python list / numpy array indexing using tuple unpacking possible?

Question

I was wondering if tuple unpacking can be used in a "index from ... to" kind of style, so with inds = (a,b), M[*inds] would lead to M[a:b].

I often have tuples which contain the indices that I use to slice my data into some interesting subset, and would think that something like the proposed above would be convenient. Is there a way to do this?

thanks for input

You can try to patch/overload the getitem method of your list objects, such thath M[inds] is supported. — jofel
– jofel, Commented Feb 11, 2016 at 10:30

tobias_k · Accepted Answer · 2016-02-11 10:41:37Z

6

You can use tuple unpacking, but you have to unpack them into a slice, and then use the slice for indexing:

>>> A = list(range(10))
>>> inds = 3, 6
>>> A[slice(*inds)]
[3, 4, 5]

Also works with numpy:

>>> B = np.array([[1,2,3],[4,5,6],[7,8,9]])
>>> ind_x, ind_y = (0,2), (1,None)
>>> B[slice(*ind_x),slice(*ind_y)]
array([[2, 3],
       [5, 6]])

Remember that A[start:stop:step] is just syntactic sugar for A[slice(start,stop,step)].

edited Feb 11, 2016 at 10:41

answered Feb 11, 2016 at 10:35

tobias_k

83.1k12 gold badges130 silver badges186 bronze badges

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

1 Comment

grg rsr Over a year ago

thank you for reminding me. Still new to this stackoverflow system.

MSeifert · Accepted Answer · 2016-02-11 10:32:34Z

0

You cannot use tuples (at least not in the start-stop-step way). But you can always use slice.

some examples:

indx = slice(start, stop, step)
indy = slice(starty, stopy, stepy)
sliced = M[indx]
sliced2 = M[:,indy]
sliced3 = M[(indx, indy)]

answered Feb 11, 2016 at 10:32

MSeifert

154k41 gold badges356 silver badges377 bronze badges

Collectives™ on Stack Overflow

python list / numpy array indexing using tuple unpacking possible?

2 Answers 2

1 Comment

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

1 Comment

Comments

Your Answer

Sign up or log in

Post as a guest

Related