Conversation
|
View rendered docs @ https://intelpython.github.io/dpctl/pulls/1299/index.html |
|
Array API standard conformance tests for dpctl=0.14.6dev0=py310h7bf5fec_24 ran successfully. |
|
Array API standard conformance tests for dpctl=0.14.6dev0=py310h7bf5fec_29 ran successfully. |
37c7b2a to
6ba7391
Compare
dpctl/tensor/libtensor/include/kernels/elementwise_functions/round.hpp
Outdated
Show resolved
Hide resolved
|
Array API standard conformance tests for dpctl=0.14.6dev0=py310h7bf5fec_29 ran successfully. |
6ba7391 to
fb35be7
Compare
|
Array API standard conformance tests for dpctl=0.14.6dev0=py310h7bf5fec_29 ran successfully. |
|
Perhaps it is worth remarking that present implementation relies on vector engines of GPU using rounding mode 'round_to_nearest'. Should this change this implementation might break. This is why I was insisting that the With this remark, it looks good to me too. |
|
Deleted rendered PR docs from intelpython.github.com/dpctl, latest should be updated shortly. 🤞 |
|
Array API standard conformance tests for dpctl= ran successfully. |
In this PR,
dpctl.tensor.roundis implemented and tested.Present implementation relies on vector engines of GPU using rounding mode 'round_to_nearest'. Should this change this implementation might break.