Make output 1 of ConcatTraining Optional and place on CPU#7199
Merged
ashbhandare merged 2 commits intomasterfrom Apr 1, 2021
Merged
Make output 1 of ConcatTraining Optional and place on CPU#7199ashbhandare merged 2 commits intomasterfrom
ashbhandare merged 2 commits intomasterfrom
Conversation
SherlockNoMad
previously approved these changes
Mar 31, 2021
Contributor
SherlockNoMad
left a comment
SherlockNoMad
left a comment
There was a problem hiding this comment.
looks good, but let's still update the name.
Contributor
|
looking good on the testing, collecting number shortly |
ashbhandare
force-pushed
the
aibhanda/concat_training
branch
from
7d66cc0 to
2224a35
Compare
SherlockNoMad
previously approved these changes
Apr 1, 2021
ytaous
previously approved these changes
Apr 1, 2021
ashbhandare
force-pushed
the
aibhanda/concat_training
branch
from
2224a35 to
48da6d4
Compare
SherlockNoMad
approved these changes
Apr 1, 2021
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
The Output1 of ConcatTraining is consumed by SplitTraining in the gradient graph, which expects it on host, while ConcatTraining places it on the device. To avoid the DtoH memcpy, this PR places the output1 on host and also makes its computation optional in case it is not consumed by any other node.