Replicate multidimensional NumPy array on an axis up-to specific value

Question

I have a two-dimensional NumPy array with shape

(2, 2)

Example array

array([[1, 2], [3, 4]])

I am trying to have it copy on just the first axis until it reaches the shape:

(5, 2)

Example result array

array([[1, 2], [1, 2], [1, 2], [3, 4], [3, 4]])

np.repeat does the job but it has to be a multiple as it repeats everything

np.repeat(arr, 3, axis=0)

array([[1, 2], [1, 2], [1, 2], [3, 4], [3, 4], [3, 4]])

giving a 6 by 2 array and not a 5 by 2 array

filippo · Accepted Answer · 2021-04-13 11:57:06Z

1

np.repeat(arr, [3, 2], axis=0)

answered Apr 13, 2021 at 11:57

filippo

5,3044 gold badges23 silver badges48 bronze badges

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

3 Comments

Azaria Gebremichael Over a year ago

Thank you, but I am looking to repeat the entire array 3 times not each array a specific amount. I have a very large array with many more axis so I am looking for a method to repeat the entire array like the normal np.repeat(arr, no, axis) but crop the remaining part for example if I want to repeat a (1475,2) array to reach (3500, 2)

Azaria Gebremichael Over a year ago

sorry for the confusing question, a normal crop can work after repeat now that I think about it but if there is a method that would be helpful

filippo Over a year ago

@AzariaGebremichael I guess cropping is the best way, you could dinamically build an array of repeats to give to np.repeat but I guess it's overengineering and it will probably be slower than cropping. Something like: q, r = np.divmod(3500, 3); repeats = [3] * q + [r] + [0] * (len(arr) - q - 1); newarr = np.repeat(arr, repeats, axis=0)

Collectives™ on Stack Overflow

Replicate multidimensional NumPy array on an axis up-to specific value

1 Answer 1

3 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

3 Comments

Your Answer

Sign up or log in

Post as a guest

Related