Delete subarray in numpy array if threshold is not met

Question

I have the following np array structure:

[[1, 2, 3 ,4]
[5, 7, 8 ,6]
     .
     .
[7, 5, 1 ,0]]

What is want to do is to remove a subarray if thresholds are not met. for example in [5, 7, 8 ,6], i want to delete this array if position 0 is not between 2 and 4. I want to do this action for the whole numpy array and intend on having a threshold on all positions in the sub array.

My thought process is something that is shown below:

for arr in data:
    if arr[0] < 2 or arr[0] > 4:
        np.delete(data, arr)

However, printing data.shape before and after show no difference. Can someone help me?

Thanks!

To understand why the np.delete approach doesn't work, first read the documentation (you will notice that it creates a new array, which you subsequently ignore). However, fixing that would still leave you with a problem because you would be trying to modify a sequence while iterating over it. — Karl Knechtel
– Karl Knechtel, Commented Apr 16, 2021 at 10:25

Karl Knechtel · Accepted Answer · 2021-04-16 10:23:19Z

2

Creating example data for testing:

>>> import numpy as np
>>> data = np.array([
... [1,2,3,4],
... [5,7,8,9],
... [7,5,1,0]
... ])

You can slice the array to get the first column:

>>> data[:, 0]
array([1, 5, 7])

Figure out which of these first-column values is in range by broadcasting the comparison operators across them (being careful that we can't chain these operators, and must combine them using a bitwise rather than logical AND, because of syntax limitations):

>>> first = data[:, 0]
>>> (4 <= first) & (first <= 6)
array([False,  True, False])

Finally, we can use that to mask the original array:

>>> data[(4 <= first) & (first <= 6)]
array([[5, 7, 8, 9]])

answered Apr 16, 2021 at 10:23

Karl Knechtel

61.4k14 gold badges131 silver badges193 bronze badges

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

Collectives™ on Stack Overflow

Delete subarray in numpy array if threshold is not met

1 Answer 1

Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related