strisunshinewentingwang/pytorch

Pull Requests Commits

Print tensor shapes and convolution parameters when cuDNN exception is thrown (#45023)

xwang233 committed 3 years ago

b4ba66ae

Move prim::tolist math.log and aten::cpu to lite interpreter for translation model (#45482)

pipibjc committed 3 years ago

93650a82

[TensorExpr] Change API for creating Load and Store expressions. (#45520)

Mikhail Zolotukhin committed 3 years ago

4aca63d3

Fix memory corruption when running torch.svd for complex.doubles (#45486)

malfet committed 3 years ago

772ce9ac

Fix D23995953 import.

Taylor Robie committed 3 years ago

ccad73ab

Enable transposed tensor copy for complex types (#45487)

malfet committed 3 years ago

c87ff2cb

CUDA RTX30 series support (#45489)

zasdfgbnm committed 3 years ago

0a15646e

Enable type-checking of torch.nn.quantized.* modules (#43110)

guilhermeleobas committed 3 years ago

c1e65929

Annotate torch.utils.(tensorboard/show_pickle/hypify) (#44216)

guilhermeleobas committed 3 years ago

375a83e6

Add typing annotations for torch.utils.data.* modules (#44136)

guilhermeleobas committed 3 years ago

eb39542e

Patch generate files for system protobuf (#44583)

peterjc123 committed 3 years ago

33aba57e

ROCm {emoji:2764} TensorExpr (#45506)

t-vi committed 3 years ago

22a34bcf

Disable multi tensor tesnor tests on rocm (#45535)

Iurii Zdebskyi committed 3 years ago

63757040

[PyTorch/NCCL] Fix async error handling (#45456)

jiayisuse committed 3 years ago

06a56637

Create experimental FX graph manipulation library (#44775)

gcatron committed 3 years ago

ef414725

Quantized operators template selective (#45509)

iseeyuan committed 3 years ago

d6429928

fix standard deviation gradient NaN behavior (#45468)

Randall Hunt committed 3 years ago

ab5cf16b

Update backward formula for torch.dot and add backward definition for torch.vdot (#45074)

anjali411 committed 3 years ago

18876b57

Add docs to a pytorch.github.io/doc/tag directory when repo is tagged (#45204)

mattip committed 3 years ago

147c88ef

Updates nonzero's as_tuple behavior to no longer warn. (#45413)

Mike Ruberry committed 3 years ago

b66ac1e9

Remove unnecessary __at_align32__ in int_elementwise_binary_256 (#45470)

xuhdev committed 3 years ago

0df99ad4

Move mobile specific CPUCachingAllocator to c10/mobile folder. (#45364)

kimishpatel committed 3 years ago

6e55a26e

Updates deterministic flag to throw a warning, makes docs consistent (#45410)

Mike Ruberry committed 3 years ago

b2925671

Conservative-ish persistent RNN heuristics for compute capability 8.0+ (#43165)

mcarilli committed 3 years ago

aa2bd7e1

Updated `cholesky_backward` for complex inputs (#45267)

IvanYashchuk committed 3 years ago

f47fd0eb

Support bfloat16 and complex dtypes for logical_not (#43537)

xuhdev committed 3 years ago

15f85eea

Fix model_name not logged properly issue. (#45488)

xcheng16 committed 3 years ago

ea59251f

[JIT] Enable @unused syntax for ignoring properties (#45261)

Meghan Lele committed 3 years ago

09b3e16b

Add mobile_optimized tag to optimized model. (#45479)

Akshit Khurana committed 3 years ago

5f49d14b

[vulkan][android][test_app] Add test_app variant that runs module on Vulkan (#44897)

IvanKobzarev committed 3 years ago

17be7c6e

Newer Older