strisunshinewentingwang/pytorch

Pull Requests Commits

[DOC] Use type hints to show annotation in the docs (#79086)

ShawnZhong committed 1 year ago

e552cf10

Improve NestedTensor documentation (#85186)

mikaylagawarecki committed 1 year ago

a77f2a95

Revert distributed test parallelization (#86756)

huydhn committed 1 year ago

be81f3d8

Add hooks for register_buffer/module/parameter (#86148)

antoniojkim committed 1 year ago

09a676f6

Let retried jobs advance viable/strict (#86821)

ZainRizvi committed 1 year ago

c08cbfcc

Update _torch_docs / ldexp (#86721)

vfdev-5 committed 1 year ago

3b266802

[quant][fx] Fix weight_dtype and bias_dtype backend_config checks (#86719)

jerryzh168 committed 1 year ago

363b108e

[ao] fixing public v private for fx.pattern_utils.py (#86033)

HDCharles committed 1 year ago

d6bfbdf5

[ao] fixing public v private for fx.graph_module.py (#86032)

HDCharles committed 1 year ago

bf0116d1

[ao] fixing public v private for quantization_types (#86031)

HDCharles committed 1 year ago

25476f2e

Use CUTLASS GEMM for NT bmm [OSS-only] (#85894)

cpuhrsch committed 1 year ago

ef58a132

Display unexpected exceptions raised from test_dtypes (#86599)

peterbell10 committed 1 year ago

73c43ce2

Add AutocastHPU support (#84927)

australopitek committed 1 year ago

6be9d9a6

Disable tf32 in functorch transform tests (#86799)

zou3519 committed 1 year ago

553eaaba

[primTorch] Add ref for `triplet_margin_loss`, improve `triplet_margin_with_distance_loss` (#85614)

nkaretnikov committed 1 year ago

d56017a1

Extend torch.cuda.is_available() to attempt an NVML-based CUDA availability assessment when explicitly requested by the user (#85951)

speediedan committed 1 year ago

ce56ee11

Add prims.clone (#86705)

IvanYashchuk committed 1 year ago

cd7c86ea

[BE] Store helper functions C++ for python API parity (#82136)

H-Huang committed 1 year ago

3356d038

[ONNX] Support device().type() string comparison with constant (#86168)

BowenBao committed 1 year ago

cc7ea93c

[ao] fixing public v private for backend_config.native.py (#86030)

HDCharles committed 1 year ago

58542eb2

Added define to fix issue with compatibility with latest Windows SDK (#85408)

vladimir-aubrecht committed 1 year ago

409efeba

Allow PrivateUse1 backends to not have Storage (#86557)

smk2007 committed 1 year ago

f24d174f

use cff standard for citation information (#86200)

pmeier committed 1 year ago

61a58986

[primTorch] decomposition for bucketize (#86366)

Fabio Rocha committed 1 year ago

493ded24

Patching getitem in partitioner (#86713)

jjsjann123 committed 1 year ago

f903f1ab

[primTorch] special: entr, expit (#86592)

khushi-411 committed 1 year ago

23441351

Add type and shape annotation for gm.print_readable() (#86562)

SherlockNoMad committed 1 year ago

a47f93b6

Revert "Backport currently dont work with some models if: (#86510)"

pytorchmergebot committed 1 year ago

e0d6898c

(Re-open) Adds cudaMallocAsync as an alternative backend for the CUDA allocator (#82682)

eqy committed 1 year ago

25725fd6

Add testing on A10G GPU to periodic workflow (#85524)

malfet committed 1 year ago

a216f470

Newer Older