Skip to content

Add instance methods tensor.isnan(), tensor.isinf(), tensor.isfinite()#37942

Closed
ShawnZhong wants to merge 1 commit intopytorch:masterfrom
ShawnZhong:instance_methods
Closed

Add instance methods tensor.isnan(), tensor.isinf(), tensor.isfinite()#37942
ShawnZhong wants to merge 1 commit intopytorch:masterfrom
ShawnZhong:instance_methods

Conversation

@ShawnZhong
Copy link
Copy Markdown
Contributor

@ShawnZhong ShawnZhong commented May 6, 2020 •
edited
Loading

Fix #37736

@ShawnZhong ShawnZhong marked this pull request as ready for review May 6, 2020 18:26
@dr-ci
Copy link
Copy Markdown

dr-ci Bot commented May 6, 2020 •
edited
Loading

💊 CI failures summary and remediations

As of commit 341fb19 (more details on the Dr. CI page):

💚 💚 Looks good so far! There are no failures yet. 💚 💚

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 on the GitHub issue tracker.

See how this bot performed.

This comment has been revised 9 times.

@choidongyeon
Copy link
Copy Markdown

choidongyeon commented May 6, 2020

Hey @ShawnZhong, could you help me understand how this PR fixes the issue? Just curious 😄

@ShawnZhong
Copy link
Copy Markdown
Contributor Author

ShawnZhong commented May 7, 2020

@choidongyeon

pytorch/aten/src/ATen/native/README.md

Lines 197 to 220 in 681c6fb

### `variants`
```
variants: function, method
```
Controls whether Tensor method (`t.foo()`) or namespace Function (`at::foo()`) is
generated as a result of this declaration. If the declaration is a method,
you must have an argument `Tensor self` at some position in the method;
in the method variant this argument will be elided from the argument
list. For example, given the declaration `where(BoolTensor cond, Tensor self, Tensor other)`,
this generates the function `at::where(cond, self, other)` and the method
`self.where(cond, other)`.
By default, ATen generates only the function variant for a native function.
When should you also generate a method variant? Tensor operations as methods
are appropriate for "core" Tensor operations (e.g., add, sub, etc.), but not for
more complicated neural network layers (e.g., `conv2d`) and internal functions
designed specifically for binding (e.g., `cudnn_convolution`).
As we progress along our schema unification of the `func` schema with the JIT
signature schema, we must introduce features that allow us to increase compliance.
One of these features are Tensor annotations. As of now we use naming conventions
to indicate whether an argument of a function is going to be mutated and returned.

@ShawnZhong
Add instance methods tensor.isnan(), tensor.isinf(), tensor.isfinite()
341fb19 
Update _tensor_docs.py

Update tensors.rst

Update _tensor_docs.py
@ShawnZhong
Copy link
Copy Markdown
Contributor Author

ShawnZhong commented May 11, 2020

Since this PR has been around for a while, is there anyone who'd like to review this PR?

@gchanan @soumith

soumith
soumith approved these changes May 11, 2020
View reviewed changes
Copy link
Copy Markdown
Collaborator

@soumith soumith left a comment

Choose a reason for hiding this comment

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

lgtm

facebook-github-bot
facebook-github-bot reviewed May 11, 2020
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.

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

@ShawnZhong ShawnZhong deleted the instance_methods branch May 11, 2020 22:58
@ShawnZhong ShawnZhong mentioned this pull request May 11, 2020
Closed
@facebook-github-bot
Copy link
Copy Markdown
Contributor

facebook-github-bot commented May 12, 2020

@soumith merged this pull request in 5f9b903.

laurentdupin pushed a commit to laurentdupin/pytorch that referenced this pull request Apr 24, 2026
@ShawnZhong @facebook-github-bot
Add instance methods tensor.isnan(), tensor.isinf(), tensor.isfinite() (
4c02c0b 
pytorch#37942)

Summary:
Fix pytorch#37736
Pull Request resolved: pytorch#37942

Differential Revision: D21503150

Pulled By: soumith

fbshipit-source-id: cf6bf57ca67013efe119543f3d9a698473960dec
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@soumith soumith soumith approved these changes

+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

Merged open source

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

[discussion] Add convenience instance methods tensor.isnan(), tensor.isinf(), tensor.isfinite()

6 participants

@ShawnZhong @choidongyeon @facebook-github-bot @soumith @pytorchbot @mruberry