strisunshinewentingwang/pytorch

Pull Requests Commits

[numpy] Add unary-ufunc tests for `erf` variants (#47155)

kshitij12345 committed 3 years ago

9ecaeb09

[docs] fix torch.swap{dim/axes} to showup in docs (#48376)

kshitij12345 committed 3 years ago

33cc1d6a

quant: make each line of fx/quantize.py <=80 chars (#48357)

vkuzo committed 3 years ago

bc2c1d7d

quant fx: fix typo (#48356)

vkuzo committed 3 years ago

1d984410

quant: add type annotations on quantization.fx.Quantizer matches (#48350)

vkuzo committed 3 years ago

8581c02a

Use libkineto in profiler (#46470)

Ilia Cherniavskii committed 3 years ago

f7a8bf28

[numpy] `torch.log1p` : promote integer inputs to float (#48002)

kshitij12345 committed 3 years ago

e9efd8df

An implementation of torch.tile as requested in pytorch/pytorch#38349 (#47974)

arbfay committed 3 years ago

2e0a8b75

Fix typos in comments (#48316)

malfet committed 3 years ago

2dff0b3e

[AutoAccept][Codemod][FBSourceClangFormatLinter] Daily `arc lint --take CLANGFORMAT`

generatedunixname89002005325676 committed 3 years ago

671ee71a

Added linalg.eigh, linalg.eigvalsh (#45526)

IvanYashchuk committed 3 years ago

4ed7f36e

Fix assertEqual's handling of numpy array inputs (#48217)

kurtamohler committed 3 years ago

b6654906

Add USE_KINETO build option (#45888)

Ilia Cherniavskii committed 3 years ago

f2da18af

quant: add type annotations on quantization.fx.Quantizer class vars (#48343)

vkuzo committed 3 years ago

c5e380bf

quant: enable mypy on torch/quantization/fx (#48331)

vkuzo committed 3 years ago

6b80b664

[Gradient Compression] clang-format test_c10d.py (#48349)

Yi Wang committed 3 years ago

cac553cf

[Gradient Compression] Define a customized state for PowerSGD comm hook (#48348)

Yi Wang committed 3 years ago

6400d27b

[TensorExpr] Fix lowering for aten::div. (#48329)

Mikhail Zolotukhin committed 3 years ago

b9671199

[TensorExpr] Fix aten::atan2 lowering and disable aten::pow lowering on CPU. (#48326)

Mikhail Zolotukhin committed 3 years ago

5e1faa1d

Fix mypy error (#48359)

Chillee committed 3 years ago

f1d32863

Disable old fuser internally (#48322)

Elias Ellison committed 3 years ago

a00ba630

[quant] Add backend_independent option for quantized linear module (#48192)

jerryzh168 committed 3 years ago

636fa8fd

Add Kineto submodule (separate PR) (#48332)

Ilia Cherniavskii committed 3 years ago

fdc62c74

[Pytorch Mobile] Disable OutOfPlace calls for mobile (#48255)

iseeyuan committed 3 years ago

6615edaf

Enable creation and transfer of ScriptModule over RPC (#48293)

rohan-varma committed 3 years ago

16d08973

[quant][fix] Fix quantization for qat.ConvBnReLU1d (#48059)

jerryzh168 committed 3 years ago

44def9ad

Explicitly cast an implicit conversion from some macro defined type to a double (#48290)

lanza committed 3 years ago

50e42b90

[FX] Prototype Conv/BN fuser in FX (#47657)

Chillee committed 3 years ago

0a3db1d4

Revert D25093315: [pytorch][PR] Fix inf norm grad

malfet committed 3 years ago

6d0947c8

Compare Weights FX Implementation (#48056)

DeepDriving committed 3 years ago

f8722825

Newer Older