strisunshinewentingwang/pytorch

Pull Requests Commits

skip cuda test_cholesky_solve_batched_many_batches due to illegal memory access (#48999)

xwang233 committed 3 years ago

a849f382

[ROCm] restore autograd tests (#48431)

jeffdaily committed 3 years ago

e8b00023

Add high level profiling trace for dataloading and optimizer (#47655)

gaoteng-git committed 3 years ago

1c31f762

[quant][graphmode][fx] Add test for ResnetBase (#48939)

jerryzh168 committed 3 years ago

2d9585a6

[pt][quant] Remove contiguous calls in qembeddingbag (#48993)

Hao Lu committed 3 years ago

59a3e766

Annotate torch._tensor_str (#48584)

guilhermeleobas committed 3 years ago

7c0a3e3a

Torch onnx (#48980)

guilhermeleobas committed 3 years ago

34cc77a8

Correctly apply WIN32_LEAN_AND_MEAN to the whole repo (#49025)

peterjc123 committed 3 years ago

5450614c

[quant][fix] Support quantization of ops where input is quantizable (#49027)

supriyar committed 3 years ago

4434c07a

[quant][graphmode][fx] Add MatchAllNode in pattern matching (#48979)

jerryzh168 committed 3 years ago

993ce4b2

Add a pass to fetch attributes of nn.Module to fx.node (#47935)

842974287 committed 3 years ago

107c31f2

[JIT] Allow del statements with multiple targets (#48876)

Meghan Lele committed 3 years ago

3f9ff48e

fx quant: move more functions to utils (#48908)

vkuzo committed 3 years ago

d033e185

fx quant: move qconfig utils to utils file (#48907)

vkuzo committed 3 years ago

2668ea80

fx quant: quick cleanup for model_device (#48906)

vkuzo committed 3 years ago

17e71509

[collect_env] Add candidate paths for nvidia-smi on Windows (#49021)

peterjc123 committed 3 years ago

e538bd66

[CUDAExtension] support all visible cards when building a cudaextension (#48891)

stas00 committed 3 years ago

02b63858

add a unit test for large node error (#48938)

scottxu0730 committed 3 years ago

60004814

CUDA BFloat16 batchnorm (non-cuDNN) (#44994)

zasdfgbnm committed 3 years ago

59605811

[StaticRuntime] Add aten::narrow (#48991)

Hao Lu committed 3 years ago

e8ec8486

Put Flake8 requirements into their own file (#49032)

samestep committed 3 years ago

d1fb4b4f

[TensorExpr] Enable inlining for output tensors too. (#48967)

Mikhail Zolotukhin committed 3 years ago

2b70bcd0

Delete ATen mirror stuff (#49028)

samestep committed 3 years ago

0fb9d366

Add LKJCholesky distribution (#48798)

neerajprad committed 3 years ago

dee82ef3

[FX][2/2] Make docstrings pretty when rendered (#48871)

James Reed committed 3 years ago

c92c8598

Add fftw3 cmake as alternative for FFT/DFT (#48808)

Rong Rong committed 3 years ago

b89c3284

Avoid initializing gradInput twice in the backward phase of replication (#48890)

Jokeren committed 3 years ago

b0e919cf

[static runtime] Add Internal Ops to the registry (#48616)

bwasti committed 3 years ago

274ce26f

[BE] Fix signed-unsigned warnings (#48848)

malfet committed 3 years ago

ad3fed8b

Modify NEON check for ARM64 on OS X (#48982)

malfet committed 3 years ago

c29f5164

Newer Older