Skip to content

Remove deprecate warning for Upsampling#11568

Closed
ailzhang wants to merge 3 commits intopytorch:masterfrom
ailzhang:interpolate_module
Closed

Remove deprecate warning for Upsampling#11568
ailzhang wants to merge 3 commits intopytorch:masterfrom
ailzhang:interpolate_module

Conversation

@ailzhang
Copy link
Copy Markdown
Contributor

@ailzhang ailzhang commented Sep 12, 2018

Fixes #11452 .

Based on the discussion with @ssnl and @soumith , we want to bring back Upsample as a module instead of introducing a new nn.interpolate module for now. If anyone want to do downsample, they should use nn.functional.interpolate instead.

ssnl
ssnl reviewed Sep 12, 2018
View reviewed changes
Copy link
Copy Markdown
Collaborator

@ssnl ssnl left a comment

Choose a reason for hiding this comment

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

That was fast! Could you add a .. note:: to the doc and tell users if they want to resize/downsample, they should use torch.nn.functional.interpolate? (Btw, does interpolate support general resizing, i.e., to arbitrary size?)

@ailzhang
Copy link
Copy Markdown
Contributor Author

ailzhang commented Sep 12, 2018

@ssnl Yep it does! Actually the tricky part is, if users use nn.Upsample for downsampling, that still works. Since it calls F.interpolate inside nn.Upsample :P So they don't have to use F.interpolate despite the misleading name.

ssnl
ssnl reviewed Sep 12, 2018
View reviewed changes
Comment thread torch/nn/modules/upsampling.py
size (tuple, optional): a tuple of ints `(H_out, W_out)` output sizes
scale_factor (int, optional): the multiplier for the image height or width

.. warning::

This comment was marked as off-topic.

Sign in to view

This comment was marked as off-topic.

Sign in to view

@ssnl
Copy link
Copy Markdown
Collaborator

ssnl commented Sep 12, 2018

@ailzhang Okay. But we can still link to F.interpolate :)

@vadimkantorov
Copy link
Copy Markdown
Contributor

vadimkantorov commented Sep 12, 2018 •
edited
Loading

@ssnl A crazy idea: how about a generic nn.Wrap(F.interpolate, scale_factor = 2) - can make sense for stateless functions, will unblock people on absence of some modules, and even crazier nn.Sequential().add(F.interpolate, **kwargs) can exist.

facebook-github-bot
facebook-github-bot reviewed Sep 12, 2018
View reviewed changes
Copy link
Copy Markdown
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.

ailzhang has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator.

@ailzhang
Copy link
Copy Markdown
Contributor Author

ailzhang commented Sep 13, 2018

@pytorchbot retest this please

ssnl
ssnl approved these changes Sep 13, 2018
View reviewed changes
Comment thread torch/nn/modules/upsampling.py Outdated
.. warning::
This class is deprecated in favor of :func:`~nn.functional.interpolate`.
.. note::
If you want down sampling, you should use :func:`~nn.functional.interpolate`.

This comment was marked as off-topic.

Sign in to view

facebook-github-bot
facebook-github-bot reviewed Sep 13, 2018
View reviewed changes
Copy link
Copy Markdown
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.

ailzhang has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator.

@ailzhang
Copy link
Copy Markdown
Contributor Author

ailzhang commented Sep 14, 2018

@pytorchbot retest this please

@nehz
Copy link
Copy Markdown
Contributor

nehz commented Mar 11, 2019

Any updates on this ?

@gchanan
Copy link
Copy Markdown
Contributor

gchanan commented Mar 11, 2019

@nehz this has already been committed.

@nehz
Copy link
Copy Markdown
Contributor

nehz commented Mar 12, 2019 •
edited
Loading

I'm still seeing it on the master and 1.0.1 branch ?
https://github.com/pytorch/pytorch/blob/master/torch/nn/modules/upsampling.py
warnings.warn("nn.{} is deprecated. Use nn.functional.interpolate instead.".format(self.name))

@ssnl ssnl mentioned this pull request Mar 12, 2019
Closed
@ssnl
Copy link
Copy Markdown
Collaborator

ssnl commented Mar 12, 2019

blaming #14123 cc @eellison

@gchanan
Copy link
Copy Markdown
Contributor

gchanan commented Mar 12, 2019

@ailzhang ?

@ailzhang
Copy link
Copy Markdown
Contributor Author

ailzhang commented Mar 12, 2019

Yea this got added back accidentally in another PR. I'm sending a quick patch to fix. Thanks @ssnl @gchanan !

@ailzhang ailzhang mentioned this pull request Mar 12, 2019
Closed
facebook-github-bot pushed a commit that referenced this pull request Mar 12, 2019
@facebook-github-bot
remove warning for upsample code (#17921)
3e00f79 
Summary:
IIRC we decided to remove warning in code in #11568. This got reverted accidentally in #14123.
Pull Request resolved: #17921

Differential Revision: D14422811

Pulled By: ailzhang

fbshipit-source-id: 7067264bd1d3e3b7861d29e18ade2969ed705ca1
@ezyang ezyang added the merged label Jun 26, 2019
laurentdupin pushed a commit to laurentdupin/pytorch that referenced this pull request Apr 24, 2026
@facebook-github-bot
Remove deprecate warning for Upsampling (pytorch#11568)
3597f9b 
Summary:
Fixes pytorch#11452 .

Based on the discussion with SsnL  and soumith , we want to bring back Upsample as a module instead of introducing a new nn.interpolate module for now. If anyone want to do downsample, they should use `nn.functional.interpolate ` instead.
Pull Request resolved: pytorch#11568

Differential Revision: D9804359

Pulled By: ailzhang

fbshipit-source-id: 2b232d55fc83c2b581bf336f1ee8d1cf1c1159ca
laurentdupin pushed a commit to laurentdupin/pytorch that referenced this pull request Apr 24, 2026
@facebook-github-bot
remove warning for upsample code (pytorch#17921)
810ad1b 
Summary:
IIRC we decided to remove warning in code in pytorch#11568. This got reverted accidentally in pytorch#14123.
Pull Request resolved: pytorch#17921

Differential Revision: D14422811

Pulled By: ailzhang

fbshipit-source-id: 7067264bd1d3e3b7861d29e18ade2969ed705ca1
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@ssnl ssnl ssnl approved these changes

@apaszke apaszke Awaiting requested review from apaszke

@colesbury colesbury Awaiting requested review from colesbury

@ezyang ezyang Awaiting requested review from ezyang

@gchanan gchanan Awaiting requested review from gchanan

@soumith soumith Awaiting requested review from soumith

@zdevito zdevito Awaiting requested review from zdevito

+1 more reviewer

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

Reviewers whose approvals may not affect merge requirements

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

Upsampling modules deprecated

7 participants

@ailzhang @ssnl @vadimkantorov @nehz @gchanan @facebook-github-bot @ezyang