Python array, conditionally change elements on classification

Question

I have an array and want to change elements based on a classification using another array.

That is, I import one array, and then if the value of cell[i,j] is within a certain limit (say between 1 and 8) then multiply secondArray[i,j] by 0.3 and put result into an ouput array at place [i,j]

I have some code that does this (and probably explains what I mean a bit clearer) but it takes a 'very' long time (as my arrays are about 1000*1000 elements) and wondered if there is a more efficient solution.

Currently I have:

..
import numpy as np

def Factor_AM(cell):            
    if cell < 1     return 0.2;
    elif cell < 8:      return 0.3;
    else:       return 0.5;



mat = np.array(..some code to get an array from an external source..) //size 1000*1000
mat_out_1 = np.zeros_like(mat)
mat_out_1 = np.zeros_like(mat)

mat_ClassifyMe = np.array(..some code to import another 1000*1000 array with values in)

for index, x in np.ndenumerate(mat):
    mat_out_1[index] = Factor_AM(x) * mat_ClassifyMe[index]
    mat_out_2[index] = (1-Factor_AM(x)) * mat_ClassifyMe[index]

..some code to output mat_out_1 and mat_out_2

I saw some documentation on the np.where and np.argwhere functions that look hopeful, but given I have more than 1 test (in the above code I have 3 but in reality I have 12) I can't think of a way without making very ugly nested statements.

Is there another way to do this, or is this as efficient as Python gets?

Jan Kuiken · Accepted Answer · 2013-05-08 16:58:01Z

1

You can use Boolean or “mask” index arrays for this purpose, for example:

import numpy as np

mat = np.arange(16.0).reshape((4,4))

mat_out = np.zeros_like(mat)

mat_out[mat < 6] = 0.2 * mat[mat < 6]  # using a conditional for indexing
# or use a variable for the boolean 'masking' index array
mask1 = np.logical_and(mat >= 6, mat < 10)
mat_out[mask1] = 0.3 * mat[mask1]
mask2 = mat >= 10
mat_out[mask2] = 0.5 * mat[mask2]

print mat
print mat < 6
print mask1
print mat_out

edited May 8, 2013 at 16:58

answered May 8, 2013 at 9:59

Jan Kuiken

2,02018 silver badges20 bronze badges

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

2 Comments

CastleH Over a year ago

thanks, that looks like what I need. I'll run the code through with some timers to see what the difference is in a bit.

CastleH Over a year ago

Time results: inner loop (i.e. minus overheads for accessing and outputing arrays to other programmes, just looking at the filtering of the matrices) ~25s to ~0.5s Given I'm doing this a few hundred times, thats a very meaningful speed increase. Cheers.

AGN Gazer · Accepted Answer · 2017-06-28 22:37:14Z

0

mat = np.array([0.1, 0.2, 5, 4, 9, 0.05, 2, 11]) # an example of input #1
mat_ClassifyMe = np.ones_like(mat) # an example of input #2
mask1 = mat < 1
mask2 = mat >= 8
f = np.empty_like(mat)
f[mask1] = 0.2
f[mask2] = 0.5
f[np.logical_not(mask1 | mask2)] = 0.3
mat_out_1 = f * mat_ClassifyMe
mat_out_2 = (1 - f) * mat_ClassifyMe

answered Jun 28, 2017 at 22:37

AGN Gazer

8,4272 gold badges31 silver badges49 bronze badges

Collectives™ on Stack Overflow

Python array, conditionally change elements on classification

2 Answers 2

2 Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

2 Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related