strisunshinewentingwang/pytorch

Pull Requests Commits

CMake/Ninja: fix dependencies for .cu files (#36938)

Baranowski committed 4 years ago

db84689c

make merge_fp32_into_fp16_inputs to generate ops for each partition (#36973)

Bangsheng Tang committed 4 years ago

246b208e

Documentation LU Decomposition: deriving L, U, and P (#36907)

twoertwein committed 4 years ago

be52b7f0

[pytorch] add test for empty tensor support in nn.Linear (#36983)

ppwwyyxx committed 4 years ago

ff435a0e

Revert D20961463: [TEST] add ops for portal TTS model

Mike Ruberry committed 4 years ago

0f3af852

Do not use VLAs in vec256_qint.h (#36855)

malfet committed 4 years ago

98c293c1

[TEST] add ops for portal TTS model (#36971)

linbinyu committed 4 years ago

68f6b987

Clean-up non-AVX variant of `bitwise_binary_op` template (#36966)

malfet committed 4 years ago

3d2d5c82

fmadd in vec256_base should be on Vec256<T>, not T (#36751)

xuhdev committed 4 years ago

a1eb591e

Enable test_hardsigmoid_grad_xla on pytorch side (#36967)

JackCaoG committed 4 years ago

cdc1ca04

[ROCm] Enable wrongly skipped tests on CPU on ROCm (#36968)

rohithkrn committed 4 years ago

742d9796

Ensure linearIndex of advanced indexing backwards is contiguous. (#36959)

gchanan committed 4 years ago

ce0500eb

Update NNPI backend to 0.5.1.8 (#4397)

omromano committed 4 years ago

59f923e8

[jit] Adding vectorized load/store support for JIT generated CUDA kernel (#36555)

FDecaYed committed 4 years ago

346215ca

Add RecordFunctionGuard (#36215)

Ilia Cherniavskii committed 4 years ago

3ae70cb8

Link NCCL lib to TORCH_PYTHON_LINK_LIBRARIES when USE_NCCL=1 (#36948)

Shen Li committed 4 years ago

a14a8376

Fix flaky test_barrier_timeout* tests for test_distributed. (#36963)

pritamdamania committed 4 years ago

32307efd

Add sync-point insertions and block/thread local memory allocations (#36563)

zheng-xq committed 4 years ago

5e504e83

Delete docker build job for pytorch-linux-bionic-clang9-thrift-llvmdev (#36930)

Ailing Zhang committed 4 years ago

0647f344

[quant][graph] Add quantizedmul_relu and quantized::mul_scalar_relu ops (#36844)

supriyar committed 4 years ago

c03d1494

[quant][graph] Support for quantized::mul and quantized::mul_scalar (#36818)

supriyar committed 4 years ago

ee2a9ac5

Add bundled input support to speed_benchmark_torch (#36765)

dreiss committed 4 years ago

1c15cb47

[quant] Minor refactor in fused conv names (#36883)

z-a-f committed 4 years ago

dbdd0f50

add absolute alias for abs (#36597)

jessebrizzi committed 4 years ago

28f439d4

hardsigmoid: add PyTorch wrapper for the QNNPACK path (#36699)

vkuzo committed 4 years ago

4d171c0e

hardsigmoid operator for QNNPACK, using LUTs (#36698)

vkuzo committed 4 years ago

1d720228

Canonicalize includes in aten, and add tests for it (#36301)

zasdfgbnm committed 4 years ago

97f2513c

Convert C casts to static casts (UnaryOpsKernel) (#36400)

kshitij12345 committed 4 years ago

47023148

[quant][graph] Update quantize_dynamic_script API to take sample model args (#36817)

supriyar committed 4 years ago

a2951a1e

Remove hacked twins from codegen (#36666)

smessmer committed 4 years ago

63e9d95c

Newer Older