strisunshinewentingwang/pytorch

Pull Requests Commits

[CUBLAS][CUDA GRAPHS] (re-re-open of #83461) Explicitly set the workspace for cuBLAS handles (#85447)

eqy committed 1 year ago

b04b2fa9

Revert "Symintified mmm/addmm derivative formulas (#85794)"

pytorchmergebot committed 1 year ago

823dc33b

[ROCm] Retry loop implemented to avoid transient memory leak errors (#82607)

alugorey committed 1 year ago

5709c67f

[NN module] speed up _load_from_state_dict (#85743)

zhengwy888 committed 1 year ago

b2311192

[BE] small typo+lint fixes for einsum/sumproduct_pair (#85709)

janeyx99 committed 1 year ago

cef8dfc8

Symintified mmm/addmm derivative formulas (#85794)

Chillee committed 1 year ago

230edd25

Registered _like metas (#85793)

Chillee committed 1 year ago

a4e75ccf

Added symbolic shape testing for AOTAutograd (#85789)

Chillee committed 1 year ago

35fe4abd

skip test TestCompositeComplianceCUDA::test_forward_ad_nn_functional_max_unpool2d_cuda_float32 (#85767)

jeffdaily committed 1 year ago

0b251d98

[4/N] [Dispatchable Collectives] Update all_reduce_ with CPU / CUDA implementations (#83810)

H-Huang committed 1 year ago

06e0583f

removed compile cache and static argnums (#85783)

Chillee committed 1 year ago

0e256c25

Fix Use obj1.is(obj2) warnings (#85688)

r-barnes committed 1 year ago

614d6f19

Revert "Revert "Symintifying slice ops (#85196)"" (#85746)

ezyang committed 1 year ago

793488cd

Remove some dead code from fake tensor (#85764)

ezyang committed 1 year ago

049be5ac

Make Python reference for permute accept varargs (#85460)

t-vi committed 1 year ago

795028a3

[3/N] [Dispatchable Collectives] Update broadcast_ with CPU and CUDA implementations (#83735)

H-Huang committed 1 year ago

ccac8d13

Rename test file from test_pythonkey to test_aotdispatch (#85769)

Chillee committed 1 year ago

01f946d7

Added floordiv simplification rule needed for models (#85768)

Chillee committed 1 year ago

fc2e7eba

[aarch64] Use the correct binary when cross building //caffe2/torch/csrc/deploy:remove_dt_needed (#85632)

psaab committed 1 year ago

a8be5954

Add environment parse function that supports default value (#85563)

kwen2501 committed 1 year ago

3276b512

[Profiler] tracking Optimizer (part 2 of Record Optimizer) (#84920)

slgong-fb committed 1 year ago

f80ef73d

Update `amax/amin/norm/count_nonzero` signatures with `int[*]? dim` (#83300)

kurtamohler committed 1 year ago

1c0f0b33

add itt unit test and docstrings (#84848)

jingxu10 committed 1 year ago

80b88862

Revert "Update `amax/amin/norm/count_nonzero` signatures with `int[*]? dim` (#83300)"

pytorchmergebot committed 1 year ago

572dd862

[FSDP] Handle the state_dict on CPU cases (#85640)

fegin committed 1 year ago

1c1f3a99

[MPS] Add tensor::index_put op (#85672)

DenisVieriu97 committed 1 year ago

ce4f187f

[quant][docs] Add types for scale and zero_point tensor for torch.fake_quantize_per_channel_affine docs (#85733)

jerryzh168 committed 1 year ago

9858f043

[MPS] Handle 1D weight in linear layer (#85752)

kulinseth committed 1 year ago

7ff6a00a

[Quant][fx] Add quant and scale ranges to BackendConfig (#85200)

andrewor14 committed 1 year ago

4ca125a9

Directly access has_symbolic_sizes_strides, avoid expensive test (#85754)

ezyang committed 1 year ago

24a26814

Newer Older