How to do numpy combined slicing and array indexing with unknown array dimension

Question

I have a list of index tuples for all dimensions of an array bar the first two.

For each index tuple, I would like to return a slice of the array at this tuple position.

I don't know in advance the number of dimensions in the array, or equivalently the number of elements in the index tuple.

For example, if there were two elements in the index tuple (pos) I would write something like:

arr[:, :, pos[0], pos[1]]

to return a slice of the array at this position. I would like to be able to do this without knowing in advance the length of pos. Naively, I thought arr[:, :, *pos] would work, but of course it doesn't.

Thanks for the help.

Are the elements of index integers or a lists?

anishtain4
– anishtain4

2018-06-14 15:20:32 +00:00
Commented Jun 14, 2018 at 15:20 — anishtain4
– anishtain4, Commented Jun 14, 2018 at 15:20

BenBoulderite · Accepted Answer · 2018-06-14 15:24:50Z

1

You can concatenate the Ellipsis with your tuple p, to obtain the tuple Ep that can be used to slice the array:

Ep = (Ellipsis,)+p
sliced_arr = arr[Ep]

Edit: kudos to the previous answer (which has been deleted) for mentioning the Ellipsis, which is really half of my answer here.

edited Jun 14, 2018 at 15:24

answered Jun 14, 2018 at 15:22

BenBoulderite

3361 silver badge9 bronze badges

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

2 Comments

anishtain4 Over a year ago

This is much smarter than what I was trying to do

hpaulj Over a year ago

* expansion does work inside a tuple, e.g. arr[(Ellipsis, *p)].

Collectives™ on Stack Overflow

How to do numpy combined slicing and array indexing with unknown array dimension

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