Convert 1D numpy.array([]) to numpy.array([[]])

Question

Im calculating results for each rows in 2D-array.

import numpy as np
xyz = np.array([[1, 2, 100],[2, 5, 100]])
resultcolumn1 = (xyz[:,0]+xyz[:,1])*xyz[:,2]
>>array([300, 700])
>>type 'numpy.ndarray'>

But when I'm trying to add this resultvector to rest of the 2D-array as a column

print np.concatenate((xyz, resultcolumn1.T), axis=1)

I get this error:

ValueError: all the input arrays must have same number of dimensions.

Declaring the same vector with double brackets. Works:

resultcolumn2=np.array([[300, 700]])
>>>array([[300, 700]])
>>><type 'numpy.ndarray'>
np.concatenate((xyz, resultcolumn2.T), axis=1)
>>>[[  1   2 100 300]
    [  2   5 100 700]]

How do I change resultcolumn1 to match resultcolumn2? Also could this process be done in better fashion?

Possible duplicate of Python reshape list to ndim array

languitar
– languitar

2017-03-30 12:37:28 +00:00
Commented Mar 30, 2017 at 12:37 — languitar
– languitar, Commented Mar 30, 2017 at 12:37

Moses Koledoye · Accepted Answer · 2017-03-30 16:02:24Z

1

Simply add an extra dimension:

>>> np.concatenate((xyz, resultcolumn1.T[..., None]), axis=1)
array([[  1,   2, 100, 300],
       [  2,   5, 100, 700]])

Also, note that for a 1d array, .T (i.e. transpose) does nothing so you can simply drop it:

>>> np.concatenate((xyz, resultcolumn1[..., None]), axis=1)
array([[  1,   2, 100, 300],
       [  2,   5, 100, 700]])

edited Mar 30, 2017 at 16:02

answered Mar 30, 2017 at 12:39

Moses Koledoye

78.8k8 gold badges139 silver badges141 bronze badges

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

2 Comments

hpaulj Over a year ago

If resultcolumn1 is 1d, the .T does nothing.

Moses Koledoye Over a year ago

@hpaulj Ah yes, adding that to the answer. Thanks.

Collectives™ on Stack Overflow

Convert 1D numpy.array([]) to numpy.array([[]])

1 Answer 1

2 Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

2 Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related