strisunshinewentingwang/pytorch

Pull Requests Commits

[PyTorch Edge] Add OwnedOrBorrowedVector for QNNPack BCSR Indices/Values (#80476)

salilsdesai committed 2 years ago

98e4524d

[PyTorch Edge] Add qnnpack bcsr matrix unpacking and use unpacking in Linear module (#80475)

salilsdesai committed 2 years ago

5c12cd22

[PyTorch Edge] Add serialization/deserialization of Sparse Quantize Linear Packed Params (#80474)

salilsdesai committed 2 years ago

eaf817df

[FSDP] Fix `full_optim_state_dict()` hang (#80712)

awgu committed 2 years ago

ab09f346

[PyTorch Edge] Remove Original Weight Tensor from QNNPack Sparse Quantized Linear Packed Params (#80473)

salilsdesai committed 2 years ago

523b081a

[Expanded Weights] fix unbatched support issue (#80891)

samdow committed 2 years ago

d96d1865

[composite compliance] preserve stride correctly for non-contiguous tensor with requires_grad=True (#81035)

kshitij12345 committed 2 years ago

e51c63da

Fix macos public bindings failures (#80970)

anjali411 committed 2 years ago

120987ff

ci: Remove extra docker pull (#81038)

seemethere committed 2 years ago

fcf027c4

Added logic for pushing docker images to ghcr.io (#80950)

DanilBaibak committed 2 years ago

91220799

[MPS] Fix LSTM batch_first output transposed (#80597)

qqaatw committed 2 years ago

b0b24b42

Add `-Werror=non-virtual-dtor` (reland) (#81012)

malfet committed 2 years ago

2beb57a8

[ci] simplify sccache stats uploading (#80806)

suo committed 2 years ago

07e41652

Stopped ProxyTensor from turning aten::lift tensors into proxy objects (#81024)

Chillee committed 2 years ago

b7046e9b

fix weight norm backward bug on CPU when OMP_NUM_THREADS <= 2 (#80930)

mingfeima committed 2 years ago

6ee54a87

Revert "Test xla llvm_build_cache patch (#80178)"

pytorchmergebot committed 2 years ago

7234b027

Correctly setup ancestors on explicit push mode. (#80995)

ezyang committed 2 years ago

3ca309c4

[Profiler] Eliminate transitive include of profiler implementation headers. (#80564)

Taylor Robie committed 2 years ago

b42c50da

Add const modifier for lazy shape inference for boolean logic ops (#79999)

wonjoolee95 committed 2 years ago

f98bfce6

Load CPU MLModel first, and configured MLModel async (#80941)

jmdetloff committed 2 years ago

665fd30a

[MPS] Add argmin (#80828)

qqaatw committed 2 years ago

2458b3cd

Test xla llvm_build_cache patch (#80178)

yeounoh committed 2 years ago

457e400a

Don't invoke mode as overloaded argument in torch dispatch (#80992)

ezyang committed 2 years ago

74877943

[caffe2] Use `arch_deps` instead of host info for arch-specific deps (#80814)

andrewjcg committed 2 years ago

74b41af2

Use miopen_LIBRARIES and rccl_LIBRARIES directly, when they are valid target (#80446)

dmikushin committed 2 years ago

e08026d4

Remove dead code from torch.ops torch function handling (#80993)

ezyang committed 2 years ago

91b02506

[caffe2/perfkernels] Avoid `native.host_info()` in build files (#80812)

andrewjcg committed 2 years ago

13de7275

Detect ProxyTensor layering violations (#80994)

ezyang committed 2 years ago

e89b1991

Revert "Add -Werror=non-virtual-dtor (#80584)"

pytorchmergebot committed 2 years ago

0491c10a

Revert "Revert "[FSDP Optim State] Remove checkpoint prefix (#80480)"" (#80936)

rohan-varma committed 2 years ago

0c5fdfd9

Newer Older