strisunshinewentingwang/pytorch

Pull Requests Commits

Add torch.memory_format to base types

Chillee committed 3 years ago

Verified fd8df2c7

[NNC] Fix lowering of aten::remainder (#47611)

nickgg committed 3 years ago

f2eac5df

[NNC] Simplify and fix some bugs in Bounds Inference (#47450)

nickgg committed 3 years ago

0b30a8d0

Allow large inputs to svd_lowrank. Fix inaccuracy in torch.svd docs. (#47440)

pearu committed 3 years ago

c8a42c32

Enable Python code coverage for onnx runs (#47387)

janeyx99 committed 3 years ago

52fe73a3

Utility that loads a DP/DDP model state dict into a non-DDP model with the same architecture. (#45643)

ganprad committed 3 years ago

b631c872

move helper functions out of Partitioner class (#47515)

scottxu0730 committed 3 years ago

49d5b4d1

Add Vulkan op Conv2D. (#46900)

Ashkan Aliabadi committed 3 years ago

4841e9ef

Vulkan tweaks (#47261)

Ashkan Aliabadi committed 3 years ago

ce11dbbb

Add diagflat complex support (#47564)

rjkilpatrick committed 3 years ago

8aca85db

[ONNX] Add export of aten::is_floating point (#46442)

KsenijaS committed 3 years ago

79f85822

Fix inaccurate note in DistributedDataParallel (#47156)

0x4f5da2 committed 3 years ago

3dd26630

[caffe2] Add __slots__ to all classes in schema.py (#47541)

dongyuzheng committed 3 years ago

8b3f1d12

skip GPU test on sandcastle if sanitizer is enabled (#47626)

Rong Rong committed 3 years ago

2f617c51

Optimize backward for torch.repeat (#46726)

ejguan committed 3 years ago

86bb4136

[caffe2] Properly call super init in schema.py (#47542)

dongyuzheng committed 3 years ago

4c52a56c

[WIP] Adding bunch of unary foreach APIs (#47383)

Iurii Zdebskyi committed 3 years ago

b6a2444e

[ONNX] Slightly improve indexing with ellipsis under scripting (#46571)

BowenBao committed 3 years ago

5686d242

Update the docs of torch.eig about derivative (#47598)

mfkasim1 committed 3 years ago

a49367e9

[pytorch] split out trace type generator and migrate to new codegen model (#47438)

ljk53 committed 3 years ago

4159191f

[pytorch] factor out return_names api (#47437)

ljk53 committed 3 years ago

499d2fad

[pytorch] TraceType codegen tweak - newline before redispatch call (#47436)

ljk53 committed 3 years ago

8d1a6ae5

[ONNX] Fix scripting rand/randn/where (#45793)

BowenBao committed 3 years ago

e26c1726

Fix python 3.9 builds on Windows (#47602)

peterjc123 committed 3 years ago

a08e8dd7

Update nccl commit tag to head of v2.8 branch (#47603)

malfet committed 3 years ago

6214d0ad

Add batching rule for torch.clone(tensor, torch.contiguous_format) (#47365)

zou3519 committed 3 years ago

ead86b24

as_strided batching rule (#47364)

zou3519 committed 3 years ago

7bc8fdb6

[tensorexpr] Fix registration of intrinsics on llvm-fb (#47540)

bertmaher committed 3 years ago

77c49e65

[fx] add missing modules for type annoations (#47537)

zdevito committed 3 years ago

70d34718

Fix compiler warning variable "num_ivalue_args" was declared but never referenced detected during: (#47494)

zasdfgbnm committed 3 years ago

fbffd959

Older