strisunshinewentingwang/pytorch

Pull Requests Commits

Eliminate unused parameters in PyTorch (#73751)

r-barnes committed 2 years ago

a7d2f39b

Don't discard stacktrace when rewriting AttributeError (#73720)

ezyang committed 2 years ago

c24783fb

Add `.jenkins/caffe2/*` to ONNX merge rules

malfet committed 2 years ago

0642285e

[Quant][core][devs] Removed support for non-quantized tensors in reflection_pad1d_cpu (#72485)

dzdang committed 2 years ago

a4160223

GHF: Add clickable links for Approved By

seemethere committed 2 years ago

0b79715d

[Quant][test] Added test to check if fp16 packing->unpacking yields the same result as to(torch.float16).to(torch.float32) (#73685)

dzdang committed 2 years ago

37d1ec9d

[SR] Fix a kwargs API default value bug (#73681)

Mike Iovine committed 2 years ago

818bf361

[static runtime] Fix aten::relu TE lowering to enable vectorization (#73713)

huiguoo committed 2 years ago

75f5ceaa

Refactor tools/actions_local_runner.py to allow custom remote/branch names

Thiago Crepaldi committed 2 years ago

cf143bdc

[shard] fix init_from_local_shards issue with deepcopy (#73400)

wanchaol committed 2 years ago

b955a046

Improve Python 3.10 compatibility

ezyang committed 2 years ago

22a3a17e

[Core ML] Attemp to fix the OOM issue (#73750)

xta0 committed 2 years ago

e6f8e521

Add a default implementation of __torch_dispatch__

ezyang committed 2 years ago

35cfa74f

[ci] don't build pytorch twice on windows

suo committed 2 years ago

7c0f166b

Add Ed to superusers

malfet committed 2 years ago

4267698a

Add Gflags for fusion strategy and make it local to executor (#73668)

Elias Ellison committed 2 years ago

fc362290

[fix] `torch.broadcast_shapes` should not handle shapes with negative dimensions. (#72999)

khushi-411 committed 2 years ago

905efa82

Remove istft python functional wrapper (#71993)

peterbell10 committed 2 years ago

9ad0578c

[FSDP] s/local_state_dict/_local_state_dict (#73532)

rohan-varma committed 2 years ago

6bb13f77

retry step for Build and upload nightly docker

cdeepali committed 2 years ago

5ae5f00a

Fix alignment, make sure release labels are included (#73739)

atalman committed 2 years ago

fc914346

Support `0`s in `out_size` of `FractionalMaxPoolNd`

Emilio Castillo committed 2 years ago

3186e366

dbr quant: add torchscript pass to remove redundant aliases (#71230)

vkuzo committed 2 years ago

bf896a29

quantization: fix bug in QuantWrapper with DeQuant qconfig (#73671)

vkuzo committed 2 years ago

eb8d0659

[GHA] Increase ASAN test timeout to 5h (#73740)

malfet committed 2 years ago

Verified 71ed3889

[Easy][c10d][DDP] (Reland) Minor fixes (#73569)

awgu committed 2 years ago

a37d54b6

[DDP][BE] (Reland) Remove bucket replicas (#73567)

awgu committed 2 years ago

7cce2d9d

Back out "[pytorch][PR] Support dataclasses in TorchScript"

Apoorva Garg committed 2 years ago

63932edc

(torch/elastic) add documentation clarifying that torchrun is a console script to torch.distributed.run (#73598)

Kiuk Chung committed 2 years ago

1a8bd1a7

Use intrusive_ptr for LazyTensor (#73445)

wconstab committed 2 years ago

b4173b80

Newer Older