Let's say I have a matrix:
>> a = np.arange(25).reshape(5, 5)`
>> a
[[ 0 1 2 3 4]
[ 5 6 7 8 9]
[10 11 12 13 14]
[15 16 17 18 19]
[20 21 22 23 24]]
and two vectors of indices that define a span of matrix elements that I want to extract:
>> indices1 = np.array([0, 1, 1, 0, 0])
>> indices2 = np.array([2, 3, 3, 2, 2])
As you can see, difference between each corresponding index is equal to 2.
I would like to do sth like this extract a part of the matrix:
>> submatrix = a[indices1:indices2, :]
so that the result would be 2x5 matrix:
>> submatrix
[[ 0 6 7 3 4],
[ 5 11 12 8 9]]
For all I know, numpy allows to provide indices as a boundaries, but does not allow to provide arrays, only integers, e.g. a[0:2]
.
Note what I want to subtract is not a submatrix:
Do you know of some other way of indexing a numpy matrix so that it is possible to provide arrays defining spans? For now I managed to do it only with for loops.
indices1
andindices2
always a constant? Like it is2
here? – Divakarindexing
. – dankal444