-
Notifications
You must be signed in to change notification settings - Fork 6.8k
Use cuDNN for conv bias and bias grad #20771
Conversation
Hey @mk-61 , Thanks for submitting the PR
CI supported jobs: [clang, centos-cpu, unix-cpu, windows-gpu, windows-cpu, miscellaneous, unix-gpu, website, sanity, centos-gpu, edge] Note: |
464a07e
to
ff7e5a1
Compare
de9b39a
to
0ac4a0b
Compare
@mxnet-bot run ci [unix-cpu, windows-gpu] |
Jenkins CI successfully triggered : [unix-cpu, 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] |
@mxnet-bot run ci [unix-gpu] |
Jenkins CI successfully triggered : [unix-gpu] |
@mxnet-bot run ci [unix-cpu] |
Jenkins CI successfully triggered : [unix-cpu] |
@mxnet-bot run ci [unix-cpu] |
Jenkins CI successfully triggered : [unix-cpu] |
@mxnet-bot run ci [unix-cpu] |
Jenkins CI successfully triggered : [unix-cpu] |
@szha please help review. |
@josephevans Could you help to check these CI failure? It seems not from the MXNet. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Looks good, thanks!
Description
This change uses cuDNN implementation of bias and bias gradient, if available. It mitigates performance regression in convolutional networks in 2.0 vs 1.x, as tested with ResNet training script.
Checklist
Essentials
Other changes