Juhyeong (Odd) Kim c77857cb83 Improve numerical stability of torch.distributions.wishart.Wishart (#72993) ... Summary: Maintanance of https://github.com/pytorch/pytorch/pull/70377 Multiple modifications of the merged initial implementation of Wishart distribution. Key modifications: * torch/distributions/wishart.py: Clamp (Clip) float type values to calculate reciprocal in numerically stable manner, by using the eps value paired to each torch.dtype * test/distributions/test_distributions.py: Test Wishart distribution implementation in numerically unstable zones, i.e df values are at ndim - 1 < df < ndim where ndim is the one dimenstion of the Wishart parameter & sample matrix. Re-opened reverted PR https://github.com/pytorch/pytorch/issues/72059 cc neerajprad vadimkantorov Pull Request resolved: https://github.com/pytorch/pytorch/pull/72993 Reviewed By: samdow Differential Revision: D34853807 Pulled By: neerajprad fbshipit-source-id: eb62dca19bf8a934fdf59b4ffc58587447fe8378 (cherry picked from commit 99240c0d947e2faeb7beae91164ede41f61d8448)		2022-03-15 18:30:08 +00:00
..
test_constraints.py	[torch.distributions] Implement positive-semidefinite constraint (#71375)	2022-01-20 17:33:08 +00:00
test_distributions.py	Improve numerical stability of torch.distributions.wishart.Wishart (#72993)	2022-03-15 18:30:08 +00:00
test_transforms.py	Add SoftplusTransform (#52300)	2022-02-25 02:30:03 +00:00
test_utils.py	Set test owner for distributions tests (#66842)	2021-10-19 11:00:29 -07:00