strisunshinewentingwang/pytorch

Pull Requests Commits

directly init a zero immediate buffer to reduce overhead for batch_norm cpu path (#82558)

XiaobingSuper committed 2 years ago

d08157d5

Make SizesAndStrides able to handle nontrivial objects (#82467)

ezyang committed 2 years ago

3141379f

Delete SymInt::data() in favor of as_int_unchecked() (#82477)

ezyang committed 2 years ago

a9320e6d

Change SymIntNode into an intrusive pointer (#82548)

ezyang committed 2 years ago

50e8abbc

Adding 3.11 nightlies for linux PyPi (#82302)

albanD committed 2 years ago

36120ce5

Fix docker build error related to libc6 (#82476)

atalman committed 2 years ago

c94706c0

[HPU] Enable torch.jit.load for HPU (#81759)

amitchawla1 committed 2 years ago

6592259e

Fix crash on unload torch cpu dll (#67632)

username12 committed 2 years ago

a54c9a41

Fix nondeterminism in torchgen (#82536)

peterbell10 committed 2 years ago

53f56894

add nested tensor matmul support (#81957)

YifanShenSZ committed 2 years ago

4bb7e148

Remove `THPTensor` defs, override macros, and `GenerateByteType.h` (#82503)

kurtamohler committed 2 years ago

32cf6c6f

Rename `_Typed/_UntypedStorage` to `Typed/UntypedStorage` and update docs (#82438)

kurtamohler committed 2 years ago

14d0296e

[vision hash update] update the pinned vision hash (#82522)

pytorchmergebot committed 2 years ago

28304dd4

[xla hash update] update the pinned xla hash (#82521)

pytorchmergebot committed 2 years ago

d74d79a8

[MPS] Add prelu (#82401)

qqaatw committed 2 years ago

7496937e

Introduce discontinuity to nested tensor (#80981)

YifanShenSZ committed 2 years ago

5f9939f6

Added AOT_PARTITIONER_DEBUG and added list of ops that inductor supports fusing (#82283)

Chillee committed 2 years ago

dc53cb45

[CUDNN] Update tests and dispatching for CUDNN V8 API behavior for `bfloat16` convs (#81139)

eqy committed 2 years ago

0b256645

Do not assume what is in `os.environ` (#82495)

malfet committed 2 years ago

b9cdd6d0

[torchgen] Fix multiple backends with custom namespace (#82133)

larryliu0820 committed 2 years ago

301fe8c2

redispatch sym overloads to sym versions correctly (#82380)

Krovatkin committed 2 years ago

3ca78a4c

[JIT] Revert SchemaInfo usage in AliasDb (#82475)

davidberard98 committed 2 years ago

6830573c

[PyTorch][Kineto] add ActivityType.h when USE_KINETO is not set (#82028)

shintaro-iwasaki committed 2 years ago

ccd30a12

Revert "Change SymIntNode into an intrusive pointer (#82432)"

pytorchmergebot committed 2 years ago

3b9cbb17

Revise sparse docs regarding Sparse Compressed tensors (#82108)

pearu committed 2 years ago

ff5399e5

Fix multiple docstring type mistakes (#82474)

ProGamerGov committed 2 years ago

8def154e

Change SymIntNode into an intrusive pointer (#82432)

ezyang committed 2 years ago

7be44f81

fix functionalization handling for mixed functional/nonfunctional tensorlists (#82326)

bdhirsh committed 2 years ago

7eed83e0

Codegen: Parse deprecated signatures as a full FunctionSchema (#82179)

peterbell10 committed 2 years ago

ba4727d4

Revert "Revert "Make factory functions CompositeExplicitAutograd (#82251)"" (#82470)

ezyang committed 2 years ago

2f95b61c

Newer Older