strisunshinewentingwang/pytorch

Pull Requests Commits

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

peterbell10 committed 1 year ago

66e39895

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

peterbell10 committed 1 year ago

b256091c

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

peterbell10 committed 1 year ago

ef512db0

update gloo submodule (#101472)

xunnanxu committed 1 year ago

df6acf27

[CI] Distribute bot workload (#101723)

malfet committed 1 year ago

8c0b1489

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

yanboliang committed 1 year ago

29de5817

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

pytorchmergebot committed 1 year ago

5f07c589

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

janeyx99 committed 1 year ago

3135bec4

Revert "match sdpa patterns from HF (#100609)"

pytorchmergebot committed 1 year ago

5d3cfda1

Update public_api to remove duplicated `randn_like` (#101302)

liang-hou committed 1 year ago

b429a4de

[Security] Move mergebot workflows in its own env (#101718)

malfet committed 1 year ago

Verified 2236d5ef

Check for pytest extensions in run_test (#100916)

clee2000 committed 1 year ago

f3fc531e

Run dynamo tests in parallel (#101432)

clee2000 committed 1 year ago

e3c9a1e5

remove default lower bound in dynamic_dim suggestions (#101636)

avikchaudhuri committed 1 year ago

e3c66ded

Run `test_cpp_memory_snapshot_pickle` only when linux and x86_64 (#101366)

crcrpar committed 1 year ago

c8579b73

Revert "[opinfo] empty_strided (#100890)"

pytorchmergebot committed 1 year ago

dfac4364

match sdpa patterns from HF (#100609)

Natalia Gimelshein committed 1 year ago

f33725b8

[quant][pt2] Handle maxpool + conv + bn case in prepare QAT (#100941)

andrewor14 committed 1 year ago

8e51521c

Adding serialization ID to inline container (#100994)

atannous committed 1 year ago

3ed1569e

Support map autograd and pytree in/out. (#101633)

ydwu4 committed 1 year ago

326a4cc8

Handle multi-user case in split-cat simplification (#101473)

devashishshankar committed 1 year ago

38e537db

Fix determenistic typos (#101631)

kit1980 committed 1 year ago

e17d9f2c

[PT2][Quant] Move to module partitioner for linear pattern quantization (#101122)

kimishpatel committed 1 year ago

07e759ec

[transformer benchmark] sort by cuda time (#101349)

jspark1105 committed 1 year ago

ebae77e8

Fix benchmark model names printouts with tqdm (#101627)

jansel committed 1 year ago

403ce1a1

[PT] Update module partitioner to return parameter node (#101121)

kimishpatel committed 1 year ago

bec655f8

inductor(CPU): fix issue when padding/stride/dilation size is one for cpu weight packing pass(reland) (#101353)

XiaobingSuper committed 1 year ago

75375b41

[PT2][Quant] Remove None annotations (#101120)

kimishpatel committed 1 year ago

2c807a4a

[functorch] fix UB in interpreter stack (#101568)

zou3519 committed 1 year ago

783a46ad

s390x SIMD: Propagate NaN in minimum and maximum operations (#99716)

AlekseiNikiforovIBM committed 1 year ago

cb6fa890

Newer Older