strisunshinewentingwang/pytorch

Pull Requests Commits

[quant][core][improvement] Added earlier termination and improved error message for calling min and max ops on per channel quantized tensors.

dzdang committed 2 years ago

b12d951f

[ONNX] Enable script tests based on recent exporter changes (#77254)

shubhambhokare1 committed 2 years ago

d26c575f

Add onnx support for `torch.lerp` (#78891)

qqaatw committed 2 years ago

b45d303d

torch/package: add fix for implicit numpy dependency (#78979)

d4l3k committed 2 years ago

5f7b9202

Revert "[hot fix] Disable MPS tests as machines are down (#79119)"

pytorchmergebot committed 2 years ago

68ef29b4

Foward fix sharding bug for DL (#79124)

ejguan committed 2 years ago

5158a6b4

[quant][improvement] Added quantized fill test case for quantized cuda tensors

dzdang committed 2 years ago

541b9d25

Fix LeakyReLU spelling (#79102)

ShawnZhong committed 2 years ago

c1516e0c

ns for fx: further fixes for kwargs-only

vkuzo committed 2 years ago

d4aa204f

[reland][complex32, jiterator] cos, sinh, cosh, tanh (#78718)

khushi-411 committed 2 years ago

5b32c344

Test torch._refs with aten and nvfuser executors (#78926)

IvanYashchuk committed 2 years ago

ff39e349

[Vulkan] Add generic VulkanOpContext

manuelcandales committed 2 years ago

446d5bae

[hot fix] Disable MPS tests as machines are down (#79119)

janeyx99 committed 2 years ago

f2d67c54

move MPS compat into common comparison machinery (#77836)

pmeier committed 2 years ago

32593ef2

Revert "add some instructions for ios test (#79097)"

pytorchmergebot committed 2 years ago

496e294e

Add onnx support for `torch.tensor_split` (#77437)

qqaatw committed 2 years ago

e4c3e98a

[Vulkan] Fix bug in GRU op (#78945)

manuelcandales committed 2 years ago

9a0bb4b6

[mobile] Fix lightweight dispatch OOM error by introducing selective build

pytorchmergebot committed 2 years ago

272bdb14

[forward ad] Sync conj for between primal and tangent on set forward grad

soulitzer committed 2 years ago

99ffeff9

[PyTorch] Avoid initializing storage for empty Optionals

swolchok committed 2 years ago

cf3ce329

Install torchdynamo as part of most CI jobs

ezyang committed 2 years ago

65fd0cdd

Generalize update-xla-commit-hash and setup torchdynamo hash

ezyang committed 2 years ago

bbd49bca

add some instructions for ios test (#79097)

linbinyu committed 2 years ago

e1534e3f

[GHF] Workaround GH GQL limitation (#79090)

malfet committed 2 years ago

530ee1c3

Revert "[Profiler] Move python tracing to unified event type (Part 2)"

pytorchmergebot committed 2 years ago

4305f8e9

Make torch.library decorators return function

ezyang committed 2 years ago

acecd4ac

cdist meta function

ezyang committed 2 years ago

41bd5b85

[pytorch] Update Gloo submodule to include rocm fixes (#77207)

minsii committed 2 years ago

8d9e5d0b

Revert "Revert "moved logit to use torch ops instead of refs + added …a couple more decompositions"" (#79082)

pytorchmergebot committed 2 years ago

8ce310b9

[nvfuser_upstream_push] nvfuser code base bump 052422 (#78244)

jjsjann123 committed 2 years ago

9e52ad28

Newer Older