strisunshinewentingwang/pytorch

Pull Requests Commits

Add/fallback some operators for mkl-dnn (#11696)

PenghuiCheng committed 5 years ago

926e718d

Don't call cudaStreamDestroy at destruction time (#15692)

Dmytro Dzhulgakov committed 5 years ago

96ea2594

Tensor construction codemod(ResizeLike) - 1/3 (#15944)

jerryzh168 committed 5 years ago

726341fe

Move nightly binary builds to 05:05 UTC (#15966)

pjh5 committed 5 years ago

bcc88dfb

Add backend checks for batch norm (#15955)

vishwakftw committed 5 years ago

e07cca13

Add scalar_type_to_pytorch_type dict in ONNX symbolic

zrphercule committed 5 years ago

c9d7ead0

Register CPU/CUDA fuser dynamically (#15887)

zdevito committed 5 years ago

3f6b212e

Simplify cat fusion (#15633)

apaszke committed 5 years ago

d580d358

Add bindings for .cpu() & .cuda() to script (#15904)

Elias Ellison committed 5 years ago

3d0d16d3

comment out large test cases for tril(u)_indices (#15959)

mrshenli committed 5 years ago

03a570ca

Automatic update of fbcode/onnx to 7abd834091f1024c11749dcfd25126802db9fdd5 (#15942)

houseroad committed 5 years ago

7841fe4f

Match NumPy by considering NaNs to be larger than any number when sorting (#15886)

Brennan Vincent committed 5 years ago

70dd44f6

Port empty_strided to ATen. (#15948)

gchanan committed 5 years ago

b7cdeb3f

Move cudaDeviceProp to ATen (#14834)

syed-ahmed committed 5 years ago

14dcdc4c

Trivial typo fixings in nn.functional dropout* docstrings (#15951)

bombs-kim committed 5 years ago

da753b7c

Resolves ptxas warnings when compiling for CUDA_ARCH 750 and a memoryType deprecation warning (#15461)

syed-ahmed committed 5 years ago

86af14b0

Fix fallback issues to handle inplace case (#15726)

gujinghui committed 5 years ago

07ea3e03

Optimize CPU version performance of the nonzero function. (#15925)

VitalyFedyunin committed 5 years ago

0934e8de

Tensor reinitialization codemod - 5/5 (#15884)

jerryzh168 committed 5 years ago

890568a0

Add backward pass notes for eig() and symeig()

Randl committed 5 years ago

e46e572b

caffe2::Tensor::is_same() (#15407)

smessmer committed 5 years ago

da746885

Clean up Half (#15317)

smessmer committed 5 years ago

b9e1028c

Move files to/from c10/core and c10/util (#15316)

smessmer committed 5 years ago

d4083243

Remove Context from c10 operator schemas (#15312)

smessmer committed 5 years ago

6b64052e

Enable calling caffe2 LayerNorm from PyTorch and JIT (#15243)

smessmer committed 5 years ago

8136c39b

zdevito committed 5 years ago

91378544

Remove USE_CUDA and USE_ROCM in engine.cpp

bddppq committed 5 years ago

27f6a29f

Extend note about contributing to the C++ frontend (#15902)

goldsborough committed 5 years ago

c5012d86

Fix different env variables in schedules runs pt 2 (#15934)

pjh5 committed 5 years ago

3ec33513

Change PoolOp Functors design to support CuDNN CUDA fallback (#15903)

xiaomengy committed 5 years ago

4b427780

Newer Older