Allow creation of *_like() arrays of arbitrary shapes #13043
Labels
Comments
|
This sounds like a good idea to me. For context, @pentschev has been
trying to get dask.array and cupy working for a while and apparently has
run into this limitation several times.
…On Tue, Feb 26, 2019 at 5:10 AM Peter Andreas Entschev < ***@***.***> wrote:
Following the discussion in dask/dask#4462
<dask/dask#4462>, in some cases it's necessary
to create a new array of the same NumPy-like type (e.g., sparse or CuPy) as
the input array. This can be achieved by using
{empty,full,ones,zeros}_like() functions. However, this prevents a new
array of arbitrary shape to be created.
@shoyer <https://github.com/shoyer> suggested the following change in
dask/dask#4462 <dask/dask#4462>:
What about adding a shape keyword argument to NumPy's zeros_like,
ones_like and full_like? There's some precedence for overwriting
attributes of the original array with the dtype argument already.
I think this is a good solution and I would like to see this implemented.
I will be opening a PR for this soon.
@mrocklin <https://github.com/mrocklin> @hameerabbasi
<https://github.com/hameerabbasi> FYI
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#13043>, or mute the thread
<https://github.com/notifications/unsubscribe-auth/AASszFFKqremlIGGOp6dNb9qHG5GPygrks5vRTJRgaJpZM4bSIyR>
.
|
|
The only possible downside would be if some of these get dispatched through ufuncs right now and that might make things tricky. Other then that, it seems a natural thing to just offer it. |
|
I've sent PR #13046 for evaluation/review, please also refer to that for suggested implementation. |
|
PR merged, closing. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Following the discussion in dask/dask#4462, in some cases it's necessary to create a new array of the same NumPy-like type (e.g., sparse or CuPy) as the input array. This can be achieved by using
{empty,full,ones,zeros}_like()functions. However, this prevents a new array of arbitrary shape to be created.@shoyer suggested the following change in dask/dask#4462:
I think this is a good solution and I would like to see this implemented. I will be opening a PR for this soon.
@mrocklin @hameerabbasi FYI
The text was updated successfully, but these errors were encountered: