CODEWITHSUNDEEP
pythonmatlabpython-2.7numpy
Ashish
Ashish

Reputation: 739

Replacement of matlab find function in python

I am trying to find suitable python function to replace matlab find in my script and with some google search I see np.where() solves the purpose most of the time. But in case of double condition, I have different outputs. Can someone tell me whats wrong with this approach and how to go ahead? The example code and difference is below.

In case of matlab: 

b = [1, 2, 3; 1, 2, 3; 1, 2, 3]
[I, J] = find(( b > 1) & (b <= 3))

Gives output

I =                 J = 
     1                 2
     2                 2
     3                 2
     1                 3
     2                 3
     3                 3

In case of python:

b= array([[1, 2, 3],
       [1, 2, 3],
       [1, 2, 3]])

>>> np.where((b > 1) & (b <= 3))
(array([0, 0, 1, 1, 2, 2]), array([1, 2, 1, 2, 1, 2]))

Upvotes: 4

Views: 1765

Answers (1)

Acorbe
Acorbe

Reputation: 8391

Both methods do provide the same answer, although order and indexing conditions are different.

Python indexing of arrays starts from 0, like in C, whilst matlab's one starts from 1.

Moreover, the two outputs (by matlab and numpy) do correspond one another modulo a permutation of the terms. Likely this is due to different indexing implementations.

You can see that matlab goes through your matrix by columns, while numpy by rows.

Upvotes: 4

Related Questions