Deleting slices from numpy array given a condition

Question

I have a numpy array t_i of shape (6, 2000, 2). I need to delete column slices from the array if all the elements in that slice below that row is -1. Let's say I am in ith row of t_i. Now for those columns below the ith row where every element in the column is [-1,-1] I need to delete the complete column slice from t_i.

I tried this:

t_iF = t_i[:, t_i[i+1:] != -1]

But I am getting this error:

IndexError: too many indices for array: array is 3-dimensional, but 4 were indexed

Can you all please point me to the right direction? Thanks!

Updated: Let t_i be:

t1=np.arange(32).reshape(4,4,2)
t1[1:,[1,3]] = -1

In this case the whole 1 and 3 column slices need to be deleted.

Your description is unclear. Please add an example of source data and the expected result. Another problem is that your code looks rather like pseudo-code, e.g. value of i is unknown. — Valdi_Bo
– Valdi_Bo, Commented Apr 9, 2021 at 17:36
@adr, yes t_i[i+1:] != -1 returns a 3D tensor while I require only the column indices which satisfy the inequality. Can you suggest how I can do that? Thanks! — gis_grad_student
– gis_grad_student, Commented Apr 9, 2021 at 17:46
I am trying to add one. Actually I need to do this update over a loop(i.e. perform this step for a range of values of i). So i varies over a range. Thanks! — gis_grad_student
– gis_grad_student, Commented Apr 9, 2021 at 17:47

Stefan B · Accepted Answer · 2021-04-10 07:39:11Z

1

np.any can be very helpful for tasks like these:

import numpy as np

t1 = np.arange(32).reshape(4, 4, 2)
t1[1:, [1, 3]] = -1

t_filtered = t1[:, [np.any(t1[i, i:] != -1) for i in range(t1.shape[0])], :]
print(t_filtered.shape) # (4, 3, 2)

edited Apr 10, 2021 at 7:39

answered Apr 9, 2021 at 20:51

Stefan B

1,6977 silver badges15 bronze badges

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

2 Comments

gis_grad_student Over a year ago

Hi @Stefan Thanks for the help. Actually its t1[1:,[1,3]] = -1 so we don't have the whole column slice as -1 instead we have all elements below a particular row in that slice as -1.

Stefan B Over a year ago

The problem description is still somewhat unclear. I have updated the answer now though, hoping this is what you need.

Collectives™ on Stack Overflow

Deleting slices from numpy array given a condition

1 Answer 1

2 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

2 Comments

Your Answer

Sign up or log in

Post as a guest

Related