strisunshinewentingwang/pytorch

Pull Requests Commits

xlogy: Port to structured (#60814)

Freey0 committed 3 years ago

11cc1793

Makes a streaming backward test try gradient stealing more directly (#60065)

mcarilli committed 3 years ago

9fb6b40f

Support 3 argument variant of the getattr() call where the third arg is the default return value (#61599)

Jerry Cai committed 3 years ago

873cc7a4

[CUDA graphs] Make sure graph mempool cudaMalloc_count decrement pairs with cudaFree for all allocations (#61567)

mcarilli committed 3 years ago

ffd2e602

Port other comparison ops: `ne`, `lt`, `gt`, `le`, `ge` to structured kernels. (#60942)

ysiraichi committed 3 years ago

208d06ca

Port `eq` kernel to structured kernels. (#60177)

ysiraichi committed 3 years ago

97327137

[vulkan] Adaptive local work group size (#61170)

SS-JIA committed 3 years ago

64ac4288

[vulkan] Calculate a 4x4 output tile for each invocation in conv2d_pw (#60760)

SS-JIA committed 3 years ago

f324421d

[vulkan] Convolution op cleanup (#60759)

SS-JIA committed 3 years ago

a1b5025e

[vulkan] Reduce submission rate to save CPU cycles (#60758)

SS-JIA committed 3 years ago

cacab7e9

[package] merge test_torchscript into test_package_script (#61807)

suo committed 3 years ago

554038c2

ban PyTorchStreamWriter from writing the same file twice (#61805)

suo committed 3 years ago

f02cfcc8

[package] fix storage serialization collision (#61806)

suo committed 3 years ago

04043d68

add BFloat16 support for topk on CPU (#59547)

jiayisunx committed 3 years ago

c30048fc

ignore and clear not ready errors (#61554)

jeffdaily committed 3 years ago

15210f3b

Regenerate libtorch workflow files that got lost in merge conflict (#61872)

janeyx99 committed 3 years ago

e68c0168

Fix grammatical errors on the PyTorch Contribution Guide (#61818)

ndkshr committed 3 years ago

0a6d8824

Port `signbit` to structured kernel (#57936)

Kiyosora committed 3 years ago

43c5dc40

Remote whitespace introduced by #61438 (#61863)

Dmytro Dzhulgakov committed 3 years ago

44d32671

Exclude wrapper tensors from functorch in the native::resize_output fastpath (#61846)

zou3519 committed 3 years ago

26d17ddc

Remove unnecessary Ubuntu version checks (#61738)

xuhdev committed 3 years ago

f9128897

Always use fast gradcheck for LayerNorm 3d_no_affine_large_feature (#61848)

soulitzer committed 3 years ago

1b0a7f38

[BE] Include a unit test for Save Operator with db_options

Kaige Liu committed 3 years ago

094abf5f

Upgrade CPUFallback for loops (#61722)

r-barnes committed 3 years ago

e389650f

[DDP] Add API to get model parameters in hook (#61637)

rohan-varma committed 3 years ago

04bd9d75

Update progress and error reporting in clang-tidy (#61672)

1ntEgr8 committed 3 years ago

66c8d21d

ENH Adds tests and docs for 2d & 3d modules that already support no batch (#61262)

thomasjpfan committed 3 years ago

24a6eb3f

enable check trace when tracing a mkldnn model (#61241)

XiaobingSuper committed 3 years ago

4f46943e

fmin has been ported to the structured kernel, removing the old implementation (#60810)

Freey0 committed 3 years ago

75b68def

fmod: Port to structured (#60809)

Freey0 committed 3 years ago

b526080d

Newer Older