Python numpy 3D array to 2D array (removing dimensions)

Question

I have a numpy array Z such that:

Z.shape

#Out[1]:
    (138, 112, 123)

How do I transform Z into a new array NewZ, such that:

NewZ.shape

#Out[2]:
    (138, 112)

?

What operation do you want performed on the dimension you want removed? NewZ = np.max(Z, axis=2) would yield a NewZ.shape of (138, 112), for example. — Michael B. Currie
– Michael B. Currie, Commented Aug 25, 2015 at 18:59
An array of shape (138, 112, 123) has 1901088 elements. An array of shape (138, 112) has only 15456 elements. So which 15456 elements of the original array do you want to end up in the new array? — Mark Dickinson
– Mark Dickinson, Commented Aug 25, 2015 at 19:26
@DolphinGenomePyramids: Can you edit the question to clarify what sort of reduction operation you're looking for? — Mark Dickinson
– Mark Dickinson, Commented Aug 26, 2015 at 18:30

Michael B. Currie · Accepted Answer · 2015-08-25 19:43:33Z

2

Removing a dimension means removing information, so you'll have to decide on a rule for projecting the original data down into a lower number of dimensions.

Suppose we have

import numpy as np
Z = np.random.random((138, 112, 123))

Here are two examples, both yielding a NewZ.shape of (138, 112):

NewZ = np.max(Z, axis=2), which takes the largest element of the last axis.
NewZ = Z[:,:,0], which takes the first element of the last axis.

answered Aug 25, 2015 at 19:43

Michael B. Currie

14.8k11 gold badges48 silver badges61 bronze badges

Sign up to request clarification or add additional context in comments.

Collectives™ on Stack Overflow

Python numpy 3D array to 2D array (removing dimensions)

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related