Python: replace values in a 2d array from another 1d array based on some conditions

Question

thanks in advance for your help! I would like to the do the following, but I am new to Python, kind of unsure what to do efficiently.

I have a 2d array, for example A=[[1,1],[2,3]].
Each value in the above 2d array corresponds to the index in another 1d array, for example: B=[0.1,0.2,0.8,0.9].
The end result should be like: C=[[0.2,0.2],[0.8,0.9]]. That means, C[i,j]=B[index=A[i,j]].

The above is a simple example. But in practice, A can be a huge array, so I would really appreciate if there is any way to do this efficiently. Thank you!

user16836078 · Accepted Answer · 2022-06-08 23:46:23Z

2

According to your post, you already almost got the answer. If you are really looking for a one line code, you can do this.

c = B[A]

c
Out[24]: 
array([[0.2, 0.2],
       [0.8, 0.9]])

The code above is for numpy array. On the other hand, if it is a list, list comprehension would be required.

answered Jun 8, 2022 at 23:46

user16836078

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

Comments

Paul Dlc · Accepted Answer · 2022-06-08 21:49:32Z

1

First try planning the sequence from index of first list and the relation with the result list.

A = [[1,1],[2,3]]
B=[0.1,0.2,0.8,0.9]

C = [[B[i] for i in j] for j in A]

print(C)

edited Jun 8, 2022 at 21:49

answered Jun 8, 2022 at 21:10

Paul Dlc

1331 silver badge6 bronze badges

3 Comments

Barmar Over a year ago

Isn't there a way to do this without a loop in numpy?

Vergil Over a year ago

Really thanks for your help! But I am wondering is there any way to vectorize such a loop? Thanks!

Paul Dlc Over a year ago

@QuVergil I edited my answer for a shorter. I hope that it's that you mean

BhusalC_Bipin · Accepted Answer · 2022-06-08 22:52:31Z

1

Based on your comments on answer by @PAUL ANDY DE LA CRUZ YANAC, I see that you are trying to use numpy and avoid for loop but as far as my knowledge, you need to use a for loop at least once.

import numpy as np

for x, y in np.ndindex(np.array(A).shape):
    A[x][y] = B[A[x][y]]

Note: This approach changes the original list A. But if you want to create a new list, look at the solution by @Paul Dlc.

edited Jun 8, 2022 at 22:52

answered Jun 8, 2022 at 21:51

BhusalC_Bipin

8215 silver badges12 bronze badges

Collectives™ on Stack Overflow

Python: replace values in a 2d array from another 1d array based on some conditions

3 Answers 3

Comments

3 Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

3 Answers 3

Comments

3 Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related