strisunshinewentingwang/pytorch

Pull Requests Commits

Add rpc.functions.async_execution decorator for rpc_sync/rpc_async (#39216)

mrshenli committed 4 years ago

a05ef17e

[ONNX] Bump up ONNX submodule to a82c6a7010e2e332d8f74ad5b0c726fd47c85376 (#39372)

houseroad committed 4 years ago

15ad9dd3

[resubmit] Use TensorMethods.cpp (#39385)

Ilia Cherniavskii committed 4 years ago

abe2be20

Fix for num_threads==1 in OpenMP "parallel for" (#36479)

mattip committed 4 years ago

a952f9bb

[TensorExpr] eliminate zero length Allocations in IRSimplifier (#38794)

nickgg committed 4 years ago

36607c85

[JIT] Kill _cast_* operators (#39348)

James Reed committed 4 years ago

f166b934

call DoRunWitType on Layernorm (#39409)

Hector Yuen committed 4 years ago

8638df45

[JIT] Make torch.unique_consecutive compatible (#39339)

zasdfgbnm committed 4 years ago

ebd4125e

nit on functional autograd (#35493)

albanD committed 4 years ago

c6720f0d

Also set CMAKE_C_STANDARD for MSVC (#39304)

xuhdev committed 4 years ago

89c0efb3

Updated assert to remove check on 3rd dim for MHA (#39402)

Nayef211 committed 4 years ago

71af538e

Make `_C` extension a thin C wrapper (#39375)

malfet committed 4 years ago

a864dbb3

support flip and rot90 for complex dtype (#37826)

kshitij12345 committed 4 years ago

09bea139

Replace calls to contiguous with contiguous(suggested memory format) (#38433)

kimishpatel committed 4 years ago

58cb369d

Kill THC_logical{Value, Tensor} (#39069)

ShawnZhong committed 4 years ago

0d96f264

Fix test_get_and_set_timeout for TensorPipe Agent (#39353)

osalpekar committed 4 years ago

a6f0051d

[caffe2] fix test error in video_input_op_test (#39382)

jspark1105 committed 4 years ago

fca928ca

[caffe2] format video_input_op_test.py (#39381)

jspark1105 committed 4 years ago

04ac41fe

Add rocm image to circleci docker builder (#39262)

jeffdaily committed 4 years ago

c3ddb3f7

Add prim::data to lite interpreter (#39335)

iseeyuan committed 4 years ago

8bc5a493

[Onnxifi] Support quantized output in Onnxifi (#39230)

Yinghai Lu committed 4 years ago

cca29f29

Fix some bugs of argmin/argmax and min/max (#39212)

zasdfgbnm committed 4 years ago

35719cdc

Adding lost extra_repr() and __setstate __() to activation.py (#39084)

Roffild committed 4 years ago

11f1014c

Return `None` instead `False`, and return `bool` to `None` in type stub (#39324)

KKawamura1 committed 4 years ago

b5cd3a80

Expose torch.futures.Future (#39008)

mrshenli committed 4 years ago

bb0377bb

Initial support for building on Ampere GPU, CUDA 11, cuDNN 8 (#39277)

zasdfgbnm committed 4 years ago

b3fac8af

[WIP] Layernorm Fake FP16 Op. (#39103)

venkatacrc committed 4 years ago

85b3fa03

More fixes about using Windows API through ctypes (#39376)

peterjc123 committed 4 years ago

30146d73

[TensorPipe] Acquire lock when adding message to timeout map (#39398)

lw committed 4 years ago

e358adb4

[TensorPipe] Guess IP addr in separate function (#39397)

lw committed 4 years ago

e142d703

Newer Older