How can I create a NumPy array B which is a sub-array of a NumPy array A, by specifying which rows and columns (indicated by x and y respectively) are to be included?
For example:
A = numpy.array([[1, 2, 3, 4, 5], [6, 7, 8, 9, 10], [11, 12, 13, 14, 15]])
x = [0, 2]
y = [1, 3, 4]
B = # Do something .....
Should give the output:
>>> B
array([[2, 4, 5], [12, 14, 15]])
The best way to do this is to use the ix_ function: see the answer by MSeifert for details.
Alternatively, you could use chain the indexing operations using x and y:
>>> A[x][:,y]
array([[ 2, 4, 5],
[12, 14, 15]])
First x is used to select the rows of A. Next, [:,y] picks out the columns of the subarray specified by the elements of y.
The chaining is symmetric in this case: you can also choose the columns first with A[:,y][x] if you wish.
You can use np.ix_ which allows to broadcast the integer index arrays:
>>> A = np.array([[1, 2, 3, 4, 5], [6, 7, 8, 9, 10], [11, 12, 13, 14, 15]])
>>> x = [0, 2]
>>> y = [1, 3, 4]
>>> A[np.ix_(x, y)]
array([[ 2, 4, 5],
[12, 14, 15]])
From the documentation the ix_ function was designed so
[...] one can quickly construct index arrays that will index the cross product.
a[np.ix_([1,3],[2,5])]returns the array[[a[1,2] a[1,5]], [a[3,2] a[3,5]]].
Here's a super verbose way to get what you want:
import numpy as np
a = np.array([[1, 2, 3, 4, 5], [6, 7, 8, 9, 10], [11, 12, 13, 14, 15]])
x = [0, 2]
y = [1,3,4]
a2 = a.tolist()
a3 = [[l for k,l in enumerate(j) if k in y] for i,j in enumerate(a2) if i in x]
b = np.array(a3)
But please follow @ajcr answer:
import numpy as np
a = np.array([[1, 2, 3, 4, 5], [6, 7, 8, 9, 10], [11, 12, 13, 14, 15]])
x = [0, 2]
y = [1,3,4]
a[x][:,y]
