Python: List shift

Question

Given a

list = [0,1,2,3,4,5,6,7,8,9,10]

How can I get a 2D List like:

2dList = [[0,1,2,3],[2,3,4,5],[4,5,6,7],[6,7,8,9],[8,9,10]]

(each 1st dimension of "2dList" has 4 values, there is a "shift" of 2 values from "list". When there isn't 4 values to fulfill the 1st dimension, it just store the last values from "list"]

?

thanks!

Why does it end with `[8, 9, 10]` and not `[6, 7, 8, 9]` or `[10]`? — NPE, May 26 '14 at 19:44
a/list is a terrible name for a list b/ 2dList is not a valid name for a variable — njzk2, May 26 '14 at 19:45
it's not a "formal" code... just a piece of "pseudo".. it's not right to use "list" as a name variable, I know that... — kairos, May 26 '14 at 19:50
@Veedrac it is not a duplicate because the chunks here are overlapping — Saullo G. P. Castro, May 26 '14 at 20:00
@SaulloCastro So is the linked question. I did notice there was another supposed duplicate that does not overlap, but my linked one does. — Veedrac, May 26 '14 at 20:01
And before someone says "but the shift value is **2**", that just means you take every other element from the iterator. It's the same question. — Veedrac, May 26 '14 at 20:02
@Veedrac, you are right...I checked the other one with three closing votes — Saullo G. P. Castro, May 26 '14 at 20:03

Jesse Aldridge · Accepted Answer · 2014-05-26T20:02:19.720

2

my_list = [0,1,2,3,4,5,6,7,8,9,10]

move_amount = 2
slice_size = 4
print [my_list[i:i + slice_size] for i in range(
  0, len(my_list) - move_amount, move_amount)]

edited May 26 '14 at 20:02

answered May 26 '14 at 19:45

Jesse Aldridge

7,991
9
48
75

1

this outputs an extra `[10]` at the end. it can be debated as to whether it should be there or not, as it can be considered a valid piece of the output. To remove it, use `range(0, len(l) - move_amount, move_amount)` – njzk2 May 26 '14 at 19:48
oops, fixed. thanks. – Jesse Aldridge May 26 '14 at 19:49

score 2 · Answer 2 · answered May 26 '14 at 19:50

2

IF you are willing to use numpy you can change the array strides in order to achieve what you want:

import numpy as np
from numpy.lib.stride_tricks import as_strided

a = np.array([0,1,2,3,4,5,6,7,8,9,10, 11]) # added 11 to avoid a memory mess
as_strided(a, shape=(5,4), strides=(a.strides[0]*2, a.strides[0]))
#array([[ 0,  1,  2,  3],
#       [ 2,  3,  4,  5],
#       [ 4,  5,  6,  7],
#       [ 6,  7,  8,  9],
#       [ 8,  9, 10,  11]])

and in memory this is actually what it is in array a, which can be useful for some purposes...

answered May 26 '14 at 19:50

Saullo G. P. Castro

56,802
26
179
234

ok, but let's imagine that I don't have a multiple of 4 elements, therefore I don't have "11" as last result... – kairos May 26 '14 at 19:51
1

should add `import numpy as np` to the top – Jesse Aldridge May 26 '14 at 19:53
@kairos, in that case this solution may be inadequate, or you can just add a dummy value at the end to ensure this requirement... – Saullo G. P. Castro May 26 '14 at 19:54

Python: List shift

2 Answers2