Skip to content

porting convtranspose3d to ATen #22019

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
Closed
wants to merge 4 commits into from
Closed

porting convtranspose3d to ATen #22019

wants to merge 4 commits into from
+2,126 −1,297

Conversation

xmnlab
Copy link
Contributor

@xmnlab xmnlab commented Jun 20, 2019

Resolves #18353

CPU and GPU porting for convolution transpose 3d

@pytorchbot pytorchbot added module: build Build system issues module: cpu CPU specific problem (e.g., perf, algorithm) module: cublas Problem related to cublas support module: cuda Related to torch.cuda, and CUDA support in general module: internals Related to internal abstractions in c10 and ATen module: operators labels Jun 20, 2019
@xmnlab xmnlab force-pushed the 18353-convtranspose3d branch from ce75551 to f938f43 Compare June 20, 2019 16:13
@pytorchbot pytorchbot added the module: nn Related to torch.nn label Jun 20, 2019
@xmnlab xmnlab force-pushed the 18353-convtranspose3d branch from 06b62a4 to ac61d7a Compare June 21, 2019 00:46
@xmnlab xmnlab marked this pull request as ready for review June 21, 2019 20:02
@xmnlab
Copy link
Contributor Author

xmnlab commented Jun 21, 2019

I will try to fix this conflict today

@xmnlab xmnlab force-pushed the 18353-convtranspose3d branch from 66f6b98 to 7eac504 Compare June 22, 2019 00:49
@xmnlab
Copy link
Contributor Author

xmnlab commented Jun 22, 2019

Hey @ailzhang

Do you have any thoughts related to the error with xla job? I renamed the function from conv_transpose2d_backward to aten_conv_transpose2d_backward. But the error seems to be related to the old name.

@ailzhang
Copy link
Contributor

Hi @xmnlab Yea although it's not hard for XLA to follow these changes, why are we adding aten_ prefixes? I think it used to be if in th, we have th prefix but nothing has been added for aten, is this change intended?

@xmnlab
Copy link
Contributor Author

xmnlab commented Jun 23, 2019

hi @ailzhang thanks for the feedback. this change intended to apply the recommendations from another PR #20994 (comment)

what do you recommend for this?

@xmnlab
Copy link
Contributor Author

xmnlab commented Jun 24, 2019

@ailzhang any recommendation here? should I remoe aten_ prefix?

@xmnlab
Copy link
Contributor Author

xmnlab commented Jun 24, 2019

@ailzhang @ezyang I will rebase the code and remove the prefix. If the prefix is necessary we can address that in a follow-up PR

@xmnlab xmnlab force-pushed the 18353-convtranspose3d branch from f5c06d9 to 9c28e06 Compare June 24, 2019 18:35
@xmnlab xmnlab requested a review from ezyang June 24, 2019 21:47
@xmnlab
Copy link
Contributor Author

xmnlab commented Jun 24, 2019

@ezyang it seems it is done for a review!

@soumith soumith changed the title 18353 convtranspose3d porting convtranspose3d to ATen Jun 24, 2019
@soumith soumith added the triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module label Jun 24, 2019
facebook-github-bot
facebook-github-bot reviewed Jun 25, 2019
View reviewed changes
Copy link
Contributor

@facebook-github-bot facebook-github-bot left a comment

Choose a reason for hiding this comment

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

@ezyang is landing this pull request. If you are a Facebook employee, you can view this diff on Phabricator.

@xmnlab xmnlab deleted the 18353-convtranspose3d branch June 25, 2019 18:36
zdevito pushed a commit to zdevito/ATen that referenced this pull request Jun 25, 2019
@xmnlab @facebook-github-bot
porting convtranspose3d to ATen (#22019)
bed1487 
Summary:
Resolves pytorch/pytorch#18353

CPU and GPU porting for convolution transpose 3d
Pull Request resolved: pytorch/pytorch#22019

Differential Revision: D15985353

Pulled By: ezyang

fbshipit-source-id: 1c579577a32db24a1ce38f5ab9b3f1cb9c8f2a6e
@facebook-github-bot
Copy link
Contributor

@ezyang merged this pull request in 7daa96a.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@facebook-github-bot facebook-github-bot facebook-github-bot left review comments

@ezyang ezyang ezyang approved these changes

Assignees

No one assigned

Labels

Merged module: build Build system issues module: cpu CPU specific problem (e.g., perf, algorithm) module: cublas Problem related to cublas support module: cuda Related to torch.cuda, and CUDA support in general module: internals Related to internal abstractions in c10 and ATen module: nn Related to torch.nn module: onnx Related to torch.onnx module: third_party open source triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

Port SpatialFullDilatedConvolution and VolumetricFullDilatedConvolution to ATen

7 participants

@xmnlab @ailzhang @facebook-github-bot @ezyang @soumith @pytorchbot @mruberry