-
Notifications
You must be signed in to change notification settings - Fork 6.8k
[Feature] Add bfloat to oneDNN version of binary broadcast operators. #20846
Conversation
Hey @agrabows , Thanks for submitting the PR
CI supported jobs: [windows-cpu, centos-cpu, unix-cpu, windows-gpu, centos-gpu, clang, website, edge, unix-gpu, sanity, miscellaneous] Note: |
@mxnet-bot run ci [windows-gpu] |
Jenkins CI successfully triggered : [windows-gpu] |
@mxnet-bot run ci [windows-gpu] |
Jenkins CI successfully triggered : [windows-gpu] |
@mxnet-bot run ci [windows-gpu] |
Jenkins CI successfully triggered : [windows-gpu] |
f4f172b
to
ff895c2
Compare
@mxnet-bot run ci [windows-gpu] |
Jenkins CI successfully triggered : [windows-gpu] |
@mxnet-bot run ci [unix-cpu] |
Jenkins CI successfully triggered : [unix-cpu] |
@mxnet-bot run ci [windows-gpu] |
Jenkins CI successfully triggered : [windows-gpu] |
ff895c2
to
2fc5e4d
Compare
Description
Binary broadcast operators such as broadcast_add are supported by oneDNN but not when inputs are of type bFloat16. Goal was to add those cases.
Checklist
Essentials