 0]
  [0 0 0]
  [0 0 0]]

 [[0 0 0]
  [0 0 0]
  [0 0 0]
  [0 0 0]]]

Your Panda3D Magazine

Make Awesome Games and Other 3D Apps

with Panda3D and Blender using Python.

Cool stuff, easy to follow articles.

Get the magazine here (PDF).

There are two other methods that you can use to create arrays filled with ones or zeros, ones_like and zeros_like. They take an existing array as an argument and create a new array of the same shape. Here’s how they work. First the ones_like method:

>>> h = np.array([[3.2, 4.6, 6.4], [4.7, 7.9, 5.3]])
>>> i = np.ones_like(h)
>>> print(i)
[[1. 1. 1.]
 [1. 1. 1.]]

And now the zeros_like method:

>>> j = np.zeros_like(h, dtype = int)
>>> print(j)
[[0 0 0]
 [0 0 0]]

Python Jumpstart Course

Learn the basics of Python, including OOP.

with lots of exercises, easy to follow

The course is available on Udemy.

There is one more method that works in a similar way, empty. It should be used with caution. It returns an uninitialized array, so an array with arbitrary data of the given shape and type that must be initialized later on. Here’s an example:

>>> g = np.empty((5, 2))
>>> print(g)
[[6.23042070e-307 1.24610383e-306]
 [6.89804132e-307 8.45610231e-307]
 [9.34601642e-307 7.56587584e-307]
 [8.45559303e-307 1.06811422e-306]
 [1.05694828e-307 2.22522596e-306]]

Blender Jumpstart Course

Learn the basics of 3D modeling in Blender.

step-by-step, easy to follow, visually rich

The course is available on Udemy and on Skillshare.

Here’s the video version: