How to change values of a numpy array between two positions

Question

I have a numpy array full of 0's such as this

   [0,0,0,0,0,0,0,0,0,0,0,0]

And a list of positions such as this

   [2-4,6-10]

So what I want to do is iterate through the list of positions and then change the 0's in the numpy array to 1's within the according positions so that I should have a numpy array such as.

  [0,1,1,1,0,1,1,1,1,1,0,0,0]

Hope this is clear enough, if not just let me know.

Thanks.

Your "list of positions" is effectively [-2, 6, 10] (because 2-4 is -2). Is that what you intended? — BrenBarn
– BrenBarn, Commented Apr 26, 2017 at 19:05
Sorry my bad, I meant 2-4 and then 6-10 so the positions would be the second to the forth and then the sixth to the tenth. Thanks — NJD
– NJD, Commented Apr 26, 2017 at 19:06
So how do you represent the 2-4 in python? (the way you represent it will change the answer) — Kasravnd
– Kasravnd, Commented Apr 26, 2017 at 19:07
That still doesn't make sense. If you type [2-4,6-10] into the Python interpreter and evaluate it you will get [-2, -4]. What data structure do you actually have? Do you have a list of strings? — BrenBarn
– BrenBarn, Commented Apr 26, 2017 at 19:07
Your output array seems a little inconsistent. Are you sure it's correct? 2-4 covers positions 2 through 4 and 6-10 only covers positions 7 through 10 — lordingtar
– lordingtar, Commented Apr 26, 2017 at 19:07

Divakar · Accepted Answer · 2017-04-26 20:10:45Z

5

Here's one approach by generating those indices as a concatenated array with np.r_ and then indexing and assigning 1s -

In [64]: a = np.array([0,0,0,0,0,0,0,0,0,0,0,0])

In [65]: pos = np.r_[1:4,5:10]

In [66]: a[pos] = 1

In [67]: a
Out[67]: array([0, 1, 1, 1, 0, 1, 1, 1, 1, 1, 0, 0])

edited Apr 26, 2017 at 20:10

answered Apr 26, 2017 at 19:07

Divakar

222k19 gold badges273 silver badges374 bronze badges

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

3 Comments

kmario23 Over a year ago

this is a trivial fix; OP needs till index 10. so, it should be np.r_[1:4, 5:10]. As a side note, we could also use range()

Divakar Over a year ago

@kmario23 OP has 12 elems in input, while 13 in output. So, we need some clarification on it from OP. Posted comment on the same. Changing the ranges should be trivial anyway, so I would leave it to OP to make the needed edits.

kmario23 Over a year ago

Yeah, I noticed the discrepancy. But from his index positions [2-4, 6-10] it is somehow clear that he needs 5 elements (6-10) regardless of the input/output size. Seems like he's following matlab based convention for the index

Community · Accepted Answer · 2017-05-23 12:02:34Z

1

You could use a list of pairs to hold the positions;

l = [(2, 4), (6, 10)]
nl = [0,0,0,0,0,0,0,0,0,0,0,0,0,0,0]

Secondly, use a loop;

for a in l:
  for c, n in enumerate(nl) 
    if (a[0] >= c || a[1] <= c):
       nl[c] = 1

This is by far not the fastest way to do this but it is simple and readable.

As suggested by this user you could use this instead, which is a lot better in my opinion;

nl[a[0]:[a[1]]=[1]*(a[1]-a[0])

edited May 23, 2017 at 12:02

CommunityBot

11 silver badge

answered Apr 26, 2017 at 19:13

user7391763

3 Comments

hpaulj Over a year ago

This slice assignment might be faster: nl[a[0]:[a[1]]=[1]*(a[1]-a[0])

NJD Over a year ago

Sorry - I fixed it now

user7391763 Over a year ago

Glad I could help :)

Collectives™ on Stack Overflow

How to change values of a numpy array between two positions

2 Answers 2

3 Comments

3 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

3 Comments

3 Comments

Your Answer

Sign up or log in

Post as a guest

Related