strisunshinewentingwang/pytorch

Pull Requests Commits

change traceback

angelayi committed 2 years ago

096465c5

angelayi committed 2 years ago

65859922

angelayi committed 2 years ago

a9fec23b

angelayi committed 2 years ago

121958bf

angelayi committed 2 years ago

4a9cfef5

tracing context

angelayi committed 2 years ago

b96b45ec

angelayi committed 2 years ago

81afc354

Upstream xformers code (#100583)

drisspg committed 2 years ago

c9ba967c

adding moco to CI (#101098)

anijain2305 committed 2 years ago

794cc395

[CI] Enlarge memory for OOM models in inductor cpu HF accuracy test (#101395)

chuanqi129 committed 2 years ago

b315c9b5

Add aten.searchsorted.Tensor meta kernel (#101637)

angelayi committed 2 years ago

72a73ef6

fix missing-prototypes warnings in torch_cpu (Part 4) (#100849)

cyyever committed 2 years ago

c2f28d1c

inductor: skip weight packing when has zero shape (#101355)

XiaobingSuper committed 2 years ago

900ca4df

Fix link error on s390x (#101000)

kiszk committed 2 years ago

e48a052e

Add activation functions (ReLU and SiLU for now) for structured sparse linear operator (#101339)

alexsamardzic committed 2 years ago

bfb3941a

[inductor] send max_pool2d_with_indices and its backwand to fallback if dilation is not 1 (#100531)

cdzhan committed 2 years ago

a0e6f820

[dynamo][moco] Save global torch state to restore on graph break (#101201)

anijain2305 committed 2 years ago

dafa009c

[DataLoader] Adding `StackDataset` (#101338)

stsouko committed 2 years ago

28098cae

Fix check-labels workflow commenting on forked PRs (#101467)

kit1980 committed 2 years ago

f0f0f709

match sdpa patterns from HF (#100609)

ngimel committed 2 years ago

c7392347

Make HUD link https (#101461)

pandaninjas committed 2 years ago

18f6f30d

[BE] Fix rule not found error message (#101745)

malfet committed 2 years ago

124d812f

[inductor/decomp] Add aten._unsafe_index to disable range checks (#101602)

peterbell10 committed 2 years ago

66e39895

[inductor] Generate indirect_indexing checks even if optimized out (#100895)

peterbell10 committed 2 years ago

b256091c

[inductor] Constant and index_expr propagation pass (#101077)

peterbell10 committed 2 years ago

ef512db0

update gloo submodule (#101472)

xunnanxu committed 2 years ago

df6acf27

[CI] Distribute bot workload (#101723)

malfet committed 2 years ago

8c0b1489

[Dynamo] Graph break on torch.cuda.set_device() (#101668)

yanboliang committed 2 years ago

29de5817

Revert "[inductor] Refactor RNG operators (#100064)"

pytorchmergebot committed 2 years ago

5f07c589

[docs] Clarify when to use SparseAdam (#101465)

janeyx99 committed 2 years ago

3135bec4

Older