How can I use numpy create list out of another list with modified elements

Question

I am not new at all to Python programming, but I am completely new to the Numpy module. I need to use this module for it's very fast and efficient.

Say I have an array called noise which is defined as follows:

noise = [[uniform(0, 1) for i in range(size)] for j in range(size)]

In numpy terms, it is defined, I believe, as so:

noise = np.uniform(0, 1, (size, size))

Now say I want to generate a new array which takes the noise array and replaces every element noise[i][j] of its elements by the function function(i, j) Using python's built-in list comprehension, I would simply say:

modified_noise = [[function(i, j) for i in range(size)] for j in range(size)]

My question, is: how can I do that using the numpy module.

What's wrong with using the list comprehension? Looks like function only works with scalar inputs, right? There's no way of getting around evaluating it once for each of the size*size (i,j) pairs, is there? — hpaulj
– hpaulj, Commented Mar 4, 2018 at 3:10
I don't see how modified_noise is related to noise. function only takes coordinates as arguments, not noise. — hpaulj
– hpaulj, Commented Mar 4, 2018 at 3:56
By the way, there's no np.uniform.. noise = np.random.random_sample((size, size)) is the right expression. You also seem to be confused about the concept of 'replacing every element'. — hpaulj
– hpaulj, Commented Mar 4, 2018 at 5:46
Oh yes, there is a numpy.uniform. I used it and it actually works just as expected. — chihabeddineAourinmouche
– chihabeddineAourinmouche, Commented Mar 4, 2018 at 12:19
Also, modified_noise is to be constructed from noise. The two are related but only inside the definition of the function function which in my case is a turbulance function (I called it function to give you a more abstract version of the code) — chihabeddineAourinmouche
– chihabeddineAourinmouche, Commented Mar 4, 2018 at 12:28

jpp · Accepted Answer · 2018-03-04 10:36:26Z

2

You can use np.fromfunction for this:

modified_noise = np.fromfunction(lambda i, j: function(i, j), (size, size), dtype=float)

This constructs an array by executing a function over each coordinate.

Related: How can I use a range inside numpy.fromfunction?

edited Mar 4, 2018 at 10:36

answered Mar 4, 2018 at 0:25

jpp

166k37 gold badges301 silver badges363 bronze badges

Sign up to request clarification or add additional context in comments.

1 Comment

hpaulj Over a year ago

On use of fromfunction: How can I use a range inside numpy.fromfunction?

Stephen Rauch · Accepted Answer · 2018-03-04 00:25:16Z

0

You can build a numpy array directly from a list of lists like:

Code:

np.array(modified_noise)

Test Code:

data = [[i * j for j in range(5)] for i in range(5)]
print(data)

import numpy as np
print(np.array(data))

Results:

[[0, 0, 0, 0, 0], [0, 1, 2, 3, 4], [0, 2, 4, 6, 8], [0, 3, 6, 9, 12], [0, 4, 8, 12, 16]]

[[ 0  0  0  0  0]
 [ 0  1  2  3  4]
 [ 0  2  4  6  8]
 [ 0  3  6  9 12]
 [ 0  4  8 12 16]]

answered Mar 4, 2018 at 0:25

Stephen Rauch♦

50.1k32 gold badges118 silver badges143 bronze badges

Collectives™ on Stack Overflow

How can I use numpy create list out of another list with modified elements

2 Answers 2

1 Comment

Code:

Test Code:

Results:

Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

1 Comment

Code:

Test Code:

Results:

Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related