Trying to append a column in numpy array, getting ValueError

Question

I am trying to append column for the following arrays

train = np.append(train_data, train_labels, axis=1)

(60000, 784)
(60000,)

And I get the error

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

I cannot understand what the issue is... I need output of

(60000, 785)

When given an axis, np.append` is just np.concatenate((arr, values), axis=axis). When using concatenate the number of dimensions of the inputs has to match (in this case all must be 2d). — hpaulj
– hpaulj, Commented Jun 1, 2019 at 23:47

Hennadii Madan · Accepted Answer · 2019-06-01 23:38:49Z

1

I think you should use axis=0.

>>> np.append((1,2),(2,),axis=0)
array([1, 2, 2])

If those tuples that you have posted are array shapes, then you can use

train = np.append(train_data, train_labels[:,None], axis=1)

The arrays must have the same number of dimensions (2 in this case). Using None in indexing for non-existing dimension adds a singleton dimension to a an array.

>>> train_labels[:,None].shape
(60000, 1)

edited Jun 1, 2019 at 23:38

answered Jun 1, 2019 at 23:25

Hennadii Madan

1,6231 gold badge16 silver badges30 bronze badges

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

2 Comments

YohanRoth Over a year ago

with axis 0 I get the same error and I want to add column, axis 0 is for rows

Hennadii Madan Over a year ago

@YohanRoth You are welcome. I added a bit of explanation (not very accurate, but you can check numpy docs on broadcasting for an accurate (but confusing) explanation.

Rishav Raj · Accepted Answer · 2019-06-01 23:45:02Z

1

You should run

train = np.append(train_data, train_labels.reshape(-1, 1), axis=1)

That will fix your problem. Cheers

answered Jun 1, 2019 at 23:45

Rishav Raj

111 bronze badge

Collectives™ on Stack Overflow

Trying to append a column in numpy array, getting ValueError

2 Answers 2

2 Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

2 Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related