strisunshinewentingwang/pytorch

Pull Requests Commits

Enable inductor CI for TorchBench (#87465)

desertfire committed 1 year ago

31b10e7d

[DataPipe] Add container template for _Fork and _Demux (#89216)

ejguan committed 1 year ago

3d8a853a

Fix typo in aten/src/README.md (#89175)

keineahnung2345 committed 1 year ago

e2229a89

Add tests for replicate multiple modules (#89099)

yhcharles committed 1 year ago

a695fcf2

[JIT][Security] Do not blindly eval input string (#89189)

malfet committed 1 year ago

767f6aa4

Fix distributed test paths when running periodic multigpu job (#89225)

huydhn committed 1 year ago

fbbf3687

reland "support running test_mobile_profiler with buck1/buck2 and OSS (#89001)" (#89091)

mikey dagitses committed 1 year ago

f057a45f

Add an env var to skip cudnn version compatibility check (#89184)

xwang233 committed 1 year ago

e856a4d6

Rewrite assert statement with torch._assert under config (#88246)

tugsbayasgalan committed 1 year ago

04169c5b

Fix bug in dynamo dashboard summary stats diff (#89226)

williamwen42 committed 1 year ago

af448e84

Revert "Support masked_fill (#88736)"

pytorchmergebot committed 1 year ago

706f791a

Revert "Reland "Towards unifying symbolic and non symbolic fake tensor (#89038)" (#89143)"

pytorchmergebot committed 1 year ago

8e4c9828

Fix buffer overflow from AddressSanitizer checks due to inaccurate bfloat16 representation of large integer (#89210)

jgong5 committed 1 year ago

cd81a700

Support masked_fill (#88736)

EikanWang committed 1 year ago

2b131b1d

Reland "Towards unifying symbolic and non symbolic fake tensor (#89038)" (#89143)

ezyang committed 1 year ago

e686b8c3

Fix typo in dist_util.py (#89167)

wconstab committed 1 year ago

bdc99115

Add BFloat16 support and optimization for mish, hardtanh backward, and silu on CPU (#82460)

CaoE committed 1 year ago

3beccbc2

Add warning if tensor cores are not used (#88844)

msaroufim committed 1 year ago

37c85cf5

[Dynamo] Support typing.Mapping & Support function as argument (#88963)

yanboliang committed 1 year ago

b72f5b9a

[ONNX] Add onnx-script into ONNX docs (#89078)

titaiwangms committed 1 year ago

126e4417

[dynamo][benchmarks] HF - Fix seq len and batch sizes (#89165)

anijain2305 committed 1 year ago

74610a1c

Round out rad2deg sparse support (#88442)

amjames committed 1 year ago

a41f7060

Use software approach to catch overflow ( `c10/utils/safe_numerics.h` ) on ARM devices (#89042)

Rachel030219 committed 1 year ago

70fb673e

Fix: prefer .is_none() over .is(py::none()) for pybind11 in caffe2 (#88199)

Skylion007 committed 1 year ago

54fca6a9

Revert "Redefine the simdlen semantic: (#88482)"

pytorchmergebot committed 1 year ago

4e1d19c5

[MPS] Add binary operations dtype precedence test case (#87545)

lhoenig committed 1 year ago

81a8fdc4

Fix empty input issue of convolution for channels last memory format (#86521)

CaoE committed 1 year ago

44c9185f

[xnnpack][executorch] Pass xnnexecutor pointer to compileModel() (#89090)

mcr229 committed 1 year ago

637e764e

[torchrec] [composable] update ShardedEmbeddingBagCollection to be use registered EBCs with shardedTensors as registered modules (#758) (#88026)

colin2328 committed 1 year ago

24b9890f

Fix typos in messages under torch (#89049)

kiszk committed 1 year ago

1cd6ebe0

Newer Older