Add complex IValues#50883
Conversation
[ghstack-poisoned]
💊 CI failures summary and remediationsAs of commit c42bd1a (more details on the Dr. CI page):
2 failures not recognized by patterns:
Extra GitHub checks: 1 failed
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. |
[ghstack-poisoned]
SplitInfinity
left a comment
There was a problem hiding this comment.
Nice! After further discussion with the JIT team about the place of ComplexDoubleType in the type system, we concluded that it makes more sense for it to be a subtype of only Any, and not of NumberType. Other than that, the rest looks good!
[ghstack-poisoned]
| // Tuples should be structurally typed | ||
| auto type = TupleType::create({IntType::get(), TensorType::get(), FloatType::get()}); | ||
| auto type2 = TupleType::create({IntType::get(), TensorType::get(), FloatType::get()}); | ||
| auto type = TupleType::create({IntType::get(), TensorType::get(), FloatType::get(), ComplexDoubleType::get()}); |
There was a problem hiding this comment.
Are these changes for testing ComplexDoubleType::operator==?
There was a problem hiding this comment.
yeah and also indirectly testing ComplexDoubleType::get()
Synced offline: It probably makes sense for |
Differential Revision: [D26003682](https://our.internmc.facebook.com/intern/diff/D26003682) [ghstack-poisoned]
Differential Revision: [D26003682](https://our.internmc.facebook.com/intern/diff/D26003682) [ghstack-poisoned]
Differential Revision: [D26003682](https://our.internmc.facebook.com/intern/diff/D26003682) [ghstack-poisoned]
|
@anjali411 merged this pull request in 9ac30d9. |
Summary: Pull Request resolved: pytorch#50883 Test Plan: Imported from OSS Reviewed By: ejguan Differential Revision: D26003682 Pulled By: anjali411 fbshipit-source-id: f02967d2d236d740cd8647891f732f1d63098d3e
Stack from ghstack:
Differential Revision: D26003682