Question

I am using numpy.

I have one array Y and one matrix X. This is for a regression. They arrays has labels, e.g. 0,1,2,3,4,5. I need to create a new array that has label 0 removed for all rows and the corresponding row in X removed as well. What is the most efficient means to do this?

e.g.

for i in xrange(y.shape):
    if y==0:
       pop y pop X

Answer 1

Numpy arrays are not good at appending/removing rows. If you know which rows are to be deleted, just extract the other rows (you need) and create a new array.

I don’t understand your question very well, so please correct me if I am wrong:

x = x[y != 0]
y = y[y != 0]

Example:

import numpy as np
x = np.array([[11, 12, 13], [21, 22, 23], [31, 32, 33]])
y = np.array([1, 0, 3])
x = x[y != 0]
y = y[y != 0]

now:

x == array([[11, 12, 13],
            [31, 32, 33]])
y == array([1, 3])