strisunshinewentingwang/pytorch

Pull Requests Commits

[PyTorch] MHA: simplify gemm_nt (#72460)

swolchok committed 2 years ago

49611a33

[PyTorch] MHA: fix dim_per_head / V bug (#72459)

swolchok committed 2 years ago

f5d07808

[PyTorch] Fix ASAN errors in CPU MHA impl (#72458)

swolchok committed 2 years ago

d707114a

[PyTorch] MHA: add debug shape checks (#72457)

swolchok committed 2 years ago

f2f4847e

[PyTorch] Use DimVector in at::matmul (#72230)

swolchok committed 2 years ago

6f29313e

Nvfuser code bump 2_1_2022 (#72127)

jjsjann123 committed 2 years ago

2d110d51

[Pytorch Edge] Minor improve documentation in test_backend_with_compiler

JacobSzwejbka committed 2 years ago

52c516ec

DOC: release documentation version should be major.minor (#72706)

mattip committed 2 years ago

fb4504da

Revert D34034848: free up dispatch key space (in C++)

bdhirsh committed 2 years ago

22ccf448

Revert D34034847: DispatchKeySet perf improvements

bdhirsh committed 2 years ago

7f560fb3

Revert D34214953: Add new tls snapshot feature

bdhirsh committed 2 years ago

f1a9650e

Fix ConvTranspose2D dilation type annotaion (#72789)

Rikorose committed 2 years ago

5bb851dc

MAINT, DOC: Trivial spellings and warnings (#72745)

HaoZeke committed 2 years ago

801abc0c

ci: Update macOS binary workflows with new templates (#72727)

seemethere committed 2 years ago

831cb4b9

Add new tls snapshot feature (#72623)

albanD committed 2 years ago

6199b523

[test_fx_const_fold] Remove dependencies on acc_* (#72810)

jfix71 committed 2 years ago

454e2ec7

Operator developer guidance (#72470)

cccclai committed 2 years ago

8e8c15cf

Add wrapped Tensor autograd test (#72622)

albanD committed 2 years ago

584f1396

Clean up LoggingTensor semantic (#72620)

albanD committed 2 years ago

3c33f0bd

add optional encoding argument to fileopener (#72715)

erip committed 2 years ago

78e481d0

avoiding adding some functions to the public python API before 1.11 release (#72543)

bdhirsh committed 2 years ago

f87f753b

[ROCm] Enable sort operator BF16 support (#71226)

pruthvistony committed 2 years ago

aa44480b

[Quant][core][devs] Separated implementations for quantized & non-quantized tensors in index_select_cuda (#72407)

dzdang committed 2 years ago

9981aade

Properly initialize `grad_weight` in `raw_cudnn_convolution_backward_weight_out` (#72157)

eqy committed 2 years ago

e7985e3c

Fix type annotation for `torch.backends.cudnn.allow_tf32` (#72757)

lkct committed 2 years ago

7d542a4f

[ROCM] Navi21 Enablement 2: Depthwise kernels (#72682)

micmelesse committed 2 years ago

89c934f4

DispatchKeySet perf improvements (#72403)

bdhirsh committed 2 years ago

8aa3620d

free up dispatch key space (in C++) (#72402)

bdhirsh committed 2 years ago

66902560

Implement Tanh Gelu Approximation (#61439)

rdspring1 committed 2 years ago

4f8b986e

Mobile upgrader: add GELU test modules

jbschlosser committed 2 years ago

64faa043

Newer Older