Question

I have a 2D array in the following form:

[[X1, X2, ..., XN]
[Y1, Y2, ..., YN]]

For each Xi greater than lower_limit_X and less than upper_limit_X, I would like to get the number of Yi‘s that are greater than lower_limit_Y and less than upper_limit_Y.

I hope there is an efficient way of doing this in Numpy apart from indexing one by one.

EDIT:
So I have a 2xN array. The first row has ordered values of N X’s and second row has ordered values of N Y’s. What I would like to get is:

1. get a the lowest_index and highest_index index of X, that have a value that is greater than lower_limit_X and less than upper_limit_X

2. then slice the Y array (just one array) in the index range [lowest_index, highest_index]

3. count the number of elements in my slice, having Yi's that are greater thanlower_limit_Yand less thanupper_limit_Y`.

Answer 1

Here are two ways you could do this, the more strait forward way is probably,

mask = ((lower_x_limit < array[0]) & (array[0] < upper_x_limit) &
        (lower_y_limit < array[1]) & (array[1] < upper_y_limit))
count = sum(mask)

If your array is very large and both x and y are sorted you could use searchsorted instead,

start = array[0].searchsorted(lower_x_limit, 'right')
end = array[0].searchsorted(upper_x_limit, 'left')
temp = array[1, start:end]
start = temp.searchsorted(lower_y_limit, 'right')
end = temp.searchsorted(upper_y_limit, 'left')
count = end - start