I have a list of masks and I want to obtain the resulting mask by multiplying all of them. My Donkey Kong approach is the following:
a = [[1, 1], [1, 0], [1, 0]]
b = a[0]
for i in range(1, len(a)):
b = b * np.array(a[i])
which I think it works as returns [1,0] as value of b.
Is there a nicer way of doing this?
EDIT: I am looking for common ranges in the mask. To find all the non zero ranges I do the following: label = 0
for i in range(1, len(labels)):
label = label + np.array(labels[i])
label = [1 if x > 0 else 0 for x in label]
I see that you are already using numpy so it can be used like other answers suggested. But still, a nice built-in solution only using reduce can be:
from functools import reduce
a = [[1, 1], [1, 0], [1, 0]]
def element_wise_multiply(list1, list2):
return [x*y for x,y in zip(list1, list2)]
b = reduce(element_wise_multiply, a)
# Or as a lambda:
b = reduce(lambda list1, list2: [x*y for x,y in zip(list1, list2)], a)
This takes every two sub-lists and reduces them to one by multiplying all index-matching elements by using zip. It also gives:
[1, 0]
A built-in way with all that short circuits (no other answer short circuits now):
>>> [all(column) for column in zip(*a)]
[True, False]
where zip(*a) kindof transposes to produce "columns" as sublists here and we check if all entries are truthful. all has a short circuting behaviour.
You can int(all(sub)) to cast results to integer after all'in the sublists.