Add autograd not implemented boxed fallback #63458
Conversation
[ghstack-poisoned]
🔗 Helpful links
💊 CI failures summary and remediationsAs of commit ed27f99 (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 to the (internal) Dr. CI Users group. |
[ghstack-poisoned]
[ghstack-poisoned]
| next_edges.emplace_back(); | ||
| } | ||
| } | ||
| void operator()(const Variable* variable) { |
There was a problem hiding this comment.
Not sure how to remove this overload yet.
See description and discussion from #62450 [ghstack-poisoned]
| _foreach_tensor([&](size_t _, size_t idx_arg, const at::Tensor& t) { | ||
| if (arguments[idx_arg].type()->kind() != c10::TypeKind::OptionalType) { | ||
| TORCH_CHECK(t.defined(), "Expected argument ", idx_arg, " of ", op_name, " to be defined."); | ||
| } |
There was a problem hiding this comment.
I don't feel like this test should be necessary. In none of the original VariableType kernels do we test if the Tensor inputs are defined or not. This is doing some sort of schema checking on the passed in call stack, but actually we should just assume that the stack inputs are well typed.
See description and discussion from #62450 [ghstack-poisoned]
|
@soulitzer has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator. |
See description and discussion from #62450 Differential Revision: [D30518572](https://our.internmc.facebook.com/intern/diff/D30518572) [ghstack-poisoned]
See description and discussion from #62450 Differential Revision: [D30518572](https://our.internmc.facebook.com/intern/diff/D30518572) [ghstack-poisoned]
See description and discussion from #62450 Differential Revision: [D30518572](https://our.internmc.facebook.com/intern/diff/D30518572) [ghstack-poisoned]
|
@soulitzer has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator. |
See description and discussion from #62450 Differential Revision: [D30518572](https://our.internmc.facebook.com/intern/diff/D30518572) [ghstack-poisoned]
See description and discussion from #62450 Differential Revision: [D30518572](https://our.internmc.facebook.com/intern/diff/D30518572) [ghstack-poisoned]
ghstack-source-id: 4c8973d Pull Request resolved: pytorch#63458
See description and discussion from #62450 Differential Revision: [D30518572](https://our.internmc.facebook.com/intern/diff/D30518572) [ghstack-poisoned]
|
@soulitzer has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator. |
See description and discussion from #62450 Differential Revision: [D30518572](https://our.internmc.facebook.com/intern/diff/D30518572) [ghstack-poisoned]
|
@soulitzer has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator. |
|
@soulitzer merged this pull request in 90a6498. |
3 participants
Stack from ghstack:
See description and discussion from #62450
Differential Revision: D30518572