strisunshinewentingwang/pytorch

Pull Requests Commits

[skip ci] Set test owner for nn tests (#66850)

janeyx99 committed 2 years ago

9eab6da8

Fix BatchMatMul test and shape inference (#66733)

qxy11 committed 2 years ago

05b6dc9d

add `OpInfo` for `torch.nn.pixel_unshuffle` (#65468)

pmeier committed 2 years ago

9f782f8b

add `OpInfo` for `torch.nn.pixel_shuffle` (#65467)

pmeier committed 2 years ago

1164118f

add `OpInfo` for `torch.nn.functional.pairwise_distance` (#65460)

pmeier committed 2 years ago

8f09292c

[quant][embedding qat] Add eager QAT test for EmbeddingBag+Linear model (#66334)

b-koopman committed 2 years ago

0036e411

use irange for loops 1 (#66741)

r-barnes committed 2 years ago

0a07488e

[caffe2] Fix invalid vector accesses and polar() call (#66757)

ot committed 2 years ago

72803dbc

Add `SourceView` which doesn't own source text as base class of `Source` (#65309)

gmagogsfm committed 2 years ago

147f7559

[DDP] Track models with sync bn (#66680)

rohan-varma committed 2 years ago

bff64e84

use irange for loops 5 (#66744)

r-barnes committed 2 years ago

e0643fa3

use irange for loops 3 (#66747)

r-barnes committed 2 years ago

bceb1db8

Skip failing tests when LAPACK and MAGMA are not available (#64930)

IvanYashchuk committed 2 years ago

061baf02

[PyTorch] Pass c10::optional<bool> to Stride ctor by value (#66698)

swolchok committed 2 years ago

08a464a9

test addr type promotion in a single test (#66812)

Natalia Gimelshein committed 2 years ago

c9c52b76

Move convertShape to Generated header, prevent warning (#66821)

wconstab committed 2 years ago

Verified 81881c32

Add lazy Node base and associated infra (#66601)

wconstab committed 2 years ago

d05c1ec0

[PyTorch][easy] Fix missing move in UnionType::createWithContained (#66691)

swolchok committed 2 years ago

a17a4e93

[PyTorch] Fix missing moves in ListType (#66701)

swolchok committed 2 years ago

c9c447f4

[PyTorch][easy] Don't copy string in TensorType::repr_str unnecessarily (#66699)

swolchok committed 2 years ago

d0a63c97

[PyTorch] Avoid refcount bump in UnionType::canHoldType (#66693)

swolchok committed 2 years ago

f65b4b7a

[nn] MultiLabelSoftMarginLoss : no batch dim support (#65690)

kshitij12345 committed 2 years ago

1db50505

Move Lower() method to {Backend}Node (#66749)

wconstab committed 2 years ago

Verified b3715a86

move get_cycles_per_ms() to common_utils (#66798)

zhaojuanmao committed 2 years ago

8173d4df

ci: Move bazel download from github -> s3 (#66815)

seemethere committed 2 years ago

d024f113

[not4land][quant][fx][graphmode] lower reference linear module example (#65723)

jerryzh168 committed 2 years ago

06e49ea0

Update documentation of torch.nn.Upsample (#66756)

JBamberger committed 2 years ago

c994a7fc

Prefer mT and mH over transpose(-2, -1) and transpose(-2, -1).conj() (#64181)

lezcano committed 2 years ago

0974215c

[PyTorch] Use intrusive_ptr to save space in KernelFunction (#65618)

swolchok committed 2 years ago

44fd3126

[PyTorch] Take const Type& in TensorType::fromNumberType (#66716)

swolchok committed 2 years ago

622e19b8

Newer Older