Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

ConvTransposeGrad CUDA Kernel #17201

Merged
merged 6 commits into from
Aug 24, 2023
Merged

ConvTransposeGrad CUDA Kernel #17201

merged 6 commits into from
Aug 24, 2023

Conversation

baijumeswani
Copy link
Contributor

This pull request adds a CUDA kernel implementation for the convolution transpose gradient operation to ONNX Runtime.

Work here is heavily derived from the previous work by @er3x3 in #8613 and from PyTorch convolution kernels.

@baijumeswani baijumeswani added the training issues related to ONNX Runtime training; typically submitted using template label Aug 17, 2023
@baijumeswani baijumeswani merged commit fca81cc into main Aug 24, 2023
95 checks passed
@baijumeswani baijumeswani deleted the baijumeswani/convtransposegrad branch August 24, 2023 16:08
@baijumeswani
Copy link
Contributor Author

Thank you for the review @er3x3 :)
After the release, please let me know if you feel there are any significant changes I should make.

@faxu faxu added the triage:approved Approved for cherrypicks for release label Aug 24, 2023
@tianleiwu
Copy link
Contributor

@baijumeswani, please take a look at test error, and update the test:

https://dev.azure.com/onnxruntime/onnxruntime/_build/results?buildId=1122071&view=logs&j=5076e696-f193-5f12-2d8a-703dda41a79b&t=318cbefa-99dd-503e-0fe4-581779019aa9&l=14091
1: [ RUN ] GradientCheckerTest.ConvTransposeGrad
1: D:\a_work\1\s\orttraining\orttraining\test\gradient\gradient_ops_test.cc(3128): error: Value of: IsErrorWithinTolerance(max_error, error_tolerance)
1: Actual: false
1: Expected: true
1: max_error: 0.18644331395626068; tolerance: 0.10000000149011612; ORT test random seed: 2294411604;
1: [ FAILED ] GradientCheckerTest.ConvTransposeGrad (33043 ms)

@er3x3 er3x3 mentioned this pull request Aug 28, 2023
Merged
@baijumeswani baijumeswani mentioned this pull request Aug 28, 2023
Merged
snnn pushed a commit that referenced this pull request Aug 28, 2023
@er3x3
Cherry-pick 1st Round (#17308)
198fc90 
Cherry-pick 1st round for rel-1.16.0 from
https://github.com/microsoft/onnxruntime/issues?q=label%3Arelease%3A1.16+label%3Atriage%3Aapproved+is%3Aclosed
except #17201 because it caused UT failure and is not fixed yet.

PR list:
#16417
#16936
#17000
#17236
#17238
#17240
#17252
#17255
#17258
#17265
#17267
#17277
er3x3 pushed a commit that referenced this pull request Sep 1, 2023
@baijumeswani @er3x3
ConvTransposeGrad CUDA Kernel (#17201)
8d07173
snnn pushed a commit that referenced this pull request Sep 7, 2023
@er3x3
Cherry-pick 2nd Round (#17386)
4296043 
Cherry-pick 2nd round for 1.16.0 release.
PR List:

#17201
#17270
#17311
#17315
#17320
#17326
#17355
#17227
#17380
#17386
@baijumeswani baijumeswani mentioned this pull request Sep 27, 2023
Closed
kleiti pushed a commit to kleiti/onnxruntime that referenced this pull request Mar 22, 2024
@baijumeswani @kleiti
ConvTransposeGrad CUDA Kernel (microsoft#17201)
546d36d
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@er3x3 er3x3 er3x3 approved these changes

@askhade askhade Awaiting requested review from askhade

@pengwa pengwa Awaiting requested review from pengwa

Assignees
No one assigned
Labels
training issues related to ONNX Runtime training; typically submitted using template triage:approved Approved for cherrypicks for release
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.

None yet
4 participants
@baijumeswani @tianleiwu @er3x3 @faxu