projects / platform / upstream / pytorch.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: b8d1f4a) | patch
Add launch bounds for TopK kernel, be more conservative in sorting (#17296)
authorThomas Viehmann <tv.code@beamnet.de>
Wed, 20 Feb 2019 11:06:53 +0000 (03:06 -0800)
committerFacebook Github Bot <facebook-github-bot@users.noreply.github.com>
Wed, 20 Feb 2019 11:10:46 +0000 (03:10 -0800)
commit237e5438f532418ecb465ca7209e3e2241e32501
tree5ca31ad82b77f942d0c7193e6265413735c81a0ftree | snapshot
parentb8d1f4a423204c9ff2f714151e2d740fb69181facommit | diff
Add launch bounds for TopK kernel, be more conservative in sorting (#17296)

Summary:
The particular use case reported is Jetson TX2 and maskrcnn.

Fixes #17144
Pull Request resolved: https://github.com/pytorch/pytorch/pull/17296

Differential Revision: D14147886

Pulled By: soumith

fbshipit-source-id: 44d5a89aaeb4cc07d1b53dd90121013be93c419c
aten/src/THC/THCTensorTopK.cuh diff | blob | history
aten/src/THC/generic/THCTensorTopK.cu diff | blob | history
Domain: Machine Learning / ML Framework;