update elements using numpy array function

Question

I am trying to implement k-mean clustering algorithm for small project. I came upon this article which suggest that

K-Means is much faster if you write the update functions using operations on numpy arrays, instead of manually looping over the arrays and updating the values yourself.

I am exactly using iteration over each element of array to update it. For each element in dataset z, I am assigning the cluster array from nearest centroid via iteration through each element.

    for i in range(z):
        clstr[i] = closest_center(data[i], cen)

and my update function is

def closest_center(x, clist):
    dlist = [fabs(x - i) for i in clist]
    return clist[dlist.index(min(dlist))]

Since I am using grayscale image, I am using absolute value to calculate the Euclidean distance.

I noticed that opencv has this algorithm too. It takes less than 2s to execute the algorithm while mine takes more than 70s. May I know what the article is suggesting?

My images are imported as gray scale and is represented as 2d numpy array. I further converted into 1d array because it's easier to process 1d array.

Why do you want to implement this yourself? scipy already has a k-means clustering algorithm for you. — Akshat Mahajan
– Akshat Mahajan, Commented Apr 24, 2016 at 20:41
@AkshatMahajan As a small project in image processing. I already have access to it via OpenCV. Still I have to it without using inbuilt function. — S L
– S L, Commented Apr 24, 2016 at 20:43
you should share more of your code, in order to have a better idea. Also shouldn't you use the euclidean distance ? — Romain
– Romain, Commented Apr 24, 2016 at 20:45
@Romain for greyscale image, there is only one element. So, Euclidean distance is same as absolute value as in real number line. — S L
– S L, Commented Apr 24, 2016 at 20:50
@SantoshLinkha A fairly big part of it is probably implementation in C vs python. — Natecat
– Natecat, Commented Apr 24, 2016 at 21:00

Tonechas · Accepted Answer · 2016-04-24 21:44:49Z

1

The list comprehension is likely to slow down execution. I would suggest to vectorize the function closest_center. This is straightforward for 1-dimensional arrays:

import numpy as np

def closest_center(x, clist):
    return clist[np.argmin(np.abs(x - clist))]

answered Apr 24, 2016 at 21:44

Tonechas

13.8k16 gold badges52 silver badges85 bronze badges

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

Collectives™ on Stack Overflow

update elements using numpy array function

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related