np.searchSorted over 2D array

[Heli]

Dear all,

I need to check whether two 2d numpy arrays have intersections and if so I will need to have the cell indices of the intersection.

By intersection, I exactly mean the intersection definition used in set theory.

I will give an example of what I need to do:

a=[[0,1,2],[3,4,5],[6,7,8]]
b=[[0,1,2],[3,4,5]]

I would like to check whether b is subset of a and then get the indices in a where b matches.

cellindices=[[True,True,True],[True,True,True],[False,False,False]]

What is the best way to do this in numpy?


Thanks in Advance for your help,

[Peter Otten]

Heli wrote:

[Please don't open a new thread for the same problem]

Providing an example is an improvement over your previous post, but to me
it's still not clear what you want.

>>> functools.reduce(lambda x, y: x | y, (a == i for i in b.flatten()))
array([[ True, True, True],
[ True, True, True],
[False, False, False]], dtype=bool)

produces the desired result for the example input, but how do you want to
handle repeating numbers as in

>>> a = numpy.array([[0,1,2],[3,4,5],[3, 2, 1]])
>>> functools.reduce(lambda x, y: x | y, (a == i for i in b.flatten()))
array([[ True, True, True],
[ True, True, True],
[ True, True, True]], dtype=bool)

?

Try to be clear about your requirement, describe it in english and provide a
bit of context. You might even write a solution that doesn't use numpy and
ask for help in translating it.

At the very least we need more/better examples.

[Heli]

Thanks Peter,

I will try to explain what I really need.

I have a 3D numpy array of 100*100*100 (1M elements). Then I have another numpy array of for example 10*2*10 (200 elements). I want to know if in the bigger dataset of 100*100*100, there is anywhere, where the second numpy array of 200 elements with shape 10*2*10 appears. If it does, then I want the indices of the bigger dataset where this happens.

I hope I´m making myself more clear :)
Thanks for your comments,

[Oscar Benjamin]

On 10 Dec 2015 14:46, "Heli" <hem...@gmail.com> wrote:
>
> Thanks Peter,
>
> I will try to explain what I really need.
>
> I have a 3D numpy array of 100*100*100 (1M elements). Then I have another
numpy array of for example 10*2*10 (200 elements). I want to know if in the
bigger dataset of 100*100*100, there is anywhere, where the second numpy
array of 200 elements with shape 10*2*10 appears. If it does, then I want
the indices of the bigger dataset where this happens.
>

So you want to find N in M. First find all occurrences of N[0][0][0] in
M[:90][:98][:90]. Then for each of those extract the same size subview from
M and check if (Ms == N).all().

--
Oscar