strisunshinewentingwang/pytorch

Pull Requests Commits

Implement adding bias vector into structured sparse linear operator

alexsamardzic committed 2 years ago

e8f916bf

sampled_addmm: BSR support (#101163)

nikitaved committed 2 years ago

6c7410dd

inductor: align cpp floordiv with python floordiv for dyanmic shape path (#102068)

XiaobingSuper committed 2 years ago

4882cd08

[fx][2/n] Add metadata to placeholders (#102195)

s4ayub committed 2 years ago

a896962f

[c10d] add fake pg necessary collectives (#102238)

wanchaol committed 2 years ago

7b47cd0a

[c10d] conslidate barrier after init logic (#102237)

wanchaol committed 2 years ago

9a192625

Profiling doc (#101895)

eellison committed 2 years ago

aa83a527

Support resize on meta storage (#101988)

ezyang committed 2 years ago

818d92f5

Location-shift MKL Exponential Distribution (#101720)

min-jean-cho committed 2 years ago

3ca068bc

[TP] Add API logging for TP high level API (#102209)

fduwjj committed 2 years ago

d4380edb

do not raise when constraint locals are not in signature (#102198)

avikchaudhuri committed 2 years ago

d4f711b0

Add meta registrations for some foreach ops (#102225)

mlazos committed 2 years ago

69c7f710

[vision hash update] update the pinned vision hash (#102230)

pytorchupdatebot committed 2 years ago

2f08f9a6

Support unary not on lists (#102210)

mlazos committed 2 years ago

2434a205

[pytorch] add Vulkan support for the `aten::cat` operator for 1d, 2d, 3d and 4d (#102128)

copyrightly committed 2 years ago

a0e44284

Full default dict support in dynamo (#102202)

mlazos committed 2 years ago

23dbdd90

[FSDP] Start to generalize modules to ignore for mixed precision (#102010)

rohan-varma committed 2 years ago

f3e42f15

[partitioner] fix for rng ops (#102123)

anijain2305 committed 2 years ago

c2093de5

update thresholds for various ops in functorch/test_ops.py (#102016)

Fuzzkatt committed 2 years ago

2763b508

[MPS] Restride output strides to contiguous format for inverse op (#102122)

DenisVieriu97 committed 2 years ago

e274c2e4

Replace require_backend with require_backend_is_available (#101891)

H-Huang committed 2 years ago

11d1cd89

Fix redudant kernel generations (#102104)

FindHao committed 2 years ago

3e08988c

[pytorch] add Vulkan support for the `t` and `transpose` operators for 2d, 3d and 4d tensors (#101808)

copyrightly committed 2 years ago

9ce95ce1

Add fake process group (#102180)

ezyang committed 2 years ago

c903b12c

add additional stream priority for cuda streams (#101956)

Natalia Gimelshein committed 2 years ago

5da497ca

update tf32 thresholds in nn/test_convolution.py (#102015)

Fuzzkatt committed 2 years ago

f8896b7b

No need to run non-CUDA jobs in memory leak check mode (#102188)

huydhn committed 2 years ago

dedcf8f7

[inductor][decomp] Add aten._unsafe_index_put for unchecked indexing (#101812)

peterbell10 committed 2 years ago

ce420107

Populate all args with fake tensor value (#102129)

tugsbayasgalan committed 2 years ago

dbf6912b

Revert "Support resize on meta storage (#101988)"

pytorchmergebot committed 2 years ago

210fc28d

Older