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 2020/06/09 19:33:16 UTC

[GitHub] [incubator-tvm] abergeron opened a new pull request #5758: Fix the meaning of conv{1,2}d_transpose output_padding parameter.

abergeron opened a new pull request #5758:
URL: https://github.com/apache/incubator-tvm/pull/5758


   I finally had time to work on it again so this is a re-attempt at https://github.com/apache/incubator-tvm/pull/4318, which was reverted.
   
   ---
   
   For some context the tvm meaning for output_padding is to pad the output shape with zeros.
   
   For every other deep learning framework, this parameter name is used to disambiguate between the possible output shapes in the presence of padding (Unless they have an explicit output_shape parameter like tensorflow, which doesn't have output_padding).
   
   The current behaviour is in my mind redundant since you can always zero-pad manually if you need to. Also there is no way to compute the gradient of a convolution with strides that did not have the minimum possible input shape.
   
   The new behaviour makes this possible and brings us in-line with the other frameworks.
   
   ---
   
   I've rebased the code and added checks to make sure an invalid output_padding is refused.
   
   As for the comments by @icemelon9:
   
   1) The aformentioned check should cover that case and prevent going out of bounds.
   2) Since the preprocess function handles everything that we need output_padding for, we don't need to pass it forward.
   
   My best guess for why the CI didn't catch the cases is that going just a little bit out of bounds (like of by one) is unlikely to trigger any memory problems.


----------------------------------------------------------------
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.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org



[GitHub] [incubator-tvm] abergeron commented on pull request #5758: Fix the meaning of conv{1,2}d_transpose output_padding parameter.

Posted by GitBox <gi...@apache.org>.
abergeron commented on pull request #5758:
URL: https://github.com/apache/incubator-tvm/pull/5758#issuecomment-641526216


   The only thing probably missing is an update of tophub for the missing output_padding arguments.


----------------------------------------------------------------
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.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org



[GitHub] [incubator-tvm] siju-samuel commented on pull request #5758: Fix the meaning of conv{1,2}d_transpose output_padding parameter.

Posted by GitBox <gi...@apache.org>.
siju-samuel commented on pull request #5758:
URL: https://github.com/apache/incubator-tvm/pull/5758#issuecomment-651576650


   @abergeron Thanks for bringing it up and the PR. I  followed the existing conv2d_transpose implementation.  You can add another PR to fix for fixing this issue.


----------------------------------------------------------------
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.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org



[GitHub] [incubator-tvm] vinx13 merged pull request #5758: Fix the meaning of conv{1,2}d_transpose output_padding parameter.

Posted by GitBox <gi...@apache.org>.
vinx13 merged pull request #5758:
URL: https://github.com/apache/incubator-tvm/pull/5758


   


----------------------------------------------------------------
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.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org



[GitHub] [incubator-tvm] abergeron commented on pull request #5758: Fix the meaning of conv{1,2}d_transpose output_padding parameter.

Posted by GitBox <gi...@apache.org>.
abergeron commented on pull request #5758:
URL: https://github.com/apache/incubator-tvm/pull/5758#issuecomment-651238029


   Sorry about the delay, I was distracted by another project.


----------------------------------------------------------------
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.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org



[GitHub] [incubator-tvm] vinx13 commented on pull request #5758: Fix the meaning of conv{1,2}d_transpose output_padding parameter.

Posted by GitBox <gi...@apache.org>.
vinx13 commented on pull request #5758:
URL: https://github.com/apache/incubator-tvm/pull/5758#issuecomment-643704526


   The tophub PR has been merged, please restart CI to try again


----------------------------------------------------------------
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.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org



[GitHub] [incubator-tvm] abergeron commented on pull request #5758: Fix the meaning of conv{1,2}d_transpose output_padding parameter.

Posted by GitBox <gi...@apache.org>.
abergeron commented on pull request #5758:
URL: https://github.com/apache/incubator-tvm/pull/5758#issuecomment-642101273


   I've confirmed manually that with the tophub changes the vta tests pass.


----------------------------------------------------------------
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.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org



[GitHub] [incubator-tvm] abergeron commented on pull request #5758: Fix the meaning of conv{1,2}d_transpose output_padding parameter.

Posted by GitBox <gi...@apache.org>.
abergeron commented on pull request #5758:
URL: https://github.com/apache/incubator-tvm/pull/5758#issuecomment-651331355


   I saw that @siju-samuel added the conv3d_transpose op but with the wrong definition for output_padding.
   
   I would rather make a follow-up PR to fix that since I want this one to be merged if it's ok and I don't have a lot of free time currently.
   
   I will make a follow-up PR to fix the definition of output_padding in conv3d_transpose later if nobody else does it before me.


----------------------------------------------------------------
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.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org



[GitHub] [incubator-tvm] vinx13 commented on pull request #5758: Fix the meaning of conv{1,2}d_transpose output_padding parameter.

Posted by GitBox <gi...@apache.org>.
vinx13 commented on pull request #5758:
URL: https://github.com/apache/incubator-tvm/pull/5758#issuecomment-646937540


   ping @abergeron 


----------------------------------------------------------------
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.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org