strisunshinewentingwang/pytorch

Pull Requests Commits

Ensure tensors are contiguous in functional all_gather.

pritamdamania87 committed 2 years ago

500fb247

[BE][TryMerge] Do not re-import trymerge multiple times (#79760)

malfet committed 2 years ago

4661c75f

[PyTorch] Don't create a new Storage in FreeMemory unnecessarily

swolchok committed 2 years ago

6e90572b

Support dropout(nested tensor) (#79318)

YifanShenSZ committed 2 years ago

1211ab67

fx quant: unbreak Mac OS tests when MKLDNN is available

vkuzo committed 2 years ago

91c5fc32

add utility to compute self time of events

davidchencsl committed 2 years ago

9955fff4

Functionality/pickling for commhooks (#79334)

Olga Andreeva committed 2 years ago

8a6d8307

Revert "Enable `dim=None` for `torch.sum` (#75845)"

pytorchmergebot committed 2 years ago

ee6ebfc0

Add nested tensor support to autograd (#79446)

drisspg committed 2 years ago

f9656817

CI Disable xla for infra issues (#79726)

zengk95 committed 2 years ago

4b342b30

[Static runtime] Pass parent graph metadata to forked subgraphs (#79578)

akshayParashar1995 committed 2 years ago

ca7ab170

[MPS]: Add fix for squeezed input axes handling in BCE loss (#79676)

kulinseth committed 2 years ago

4615f6aa

Enable `dim=None` for `torch.sum` (#75845)

kurtamohler committed 2 years ago

e79a51f7

[BE][FSDP] Enhance summon full params test (#79642)

rohan-varma committed 2 years ago

c8b073c5

true up c10::FunctionSchema::operator<< to print native_functions.yaml syntax

bdhirsh committed 2 years ago

9d419e81

Revert "Reland PySymInt (#79617)"

pytorchmergebot committed 2 years ago

44436947

reenable buck test (#79721)

zengk95 committed 2 years ago

226a5e87

reorder cpuinfo and clog deps in TorchConfig.cmake (#79551)

powderluv committed 2 years ago

e3135946

[nvfuser_upstream_push] Reland: nvfuser code base bump 060822 (#79406)

jjsjann123 committed 2 years ago

c9c402ea

[Quant][fx][bc-breaking] Replace *custom_config_dict with config objects

andrewor14 committed 2 years ago

78144b9f

[pytorch][gh1] Fix failing test (#79632)

bigfootjon committed 2 years ago

920ac70e

canonicalize includes of form <aten/src/ATen/...>

Michael Andreas Dagitses committed 2 years ago

acd07296

[ao][sparsity] Support for L2 norm based block data sparsifier

macandro96 committed 2 years ago

751fbc4c

[ao][sparsity] L1 norm based block data sparsifier

macandro96 committed 2 years ago

419b82e0

[ao][sparsity] Support for embeddings and embedding bags in BaseDataSparsifier

macandro96 committed 2 years ago

70fc8652

[ao][sparsity] Support for the nn.Parameter in BaseDataSparsifier

macandro96 committed 2 years ago

490631aa

[ao][sparsity] Implemented state dict and serialization functionalities

macandro96 committed 2 years ago

69347e9f

[ao][sparsity] Support for sparsifying data operations on raw torch tensors.

macandro96 committed 2 years ago

596095dc

[ao][sparsity] Base class for Data Sparsifier

macandro96 committed 2 years ago

15828bcf

Enable embedding forward over reverse support (#79699)

samdow committed 2 years ago

2539efb8

Newer Older