strisunshinewentingwang/pytorch

Pull Requests Commits

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

fix multinomial for empty batch (#39873)

ngimel committed 4 years ago

f59e3897

Move pip install after setting up VS environment (#39898)

malfet committed 4 years ago

8c8d9f89

[TensorExpr] Eliminate Cond statements when each branch is a different kind of empty (#39754)

nickgg committed 4 years ago

63dc1363

[vulkan] VulkanTensor, add strides in interface (#39077)

IvanKobzarev committed 4 years ago

36501ff5

Move all torch.nn.modules type annotations inline (#38211)

ezyang committed 4 years ago

eace0533

Migrate pow kernel to c10::complex (#39286)

zasdfgbnm committed 4 years ago

e22dd561

Newer Older