strisunshinewentingwang/pytorch

Pull Requests Commits

[caffe2][torch] Clean up unused variable 'device' (#48600)

jdonald committed 3 years ago

acd4fca3

Testing: Improve interaction between dtypes and ops decorators (#48426)

peterbell10 committed 3 years ago

9500e8a0

[quant][fix] Add bias once in conv_fused (#48593)

jerryzh168 committed 3 years ago

d2e42986

add aot_based_partition (#48336)

scottxu0730 committed 3 years ago

7a59a1b5

[Gradient Compression] Add a random generator to PowerSGD state for initializing low-rank matrix Q (#48507)

Yi Wang committed 3 years ago

ddb65949

[PyTorch][JIT] Parameter passing & std::map API usage pass on ProfilingRecord::instrumentGraph (#47960)

swolchok committed 3 years ago

61936cb1

caffe2: refactor context to allow being typed (#48340)

d4l3k committed 3 years ago

dc7d8a88

[te] Fix comparison ops on booleans (#48384)

bertmaher committed 3 years ago

adb4fd3f

[TensorExpr] Add a envvar to disable LLVM backend and use IR Eval instead. (#48355)

Mikhail Zolotukhin committed 3 years ago

d9f5ac08

[TensorExpr] IREval: fix div for Half dtype. (#48354)

Mikhail Zolotukhin committed 3 years ago

a6f0c3c4

Disable fast sigmoid since it causes divergence (#48623)

bertmaher committed 3 years ago

671a9592

Fused8BitRowwiseQuantizedToFloat operator support (#48407)

Frank Seide committed 3 years ago

29f0e1e2

remove unused params in scalar_tensor_static (#48550)

Basil Hosmer committed 3 years ago

c3bb3827

[vulkan] Fix Addmm prepacking to persist after GPU flush (#48313)

SS-JIA committed 3 years ago

ea0ffbb6

Update Windows CI to CUDA 11.1, cuDNN 8.0.5 (#48469)

peterjc123 committed 3 years ago

5b6b1495

[ROCm] update debug flags (#46717)

jeffdaily committed 3 years ago

7f869dca

Fix multiple spelling and grammar mistakes (#48592)

ProGamerGov committed 3 years ago

d6ddd78e

[CUDA graphs] Make CUDAGeneratorImpl capturable (#47989)

mcarilli committed 3 years ago

2200e722

[caffe2] Register BlackBoxPredictor AllocationArenaPool as CPUCachingAllocator (#48161)

Hao Lu committed 3 years ago

4976208e

[dper] supress excessive msg (#48404)

xw285cornell committed 3 years ago

d386d332

Fix bazel build after sleef update (#48614)

malfet committed 3 years ago

d74f2d28

Tweak Vulkan memory use. (#47728)

Ashkan Aliabadi committed 3 years ago

66440d1b

[vmap] implement batching rules for clamp, clamp_min and clamp_max (#48449)

RockingJavaBean committed 3 years ago

8f8738ce

migrate export_caffe2_op_to_c10.h macros to the new dispatcher registration API, update code_analyzer regex (#48308)

bdhirsh committed 3 years ago

b5149513

Fix test comparison ops check for scalar overflow (#48597)

malfet committed 3 years ago

032e4f81

Fix the typo errror in the line #953 of the docs of 'torch/nn/modules/activation.py' (#48577)

Siyeong-Lee committed 3 years ago

b84d9b48

Back out "[pytorch][PR] [JIT] Add `__prepare_scriptable__` duck typing to allow replacing nn.modules with scriptable preparations"

suo committed 3 years ago

eba96b91

added docs to nn.rst (#48374)

AishwaryaKalloli committed 3 years ago

fe806382

Add documentation for torch.overrides submodule. (#48170)

hameerabbasi committed 3 years ago

4e15877d

Improve libuv detection on Windows (#48571)

gunandrose4u committed 3 years ago

42e7cdc5

Newer Older