CUDA kernel for ClipGradNorm for TensorSeq gradients#12412
Merged
baijumeswani merged 6 commits intomasterfrom Aug 5, 2022
Merged
CUDA kernel for ClipGradNorm for TensorSeq gradients#12412baijumeswani merged 6 commits intomasterfrom
baijumeswani merged 6 commits intomasterfrom
Conversation
baijumeswani
added
the
training
baijumeswani
force-pushed
the
bmeswani/clipgradnorm
branch
from
1eb026f to
a1119e1
Compare
pengwa
reviewed
Aug 3, 2022
orttraining/orttraining/training_ops/cuda/optimizer/clip_grad_norm/clip_grad_norm.cc
Outdated
Show resolved
Hide resolved
orttraining/orttraining/training_ops/cuda/optimizer/clip_grad_norm/clip_grad_norm.h
Outdated
Show resolved
Hide resolved
orttraining/orttraining/training_ops/cuda/optimizer/clip_grad_norm/clip_grad_norm_impl.cu
Show resolved
Hide resolved
…pull request review comments
askhade
reviewed
Aug 3, 2022
askhade
reviewed
Aug 3, 2022
askhade
reviewed
Aug 3, 2022
askhade
reviewed
Aug 3, 2022
orttraining/orttraining/training_ops/cuda/optimizer/clip_grad_norm/clip_grad_norm.cc
Show resolved
Hide resolved
askhade
reviewed
Aug 3, 2022
orttraining/orttraining/training_ops/cuda/optimizer/clip_grad_norm/clip_grad_norm.cc
Show resolved
Hide resolved
askhade
reviewed
Aug 3, 2022
orttraining/orttraining/training_ops/cuda/optimizer/clip_grad_norm/clip_grad_norm.h
Show resolved
Hide resolved
baijumeswani
force-pushed
the
bmeswani/clipgradnorm
branch
from
536797d to
b91946f
Compare
askhade
reviewed
Aug 4, 2022
askhade
previously approved these changes
Aug 4, 2022
pengwa
approved these changes
Aug 5, 2022
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
3 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Currently,
ClipGradNorminonnxruntimeis executed by running more primitive onnxruntime ops followed by aSequenceConstructbefore feeding intoAdamWOptimizer.This
SequenceConstructwill create another copy of the input tensors and put them into aTensorSeq. To avoid making these copies, theClipGradNormkernel can take in aTensorSeqand clip the gradients in place.