cleanup static_cast of AutogradMeta#54103
Closed
albanD wants to merge 3 commits intogh/albanD/73/basefrom
Closed
Conversation
[ghstack-poisoned]
This was referenced Mar 16, 2021
Contributor
💊 CI failures summary and remediationsAs of commit 860639e (more details on the Dr. CI page):
This comment was automatically generated by Dr. CI (expand for details).Follow this link to opt-out of these comments for your Pull Requests.Please report bugs/suggestions to the (internal) Dr. CI Users group. |
soulitzer
reviewed
Mar 17, 2021
soulitzer
reviewed
Mar 17, 2021
soulitzer
reviewed
Mar 17, 2021
| return static_cast<AutogradMeta*>(self.unsafeGetTensorImpl()->autograd_meta()); | ||
| } | ||
|
|
||
| DifferentiableViewMeta* get_view_autograd_meta(const Variable& self) { |
Contributor
There was a problem hiding this comment.
nit: Would we want to use Tensor here? or do we keep Variable for consistency.
Collaborator
Author
There was a problem hiding this comment.
Yes we have a using Variable = at::Tensor at the top of the header.
We should remove that and rename all of them I think yes.
The goal is to reduce the spread of static casts in the autograd code as per the comment in #49097 (comment) I wasn't sure how to use a virtual method here but a simple method in impl clean it up quite nicely. [ghstack-poisoned]
Contributor
|
This seems OK, but given that all the call sites are testing for view-ness before calling the cast, it seems like it might just be better to make this function return a nullptr/ullopt when it's not castable, and then force callers to test the return output. |
albanD
added a commit
to albanD/pytorch
that referenced
this pull request
Mar 17, 2021
ghstack-source-id: 4cdb26e Pull Request resolved: pytorch#54103
The goal is to reduce the spread of static casts in the autograd code as per the comment in #49097 (comment) I wasn't sure how to use a virtual method here but a simple method in impl clean it up quite nicely. Differential Revision: [D27117840](https://our.internmc.facebook.com/intern/diff/D27117840) [ghstack-poisoned]
Collaborator
Author
|
@ezyang I updated following your recommendation. Does that look better? It does make the code simpler but it is a less clear that we use this to check if a Tensor is a view. |
ezyang
approved these changes
Mar 17, 2021
Contributor
|
One way to make it clearer is to jiggle the naming; e.g., |
Contributor
This was referenced Mar 24, 2021
Closed
albanD
added a commit
that referenced
this pull request
Mar 29, 2021
The `.is_view()` method actually only refers to backward mode views This is not a problem right now in master (and thus I didn't revert the other PR) because nothing creates forward AD views. [ghstack-poisoned]
albanD
added a commit
that referenced
this pull request
Mar 29, 2021
The `.is_view()` method actually only refers to backward mode views This is not a problem right now in master (and thus I didn't revert the other PR) because nothing creates forward AD views. [ghstack-poisoned]
albanD
added a commit
to albanD/pytorch
that referenced
this pull request
Mar 29, 2021
ghstack-source-id: b26a8be Pull Request resolved: pytorch#54610
facebook-github-bot
pushed a commit
that referenced
this pull request
Apr 5, 2021
Summary: Pull Request resolved: #54610 The `.is_view()` method actually only refers to backward mode views This is not a problem right now in master (and thus I didn't revert the other PR) because nothing creates forward AD views. Test Plan: Imported from OSS Reviewed By: gchanan Differential Revision: D27396756 Pulled By: albanD fbshipit-source-id: 64ff11c6f2486c6430714988d1cf6ecf3d80dccb
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
The goal is to reduce the spread of static casts in the autograd code as per the comment in #49097 (comment)
I wasn't sure how to use a virtual method here but a simple method in impl clean it up quite nicely.
Stack from ghstack:
Differential Revision: D27117840