pytorch/torch/autograd
Hameer Abbasi f8b3af21f2 Allow Tensor-likes in torch.autograd.gradcheck (#45732)
Summary:
Fixes https://github.com/pytorch/pytorch/issues/42942

Re-do of https://github.com/pytorch/pytorch/issues/43877.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/45732

Reviewed By: mruberry

Differential Revision: D24195820

Pulled By: albanD

fbshipit-source-id: 8f43353077f341e34371affd76be553c0ef7d98a
2020-10-09 11:51:27 -07:00
..
_functions [quant] qtensor resize (#36442) 2020-04-25 15:52:35 -07:00
__init__.py Allow consumer ops to sync on GraphRoot's gradient (#45787) 2020-10-07 08:53:53 -07:00
anomaly_mode.py Return None instead False, and return bool to None in type stub (#39324) 2020-06-02 10:46:44 -07:00
function.py annotate torch.autograd.* modules (#45004) 2020-10-07 10:53:41 -07:00
functional.py annotate torch.autograd.* modules (#45004) 2020-10-07 10:53:41 -07:00
grad_mode.py Change from self to self.class() in _DecoratorManager to ensure a new object is every time a function is called recursively (#44633) 2020-09-22 09:13:39 -07:00
gradcheck.py Allow Tensor-likes in torch.autograd.gradcheck (#45732) 2020-10-09 11:51:27 -07:00
profiler.py annotate torch.autograd.* modules (#45004) 2020-10-07 10:53:41 -07:00
variable.py annotate torch.autograd.* modules (#45004) 2020-10-07 10:53:41 -07:00