Does element exist in numpy array?

Question

numpy NOOB.

How best to see if an element exists in a numpy array?

Example:

import numpy as np
a = np.array([[1, 2], [2, 3], [3, 4]])
[2, 4] in a
# This evaluates to True because there's 
# a 2 (somewhere) and a 4 (somewhere)
# but I want to match [2, 4] ONLY. So...
[2, 4] in a  # Would like this to be False
[2, 3] in a  # Would like this to be True
[3, 2] in a  # This too should be false (wrong order [3, 2] != [2, 3])

I've looked at np.where() and that doesn't seem to be what I'm looking for. I get a similar result to the above using np.isin([2, 4], a).

Don't need the index (though if it comes along for the ride that OK), just a boolean will suffice.

Agreed. I think it is a duplicate. – PartialOrder Oct 01 '19 at 14:49 — PartialOrder, Oct 01 '19 at 14:49

score 1 · Answer 1 · answered Oct 01 '19 at 14:22

1

You're searching for 2 and 4 in a, try:

[[2, 4]] in a

answered Oct 01 '19 at 14:22

Rafa Guillermo

14,474
3
18
54

...but that evaulates to `True`. – Matt Hall Oct 01 '19 at 14:26
I've added clarification. See edited question. If [2, 3] in array, do not want to match with [3, 2] (order important). Only want to match if elements and order agree. THANKS – PartialOrder Oct 01 '19 at 14:34

score 0 · Accepted Answer · answered Jan 08 '22 at 20:45

0

This was flagged as duplicate of testing whether a Numpy array contains a given row wherein Tom10 provides the answer I sought:

np.equal([2, 3], a).all(axis=1).any()

answered Jan 08 '22 at 20:45

PartialOrder

2,870
3
36
44

1

When someone suggests a duplicate that you agree with, you should vote to close/flag as duplicate, or just accept the duplicate suggestion. You definitely shouldn't answer the question. If you have a solution that's not covered on the target, you can add an answer there. – cigien Jan 09 '22 at 01:35

Does element exist in numpy array?

2 Answers2