Skip to content

cuda4dnn(eltwise): tensor broadcasting#20782

Merged
opencv-pushbot merged 1 commit intoopencv:masterfrom
YashasSamaga:cuda4dnn-eltwise-broadcast
Oct 4, 2021
Merged

cuda4dnn(eltwise): tensor broadcasting#20782
opencv-pushbot merged 1 commit intoopencv:masterfrom
YashasSamaga:cuda4dnn-eltwise-broadcast

Conversation

@YashasSamaga
Copy link
Copy Markdown
Contributor

@YashasSamaga YashasSamaga commented Sep 30, 2021 •
edited
Loading

EltwiseOp requires all the input tensors to have the same shape. This PR generalizes it to support eltwise operations on any set of compatible tensors as input.

resolves #20778

The following tests were failing:

[  FAILED  ] Test_TensorFlow_layers.eltwise_add_vec/0, where GetParam() = CUDA/CUDA
[  FAILED  ] Test_TensorFlow_layers.eltwise_add_vec/1, where GetParam() = CUDA/CUDA_FP16
[  FAILED  ] Test_TensorFlow_layers.eltwise_mul_vec/0, where GetParam() = CUDA/CUDA
[  FAILED  ] Test_TensorFlow_layers.eltwise_mul_vec/1, where GetParam() = CUDA/CUDA_FP16
[  FAILED  ] Test_TensorFlow_layers.channel_broadcast/0, where GetParam() = CUDA/CUDA
[  FAILED  ] Test_TensorFlow_layers.channel_broadcast/1, where GetParam() = CUDA/CUDA_FP16
[  FAILED  ] Test_TensorFlow_nets.EfficientDet/0, where GetParam() = CUDA/CUDA
[  FAILED  ] Test_TensorFlow_nets.EfficientDet/1, where GetParam() = CUDA/CUDA_FP16

The listed tests now pass.

Pull Request Readiness Checklist

See details at https://github.com/opencv/opencv/wiki/How_to_contribute#making-a-good-pull-request

  • I agree to contribute to the project under Apache 2 License.
  • To the best of my knowledge, the proposed patch is not based on a code under GPL or other license that is incompatible with OpenCV
  • The PR is proposed to proper branch
  • There is reference to original bug report and related work
  • There is accuracy test, performance test and test data in opencv_extra repository, if applicable
    Patch to opencv_extra has the same branch name.
  • The feature is well documented and sample code can be built with the project CMake
force_builders=Custom
buildworker:Custom=linux-4
build_image:Custom=ubuntu-cuda-cc52:18.04
Xbuild_image:Custom=ubuntu-cuda:18.04

@alalek
Copy link
Copy Markdown
Member

alalek commented Oct 3, 2021

Draft

@YashasSamaga Thank you for the contribution! Is this ready to be merged?

@YashasSamaga
Copy link
Copy Markdown
Contributor Author

YashasSamaga commented Oct 3, 2021

Thank you for the contribution! Is this ready to be merged?

No. I will finish in 24 hours.

@YashasSamaga YashasSamaga force-pushed the cuda4dnn-eltwise-broadcast branch from 3200703 to 505dde0 Compare October 4, 2021 07:09
@YashasSamaga YashasSamaga marked this pull request as ready for review October 4, 2021 07:09
@asmorkalov asmorkalov requested a review from JulieBar October 4, 2021 08:05
alalek
alalek approved these changes Oct 4, 2021
View reviewed changes
Copy link
Copy Markdown
Member

@alalek alalek left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Well done! Thank you 👍

@opencv-pushbot opencv-pushbot merged commit 1b70f94 into opencv:master Oct 4, 2021
@alalek alalek mentioned this pull request Oct 15, 2021
Merged
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@alalek alalek alalek approved these changes

@JulieBar JulieBar Awaiting requested review from JulieBar

Assignees

@alalek alalek

Labels

category: dnn category: gpu/cuda (contrib) OpenCV 4.0+: moved to opencv_contrib feature

Projects

None yet

Milestone

4.5.4

Development

Successfully merging this pull request may close these issues.

error: (-217:Gpu API call) invalid device function in function 'make_policy'

3 participants

@YashasSamaga @alalek @opencv-pushbot