Add batched version of trtrs (#18025)
authorVishwak Srinivasan <cs15btech11043@iith.ac.in>
Wed, 20 Mar 2019 18:06:56 +0000 (11:06 -0700)
committerFacebook Github Bot <facebook-github-bot@users.noreply.github.com>
Wed, 20 Mar 2019 18:11:32 +0000 (11:11 -0700)
commita519217ee72bdef12b5d133b7e941d92d6ccd3cf
tree665982141e548653738f47418ca8922e70f9d41e
parente312801453d1121f929b956c2862dfa58d6b3ac3
Add batched version of trtrs (#18025)

Summary:
- Remove single batch TH/THC implementations
- Remove `_batch_trtrs_lower` from `multivariate_normal`
- Add tests for batched behavior
- Modify trtrs_backward to accommodate for batched case
- Modify docs

In a future PR, this will be renamed to `triangular_solve`.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/18025

Differential Revision: D14523004

Pulled By: ifedan

fbshipit-source-id: 11c6a967d107f969b60e5a5c73ce6bb8099ebbe1
19 files changed:
aten/src/ATen/Declarations.cwrap
aten/src/ATen/native/BatchLinearAlgebra.cpp
aten/src/ATen/native/LegacyDefinitions.cpp
aten/src/ATen/native/cuda/BatchLinearAlgebra.cu
aten/src/ATen/native/native_functions.yaml
aten/src/TH/generic/THLapack.cpp
aten/src/TH/generic/THLapack.h
aten/src/TH/generic/THTensorLapack.cpp
aten/src/TH/generic/THTensorLapack.h
aten/src/THC/generic/THCTensorMathMagma.cu
aten/src/THC/generic/THCTensorMathMagma.h
test/test_autograd.py
test/test_cuda.py
test/test_torch.py
tools/autograd/templates/Functions.cpp
torch/_torch_docs.py
torch/distributions/kl.py
torch/distributions/lowrank_multivariate_normal.py
torch/distributions/multivariate_normal.py