Skip to content

Use specilized kernel for f-arrays and sum by axis=1. Add keepdims support #1489

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 2 commits into from
Jul 21, 2023
Merged

Use specilized kernel for f-arrays and sum by axis=1. Add keepdims support #1489

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
63dcc33
Use specilized kernel for f-arrays and sum by axis=1. Add keepdims su…
AlexanderKalistratov Jul 19, 2023
48348c9
Review remarks
AlexanderKalistratov Jul 20, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
21 changes: 18 additions & 3 deletions dpnp/dpnp_iface_mathematical.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1828,12 +1828,18 @@ def sum(
elif where is not True:
pass
else:
if axis == (0,) and len(x.shape) == 2 and not keepdims:
if len(x.shape) == 2 and (
(axis == (0,) and x.flags.c_contiguous)
or (axis == (1,) and x.flags.f_contiguous)
):
from dpctl.tensor._reduction import _default_reduction_dtype

from dpnp.backend.extensions.sycl_ext import _sycl_ext_impl

input = dpnp.get_usm_ndarray(x)
input = x
if axis == (1,):
input = input.T
input = dpnp.get_usm_ndarray(input)

queue = input.sycl_queue
out_dtype = (
Expand All @@ -1850,7 +1856,16 @@ def sum(

if sum:
sum(input, output, []).wait()
return dpnp_array._create_from_usm_ndarray(output)
result = dpnp_array._create_from_usm_ndarray(output)

if keepdims:
if axis == (0,):
res_sh = (1,) + output.shape
else:
res_sh = output.shape + (1,)
result = result.reshape(res_sh)

return result

y = dpt.sum(
dpnp.get_usm_ndarray(x), axis=axis, dtype=dtype, keepdims=keepdims
Expand Down