strisunshinewentingwang/pytorch

Pull Requests Commits

Fix failing torch deploy tests and reenable.

ezyang committed 3 years ago

222fb8d6

.github: Add windows.8xlarge.nvidia.gpu (#58781)

seemethere committed 3 years ago

a083933d

.circleci: Disable windows CPU builds for CircleCI (#58855)

seemethere committed 3 years ago

8ae4d07d

fixing csr addmm bug (#58768)

aocsa committed 3 years ago

1fca1545

Move cublas dependency after CuDNN (#58287)

malfet committed 3 years ago

2dda8d75

.github: Enable Windows workflow for pull_request (#58418)

seemethere committed 3 years ago

bb477046

Adding a new include directory in BLIS search path (#58166)

austinpagan committed 3 years ago

007fe949

[DataLoader] Fix bugs for typing (#58450)

ejguan committed 3 years ago

0e160870

Automated submodule update: FBGEMM (#58161)

facebook-github-bot committed 3 years ago

5c7dace3

add deterministic path for scatter_add_cuda for 1D tensors (#58761)

yuguo68 committed 3 years ago

74c12da4

[deploy] temporarily disable deploy tests (#58832)

suo committed 3 years ago

50ded095

Port `kthvalue` tests to `OpInfo` (#58654)

ysiraichi committed 3 years ago

a7fdd487

Add GenericShardingSpec for generic tensor sharding. (#57409)

pritamdamania committed 3 years ago

4709fdb1

Introduce ChunkShardingSpec as a model sharding specification. (#55728)

pritamdamania committed 3 years ago

0d6fa1ad

Enabled BFloat16 support for cumsum, logcumsumexp, cumprod, cummin & cummax on CUDA (#57904)

imaginary-person committed 3 years ago

c5a1f043

OpInfo: split, split_with_sizes (#58184)

kshitij12345 committed 3 years ago

ee3ea31f

[ROCm] disable test test_Conv2d_groups_nobias_v2 for ROCm (#58701)

KyleCZH committed 3 years ago

52a8031e

Change list striding kernel implementation to handle optional integers (#58536)

tugsbayasgalan committed 3 years ago

fa0b89bb

[Gradient Compression] Disable test_ddp_hook_parity_powerSGD on Gloo backend (#58802)

Yi Wang committed 3 years ago

28840b9a

[torch][repeat_interleave] remove stream syncronization if output size is given (#58417)

Serhat Yilmaz committed 3 years ago

4ca4640b

port mm to structure kernel (#57755)

Rong Rong (AI Infra) committed 3 years ago

c1c9be16

OpInfo: clone, contiguous (#58390)

kshitij12345 committed 3 years ago

f9e8dc00

adding test_sparse_csr to run_test (#58666)

Rong Rong (AI Infra) committed 3 years ago

a7002046

[PyTorch] Remove device check from a few indexing methods (#58800)

wenleix committed 3 years ago

22776f08

[Pytorch Delegated Backend] Add python binding for (#57156)

kimishpatel committed 3 years ago

796c97a8

[Pytorch Backend delegation] Add api for backend lowering to query debug (#55462)

kimishpatel committed 3 years ago

d6d726f7

Revert D28617214: [Gradient Compression] Do not skip the comm hook tests on Gloo backend

Rong Rong (AI Infra) committed 3 years ago

e7c35a33

Separated out working tests from not working tests for NNC OpInfo (#58788)

Chillee committed 3 years ago

60931611

[PyTorch Edge] Check if open paren ( occurs in an operator name string (#58687)

dhruvbird committed 3 years ago

dc8bc6ba

Revert D28474878: Always use intrusive_ptr for Message (1 out of 2)

xw285cornell committed 3 years ago

4c961bea

Older