You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@tvm.apache.org by GitBox <gi...@apache.org> on 2022/03/24 08:45:58 UTC
[GitHub] [tvm] Fvoiretryzig opened a new pull request #10756: [Relay] Support dcnv2(mask) for pytorch in tvm
Fvoiretryzig opened a new pull request #10756:
URL: https://github.com/apache/tvm/pull/10756
Add dcnV2 support according to this [repository](https://github.com/xi11xi19/CenterNet2TorchScript/tree/master/DCNv2)
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [tvm] Fvoiretryzig edited a comment on pull request #10756: [Relay] Support dcnv2(mask) for pytorch in tvm
Posted by GitBox <gi...@apache.org>.
Fvoiretryzig edited a comment on pull request #10756:
URL: https://github.com/apache/tvm/pull/10756#issuecomment-1078598224
> I have a question there, TorchVision has also implemented a version of DCNv2： https://github.com/pytorch/vision/blob/main/torchvision/csrc/ops/deform_conv2d.h , it should [support](https://github.com/pytorch/vision/blob/56fb0bf5796ac374d4e353032e418236cd73c554/test/test_ops.py#L953-L955) exporting to torchscript. Could we also test this version?
I wrote this operator support for compiling [CenterTrack Model](https://github.com/xingyizhou/CenterTrack) in tvm. It uses dcnv2 from this [repository](https://github.com/CharlesShang/DCNv2/). This implementation doesn't support dcnv2 in torchvision, but it can modify the tvm pytorch frontend part to support. Does this need to change to support torchvision? @zhiqwang
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [tvm] Fvoiretryzig commented on pull request #10756: [Relay] Support dcnv2(mask) for pytorch in tvm
Posted by GitBox <gi...@apache.org>.
Fvoiretryzig commented on pull request #10756:
URL: https://github.com/apache/tvm/pull/10756#issuecomment-1078598224
> I have a question there, TorchVision has also implemented a version of DCNv2： https://github.com/pytorch/vision/blob/main/torchvision/csrc/ops/deform_conv2d.h , it should [support](https://github.com/pytorch/vision/blob/56fb0bf5796ac374d4e353032e418236cd73c554/test/test_ops.py#L953-L955) exporting to torchscript. Could we also test this version?
I wrote this operator support for compiling [CenterTrack Model](https://github.com/xingyizhou/CenterTrack) in tvm. It uses dcnv2 from this [repository](https://github.com/CharlesShang/DCNv2/). This implementation doesn't support dcnv2 in torchvision, but it can modify the tvm pytorch frontend part to support. Does this need to change to support torchvision?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [tvm] masahi commented on pull request #10756: [Relay] Support dcnv2(mask) for pytorch in tvm
Posted by GitBox <gi...@apache.org>.
masahi commented on pull request #10756:
URL: https://github.com/apache/tvm/pull/10756#issuecomment-1078419171
I'm aware that our deform covn2d support lacks `mask` support. If that is what this PR is about, this is great!
Please clean up your change (I see unrelated removal of existing code)
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [tvm] Fvoiretryzig commented on pull request #10756: [Relay] Support dcnv2(mask) for pytorch in tvm
Posted by GitBox <gi...@apache.org>.
Fvoiretryzig commented on pull request #10756:
URL: https://github.com/apache/tvm/pull/10756#issuecomment-1078613582
> I'm aware that our deform covn2d support lacks `mask` support. If that is what this PR is about, this is great!
>
> Please clean up your change (I see unrelated removal of existing code)
I have updated this repository. @masahi @tqchen
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [tvm] zhiqwang commented on pull request #10756: [Relay] Support dcnv2(mask) for pytorch in tvm
Posted by GitBox <gi...@apache.org>.
zhiqwang commented on pull request #10756:
URL: https://github.com/apache/tvm/pull/10756#issuecomment-1077682572
I have a question there, TorchVision has also implemented a version of DCNv2： https://github.com/pytorch/vision/blob/main/torchvision/csrc/ops/deform_conv2d.h , it should [support](https://github.com/pytorch/vision/blob/56fb0bf5796ac374d4e353032e418236cd73c554/test/test_ops.py#L953-L955) exporting to torchscript. Could we also test this version?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org