strisunshinewentingwang/pytorch

Pull Requests Commits

irange-ify 7 (#62117)

r-barnes committed 3 years ago

b5867a1b

Revert D29928698: [pytorch][PR] Use private squid proxy

janeyx99 committed 3 years ago

59bb4f2d

Port `slow_conv_transpose2d` to structured (#55503)

Meghan Lele committed 3 years ago

3a2603bc

[pytorch] Bring back RemoveInplaceOps() (#62200)

Meghan Lele committed 3 years ago

05b802d4

[Static Runtime] Fixed another build failure in OSS due to test_utils.h (#62338)

navahgar committed 3 years ago

b91a9176

ENH Adds no_batch_dim support for pad 2d and 3d (#62183)

thomasjpfan committed 3 years ago

7c588d5d

Use private squid proxy (#62244)

zhouzhuojie committed 3 years ago

6da4a255

[Model Averaging] Create a base class for model averaging (#62111)

Yi Wang committed 3 years ago

2581dfc2

Revert D29794666: Remove faulty process group code

H-Huang committed 3 years ago

a15fff0a

ENH Adds no_batch_dim tests/docs for Maxpool1d & MaxUnpool1d (#62206)

thomasjpfan committed 3 years ago

71a6ef17

[quant][graphmode][fx] Add reference pattern support for BatchNorm (#62215)

jerryzh168 committed 3 years ago

cdf85a82

optimize non lastdim softmax bf16 (#60371)

leslie-fang-intel committed 3 years ago

7443c90f

[static runtime] Implement aten::full (#62227)

d1jang committed 3 years ago

68efa186

[DDP] Run test_ddp_new_tensor_in_fwd with static graph (#61992)

rohan-varma committed 3 years ago

10c6811a

These should be equivalent per the previous formula but breaks xla (#62329)

albanD committed 3 years ago

acf8907e

[bc-breaking] reference option for conv produce a pattern instead of reference conv module (#61942)

jerryzh168 committed 3 years ago

f4baa83e

Teach pytrees about namedtuple (#62292)

zou3519 committed 3 years ago

52d1ffb7

Build M1 binaries with PocketFFT (#62222)

malfet committed 3 years ago

c06b6e44

Revert D29816592: [pytorch][PR] [fix] polygamma n>=1

malfet committed 3 years ago

cb2b5f06

[8/N] Nnapi backend delegation preprocess: New refactored design (#62225)

Amyh11325 committed 3 years ago

73f1e2d1

Update nccl to v2.10.3-1 (#62276)

malfet committed 3 years ago

7aabda6d

Revert D29943356: .github: Migrate ecr_gc to github actions

malfet committed 3 years ago

1f1d01df

[dist_optim] fix the bug of none grads on functional optimizers (#62249)

wanchaol committed 3 years ago

af0f083d

Do not use deprecated data accessor in IndexKernel.cu (#62268)

malfet committed 3 years ago

c0b80669

[PyTorch] Add KWargs support to script module forward (#62224)

m3rlin45 committed 3 years ago

e3be1850

Migrate thnn_conv_depthwise2d from THC to ATen (#62281)

peterbell10 committed 3 years ago

9776e1ff

Fix forward ad for matrix power land race (#62291)

albanD committed 3 years ago

ba9423aa

Rework PowKernel.cu (#62260)

peterbell10 committed 3 years ago

171e13fd

[reland][bc-breaking] reference option for linear produce a pattern instead of reference linear module (#61892) (#62277)

jerryzh168 committed 3 years ago

7507aede

Limit smoke tests on PRs to just one config (#62288)

janeyx99 committed 3 years ago

24d94f51

Newer Older