strisunshinewentingwang/pytorch

Pull Requests Commits

[PyTorch] Don't store multiple kernels per key on mobile (#64447)

swolchok committed 3 years ago

eedc234e

[fx const fold] fix some cases with deep model hierarchy (#64945)

842974287 committed 3 years ago

446d95a7

[Model Averaging] Revert #63895 (#64903)

Yi Wang committed 3 years ago

00e6e0c5

Drop incremental linking on Windows with REL_WITH_DEB_INFO=1. (#64892)

Nick Kreeger committed 3 years ago

882b67df

Disable target determination for now (#64921)

malfet committed 3 years ago

01cfea94

print_test_stats.py: dedup test report upload name with TEST_CONFIG (#64948)

janeyx99 committed 3 years ago

4e225da3

Make {select,slice,diagonal}_backward primitives wrt autograd (#64933)

zou3519 committed 3 years ago

e8845540

Replace composite dispatch with `CompositeExplicitAutograd` (#64641)

ysiraichi committed 3 years ago

2853c7da

Revert D30711934: [pytorch][PR] Use RDS for build size tracking

ezyang committed 3 years ago

09d221e8

[TensorExpr] Remove 'Placeholder' class. (#64887)

Mikhail Zolotukhin committed 3 years ago

f23f21da

[TensorExpr] PyBinds: improve QoL of pybind users. (#64886)

Mikhail Zolotukhin committed 3 years ago

199031c4

Fix use of deprecated tensor.type() in SegmentReduce.cpp (#64151)

peterbell10 committed 3 years ago

caaa6efc

[quant] handle empty input in fused_moving_avg_obs_fake_quant op (#64829)

supriyar committed 3 years ago

d4b4d835

Add forward AD for torch.linalg.eigh (#62163)

IvanYashchuk committed 3 years ago

0aef44cb

[THC] remove TensorTypeUtils and TensorInfo (#64965)

ngimel committed 3 years ago

35c82dbf

EmbeddingBag sort thrust->cub (#64498)

zasdfgbnm committed 3 years ago

816048e7

Speed up torch.unique_consecutive() (#64835)

oToToT committed 3 years ago

ed30afd4

[WIP] Example of DataPipes and DataFrames integration (#60840)

VitalyFedyunin committed 3 years ago

ab5e1c69

Re-land Fix test report uploading (#64958)

driazati committed 3 years ago

ee554e2e

[iOS][OSS][BE] Add Simulator tests for full JIT (#64851)

xta0 committed 3 years ago

f159f12f

add acc_ops.max, acc_ops.maximum, consolidate acc_ops.min and acc_ops.minimum

emad committed 3 years ago

fd09e564

Add BFloat16 support for cross, tril, triu, tril_indices, triu_indices and cumsum operators on CPU (#62454)

CaoE committed 3 years ago

3855c246

Use RDS for build size tracking (#64303)

driazati committed 3 years ago

1cd0252e

Add `skipIfTBB` decorator (#64942)

malfet committed 3 years ago

c4073af6

Raise TypeError on assigned grad with wrong type (#64876)

Varal7 committed 3 years ago

8131bc85

kill SkipInfo (#64878)

ngimel committed 3 years ago

1e25a849

Fix TRTOperatorSupport (#64873)

Shirong Wu committed 3 years ago

3710edc8

Revert D30878101: [pytorch][PR] Fix test report uploading

seemethere committed 3 years ago

914e3a86

torch.ao migration: fake_quantize.py, phase 1 (#64814)

vkuzo committed 3 years ago

6101cbce

[PyTorch] Reduce heap allocations in OperatorName::setNamespaceIfNotSet (#64673)

swolchok committed 3 years ago

e4314dac

Newer Older