strisunshinewentingwang/pytorch

Pull Requests Commits

Targeted documentation updates in autograd.functional (#72111)

zou3519 committed 2 years ago

f99147de

[TensorExpr] Move AOT compilation logic from aot_compiler.cpp to NNC's to_backend (#70375)

Mikhail Zolotukhin committed 2 years ago

a60e2ae0

[TensorExpr] AOT Compiler: support symbolic shape arguments. (#70374)

Mikhail Zolotukhin committed 2 years ago

64668e61

Revert D33892443: use irange for caffe2/aten directory

suo committed 2 years ago

a7cebda9

[Pytorch Edge] Fix Custom Class Parser (#72153)

JacobSzwejbka committed 2 years ago

a34c17bf

[fix] EmbeddingBag segfault for out-of-bounds idx (#71904)

kshitij12345 committed 2 years ago

aa5dab02

Fix persistent worker exits before pin_memory thread (#71579)

ejguan committed 2 years ago

67a275c2

use irange for caffe2/aten directory (#72067)

Charles Zhang committed 2 years ago

10cc66bc

Fix internal assert custom function when input does not require grad (#72008)

soulitzer committed 2 years ago

e39bf133

Revert D32053748: [pytorch] use cublas lt interface for bias fusion

malfet committed 2 years ago

26f88eb0

Update flatbuffers to v2.0.5 (#72132)

malfet committed 2 years ago

b9f9d78a

[SR] Eliminate op_name_ in ProcessedNode (#71986)

Mike Iovine committed 2 years ago

3dce68fd

fixing stride order for expanded tensor (#71665)

jjsjann123 committed 2 years ago

b1897d6d

Clean up old warnings in F.interpolate (#72093)

pejato committed 2 years ago

b8a4ee5e

Process commit update2

albanD committed 2 years ago

29d91002

Avoid unnecessary copy of padding/dilation vectors in check_shape_forward (#72019)

ezgicicek committed 2 years ago

d0f397ae

[tensorexpr][quant] Enable tensorexpr for quant,dequant (#71243)

IvanKobzarev committed 2 years ago

9e8334e3

[nnc] tensorexpr for quantized/aten::upsample_nearest2d (#71236)

IvanKobzarev committed 2 years ago

34e4418d

Add lowering path for LinearReLU module (#71427)

andrewor14 committed 2 years ago

e118d6e5

Update process_commit.py

albanD committed 2 years ago

be7ee926

Add docs for fusion strategy (#72036)

Elias Ellison committed 2 years ago

e0a0f37a

update default fusion strategy (#72038)

Elias Ellison committed 2 years ago

a55ef69e

[nnc] Update cuda codegen to use llvm for thread and block extent computations (#72040)

Elias Ellison committed 2 years ago

b44f724a

NNC Dynamic Channels last fixes (#72032)

Elias Ellison committed 2 years ago

27a4d397

[NNC] Add Tests for Dynamic Shape Fusion Change default fusion strategy (#71651)

Elias Ellison committed 2 years ago

59a63756

Refactor PE so fusion specializations are configurable (#71650)

Elias Ellison committed 2 years ago

f1499d6c

[WIP] add explicit dynamic fusion arg (#71173)

Elias Ellison committed 2 years ago

cf1833df

Add logspace test modules (#72052)

tugsbayasgalan committed 2 years ago

e305248a

Added remove_duplicate parameter to `nn.Module` (#39)

Chillee committed 2 years ago

7e821754

Upgrade oneDNN to v2.5.2 (#71546)

yanbing-j committed 2 years ago

4567d5de

Newer Older