strisunshinewentingwang/pytorch

Pull Requests Commits

[futures] Add torch.futures.collect_all()/wait_all() python api. (#39790)

jjlilley committed 4 years ago

a9aa6367

[pytorch] fix (ProfiledType|TraceType)None.cpp (#39934)

ljk53 committed 4 years ago

0d19ae5a

re-enable some corner cases in memory format transpose test (#39891)

lixinyu committed 4 years ago

fdf6d378

Int8 PTQ ops for online training (#39818)

csummersea committed 4 years ago

558c20f5

[quant][graphmode][refactor] isScalar check (#39892)

jerryzh168 committed 4 years ago

99084104

Correct #39759 for HIP. (#39801)

jeffdaily committed 4 years ago

1e05e5e0

Add file_name argument to load_state_dict_from_url (#39749)

MohamedAliRashad committed 4 years ago

f3f9415f

add optional request headers to torch.hub (#39740)

pmeier committed 4 years ago

baa60481

[CI][vulkan] android build abi x86 with USE_VULKAN (#39912)

IvanKobzarev committed 4 years ago

d367f575

Extend int8 quantize op to take scale and zero point from input

Haixin Liu committed 4 years ago

2bab9149

pin ninja version to fix windows CI (#39944)

albanD committed 4 years ago

48678aa3

Replace __host__ __device__ with C10_HOST_DEVICE in THCIntegerDivider.cuh (#39797)

pbelevich committed 4 years ago

4574abc3

Add experimental deterministic flag (#38683)

kurtamohler committed 4 years ago

124cdf22

[jit][subgraph_rewriter] Support list of filters (#39867)

jerryzh168 committed 4 years ago

004aa089

Remove LegacyComplex.h (#39834)

zasdfgbnm committed 4 years ago

38768892

[TensorPipe] Add extensive logging (#39781)

lw committed 4 years ago

ae6a68ad

Revert D22011184: [pytorch][PR] Fix CUDA device guard usage when first arg of kernel is scalar

albanD committed 4 years ago

52cc0c2c

[quant][graphmode] Quantizing traced modules (#39826)

jerryzh168 committed 4 years ago

246d7bb4

Add CHECK-SOURCE-HIGHLIGHTED to file check utils. (#39692)

Yanan Cao committed 4 years ago

c0682333

[vulkan] Conv2d with optional clamp (#39115)

IvanKobzarev committed 4 years ago

0526af1a

[vulkan] addmm support non-vulkan inputs (#39078)

IvanKobzarev committed 4 years ago

71372b45

[nvFuser] Transform replay refactor and minor updates (#39579)

csarofeen committed 4 years ago

80e5ebf9

[JIT] Allow self-referential annotations in classes (#39821)

Meghan Lele committed 4 years ago

2854811a

[quant][graphmode] Remove dedup pass (#39825)

jerryzh168 committed 4 years ago

14e841c2

Fix CUDA device guard usage when first arg of kernel is scalar (#39870)

kurtamohler committed 4 years ago

2cd27be5

Vectorize on output for reduction kernels (#37206)

zasdfgbnm committed 4 years ago

b10c53e9

Typo in Dispatch.h (#39882)

zasdfgbnm committed 4 years ago

a92231b7

move basic math ops to lite interpreter (#39861)

linbinyu committed 4 years ago

bbf364b0

[JIT] use python type resolver for all types (#39880)

Elias Ellison committed 4 years ago

bdecedd2

Delete redundant device/dtype in TensorIterator add_input/add_output (#39798)

ezyang committed 4 years ago

0aa70039

Newer Older