strisunshinewentingwang/pytorch

Pull Requests Commits

Grouping allreduces (#1644)

naoyam committed 2 years ago

Verified d0637681

Fix grid sync allocation and codegen (#1650)

naoyam committed 2 years ago

Verified 97636cfb

fixing workflow

jjsjann123 committed 2 years ago

3470a696

remove v3 option (#1647)

shmsong committed 2 years ago

Verified 3834cb95

Merge remote-tracking branch 'origin/master' into HEAD

jjsjann123 committed 2 years ago

eba22511

Merge remote-tracking branch 'upstream/master' into HEAD

jjsjann123 committed 2 years ago

2d7ee6f6

Added gelu decomposition

Chillee committed 2 years ago

ed18181d

Use scatter_reduce to support masked reductions on sparse COO tensors (sum, prod, amin, amax)

mikaylagawarecki committed 2 years ago

fc2a2e8b

[NVFuser] environment variable to turn nvfuser on or off (#76485)

davidberard98 committed 2 years ago

e33f3229

shard `pull / linux-xenial-cuda11.3-py3.7-gcc7 / test (distributed` 1->2

clee2000 committed 2 years ago

56ea57de

[xplat] add static_cast where missing (#76756)

rmaz committed 2 years ago

3a2fc312

[PyTorch][AMD] fix hipify_python (#76720)

shintaro-iwasaki committed 2 years ago

20e4d6c4

Added proxy tensor

Chillee committed 2 years ago

fc95eda2

Support no sharding config

zhaojuanmao committed 2 years ago

4fbbbed6

nit cleanup in lower_loops.cpp (#1648)

zasdfgbnm committed 2 years ago

Verified b6593d45

[lint] upgrade mypy to latest version

suo committed 2 years ago

fb0f2856

Remove breakpad dependency

malfet committed 2 years ago

8473173c

Bugfix NAN and Inf handling for scatter_reduce (amin and amax)

mikaylagawarecki committed 2 years ago

4441582f

Revert "[lint] upgrade mypy to latest version"

pytorchmergebot committed 2 years ago

3d7428d9

Revert "Add atan2 isfinite isinf isnan isneginf isposinf isreal to nvfuser and its frontend"

pytorchmergebot committed 2 years ago

4bb59441

[lint] upgrade mypy to latest version

suo committed 2 years ago

9bf18aab

[lint] fix pip init for lint when user has a global install

suo committed 2 years ago

84641d0d

[SR] Quick hack to eliminate no-op slice (#75774)

Mike Iovine committed 2 years ago

b02b3f25

[lint] register formatter linters

suo committed 2 years ago

4537ac11

Turn on memory efficient format for jit pickle files.

qihqi committed 2 years ago

aca55948

Adding output_size to to_padded_tensor (#76640)

mjanderson09 committed 2 years ago

76abbbe3

Nested Tensor Elementwise (#76470)

mjanderson09 committed 2 years ago

4d6b145b

Add binary_cross_entropy and trace decomp - fixed _log_softmax/_softmax dtype promotion semantics

Chillee committed 2 years ago

8a3e9255

Update obselete comment (#1646)

zasdfgbnm committed 2 years ago

Verified ee4c0041

[EASY] Quick Fix for broken shape function autogen.

Gamrix committed 2 years ago

db21e22b

Newer Older