Concatenating with empty numpy array

Question

I am running into errors when concatenating arrays in Python:

x = np.array([])
while condition:
    % some processing 
    x = np.concatenate([x + new_x])

The error I get is:

----> 1 x = np.concatenate([x + new_x])

ValueError: operands could not be broadcast together with shapes (0) (6)

On a side note, is this an efficient way to grow a numpy array in Python?

Given that you're copying x on every iteration, I am not sure I would necessarily call this "efficient". — NPE
– NPE, Commented Dec 5, 2012 at 17:55

Xymostech · Accepted Answer · 2012-12-05 17:52:21Z

3

Looks like you want to call

x = np.concatenate((x, new_x))

according to the docs.

answered Dec 5, 2012 at 17:52

Xymostech

9,8803 gold badges36 silver badges45 bronze badges

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

3 Comments

heltonbiker Over a year ago

Also, for greater control, there are the functions hstack, vstack and dstack, that are worth taking a look!

gerrit Over a year ago

@heltonbiker How do those have greater control than concatenate? I thought concatenate can do an arbitrary axis, whereas hstack, vstack, and dstack concatenate along a particular axis.

heltonbiker Over a year ago

By "greater control" I meant that, say, with hstack you always will concatenate along 1-axis. So, using hstack you won't "accidentally" concatenate along the wrong axis, so you would in this sense be more "in control" of the situation. But in the end it's a matter of perspective, of course.

Community · Accepted Answer · 2017-05-23 12:21:19Z

0

Alternatively:

x = np.append(x,new_x)

Regarding your side note, take a look here: How to extend an array in-place in Numpy?

edited May 23, 2017 at 12:21

CommunityBot

11 silver badge

answered Dec 5, 2012 at 18:56

daedalus12

4144 silver badges13 bronze badges

Collectives™ on Stack Overflow

Concatenating with empty numpy array

2 Answers 2

3 Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

3 Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related