Can I filter an array using a boolean array obtained from a compound statement?

Question

I have a numpy array of some dimension and I would like to make a slice of this array using a boolean array. For instance, given my array called 'verts' and I want to make a slice of this array into another array called 'upper' based on some condition. I can do the following:

cond_upper = verts[:,2]<L/2
upper = verts[cond_upper]

How can I modify the above snippet so that a double condition is satisfied, something like this:

cond_upper = verts[:,2]<L/2 and verts[:,2]<40.0
upper = verts[cond_upper]

This does not give me the desired result because python wants me to compare them element-wise, however this is not what I want to do. How can I resolve this?

I add an example to make things clearer:

verts = np.array([[1,2,3],
                  [4,5,6],
                  [7,8,9],])
cond = verts[:,2]>3 and verts[:,2]<7
upper = verts[cond]

Expected result:

upper = [[1,2,3],
         [4,5,6],]

I hope this makes it somewhat clearer

Can you update your code with a sample and the expected result please? — Corralien
– Corralien, Commented Oct 17, 2021 at 21:23
Not sure which array package you are using, but try & instead of and. — mozway
– mozway, Commented Oct 17, 2021 at 21:23
@mozway I get the following error when I use that: TypeError: ufunc 'bitwise_and' not supported for the input types, and the inputs could not be safely coerced to any supported types according to the casting rule ''safe'' — eddy
– eddy, Commented Oct 17, 2021 at 21:31
@mozway It works fine, I googled a little bit and I found out I needed to put some brackets. Can you post it as an answer so I can accept it? — eddy
– eddy, Commented Oct 17, 2021 at 21:36

Corralien · Accepted Answer · 2021-10-17 21:42:21Z

2

Try:

>>> [row for row in verts if row[2] >= 3 and row[2] < 7]
[[1, 2, 3], [4, 5, 6]]

answered Oct 17, 2021 at 21:42

Corralien

121k8 gold badges44 silver badges69 bronze badges

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

Comments

mozway · Accepted Answer · 2021-10-17 21:52:49Z

1

You should use the bitwise operator & for element comparisons, not and.

import numpy as np
verts = np.array([[1,2,3],
                  [4,5,6],
                  [7,8,9],])
cond = (verts[:,2]>=3) & (verts[:,2]<7)
upper = verts[cond]

Output:

array([[4, 5, 6]])

edited Oct 17, 2021 at 21:52

answered Oct 17, 2021 at 21:42

mozway

267k13 gold badges56 silver badges106 bronze badges

6 Comments

Corralien Over a year ago

Check your code... If it's not a numpy array, it can't work!

mozway Over a year ago

OP mentioned the input was an array, I added the import for clarity ;)

eddy Over a year ago

Indeed I use a numpy array and the answer submitted by @mozway works fine. I should have added in the question that I am using a numpy array. I have properly edited the question now

Corralien Over a year ago

Missing parenthesis: cond = (verts[:,2]>3) & (verts[:,2]<7)

Corralien Over a year ago

And the filter don't give the expect output because 3 > 3 is not True. 3 >= 3 is True

|

Collectives™ on Stack Overflow

Can I filter an array using a boolean array obtained from a compound statement?

2 Answers 2

Comments

6 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

6 Comments

Your Answer

Sign up or log in

Post as a guest

Related