strisunshinewentingwang/pytorch

Pull Requests Commits

Prepare python jiterator for multiple outputs

SherlockNoMad committed 2 years ago

f4f09ac6

Implement sym_sizes to create proper IR for sym ints representing tensor sizes (#77756)

Krovatkin committed 2 years ago

df1f9b98

[quant] Reordering the imports in the torch/__init__.py

z-a-f committed 2 years ago

a8c929b0

[symint] add support for negative integers

suo committed 2 years ago

68e22aa9

Create JSON from new FX IR and lower to LLVM (#77765)

dborkovic committed 2 years ago

f54098cd

MPS Fixes: copy operations, addmm and baddmm (#77791)

kulinseth committed 2 years ago

3d83321b

[StaticRuntime] Replace Permute with copy version only when it's followed by reshape or flatten (#77832)

Hao Lu committed 2 years ago

c60d2ef4

Revert "Fixed type promotion semantics for native_batch_norm and native_layer_norm (#77407)"

pytorchmergebot committed 2 years ago

03546e9c

Restore old names for private funcs in legacy storages (#77861)

kurtamohler committed 2 years ago

cecb2ad9

fixing trivial reduction & broadcast scheduling (#77884)

jjsjann123 committed 2 years ago

6583c038

[ONNX] Clean up module imports (#77423)

justinchuby committed 2 years ago

0d76299f

Added support for `slogdet` in LazyTensor shape inference (#77904)

miladm committed 2 years ago

e67284d9

Added support for `inverse` in LazyTensor shape inference (#77888)

miladm committed 2 years ago

d6ae6507

[GHF] Add checkruns pagination (#77922)

malfet committed 2 years ago

67c30a04

torch: Fix black linter

seemethere committed 2 years ago

1bec7f84

fix another missing c10::

davidberard98 committed 2 years ago

a5766c86

ENH: Convert finfo.tiny to finfo.smallest_normal (#76292)

HaoZeke committed 2 years ago

c915fbe2

Add missing decref to `createStorageGetType` (#77860)

kurtamohler committed 2 years ago

7892a457

Adding activation references for celu, mish, selu, softplus, and tanh (#77473)

kevinstephano committed 2 years ago

11daf200

[FSDP][Easy] Update `state_dict()` docstring

awgu committed 2 years ago

e69d13b8

[FSDP][Easy] Reword device placement warning

awgu committed 2 years ago

d9b3feb2

[FSDP][Easy] Fix `state_dict_type()` docstring example

awgu committed 2 years ago

36bf8007

[FSDP][Easy] Doc fixes

awgu committed 2 years ago

96e674a0

Populate bytecode version and operator version (#77685)

qihqi committed 2 years ago

0bc4b2af

Revert RPC Meta device support

pbelevich committed 2 years ago

94eba341

Prepare Jiterator code template for multiple outputs (#77902)

SherlockNoMad committed 2 years ago

82682aab

Revert "Switch to use nested tensor by-default in TransformerEncoder (#77217)"

pytorchmergebot committed 2 years ago

0d6fa91d

add slow path for is_contiguous (#77906)

george-qi committed 2 years ago

294fff16

aten: Removed unused-local-typedef

seemethere committed 2 years ago

009c14b0

Support saving extra files in pytorch flat buffer format via python (#77870)

Ian Graves committed 2 years ago

b3ca7fc8

Older