Looking to replace min and max of ndarray in Python

Question

I have the following ndarray :

c_dist = [[0.         5.83095189]
 [2.23606798 3.60555128]
 [5.83095189 0.        ]
 [5.83095189 2.82842712]
 [4.12310563 2.23606798]]

and I would like for each sub-array to replace the min with 1 and the max with 0, in order to obtain the following :

[[1. 0.]
 [1. 0.]
 [0. 1.]
 [0. 1.]
 [0. 1.]]

I used the following :

for i in range(len(c_dist)):
    max_of_row = c_dist[i].max()
    for elements_of_row in range(len(c_dist[i])):
        if c_dist[i][elements_of_row] == max_of_row:
            c_dist[i][elements_of_row] = 1
        else:
            c_dist[i][elements_of_row] = 0

but it is obviously not very elegant. Is there an python way of doing the comparison array by array please ?

is the cdist.shape = (n, 2) always True?

Daniel F
– Daniel F

2020-11-18 11:32:47 +00:00
Commented Nov 18, 2020 at 11:32 — Daniel F
– Daniel F, Commented Nov 18, 2020 at 11:32

Mehrdad Pedramfar · Accepted Answer · 2020-11-18 10:59:39Z

3

Try this in one line:

c_dist = [[0. ,5.83095189], 
          [2.23606798 ,3.60555128], 
          [5.83095189 ,0.        ], 
          [5.83095189 ,2.82842712], 
          [4.12310563 ,2.23606798]]  

new_list = [[int(i<=j), int(i>j)] for i,j in c_dist]

The result will be:

In [6]: new_list                                                                                                                                                                                                                                                                          
Out[6]: [[1, 0], [1, 0], [0, 1], [0, 1], [0, 1]]

answered Nov 18, 2020 at 10:59

Mehrdad Pedramfar

11.1k4 gold badges43 silver badges61 bronze badges

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

3 Comments

Mehrdad Pedramfar Over a year ago

Your welcome. please check the gray button and turn it to green if it was helpful.

Rene Chan Over a year ago

Hi @Mehrdad, actually my example should state that there are cases when i an have more than 2 column. ei: the number of column is not fixed.

Mehrdad Pedramfar Over a year ago

@FrankyDoul In those cases... what the logic. can you add an example of what you are saing.

Daniel F · Accepted Answer · 2020-11-18 11:38:31Z

1

If you have more than 2 columns:

out = c_dist.copy()
np.put_along_axis(out, c_dist.argmax(0), 1)
np.put_along_axis(out, c_dist.argmin(0), 0)

Or if there are multiple max and min values per row:

out = np.where(c_dist == c_dist.max(0, keepdims = True), 1, c_dist)
out = np.where(c_dist == c_dist.min(0, keepdims = True), 0, out)

answered Nov 18, 2020 at 11:38

Daniel F

14.5k2 gold badges34 silver badges59 bronze badges

Collectives™ on Stack Overflow

Looking to replace min and max of ndarray in Python

2 Answers 2

3 Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

3 Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related