strisunshinewentingwang/pytorch

Pull Requests Commits

[MPS] Add support for MPSProfiler Python bindings (#101002)

razarmehr committed 1 year ago

0be53d83

Add branch change name for composite action (#101309)

malfet committed 1 year ago

816400a2

Update PyTorch docker base image to Ubuntu-20.04 (take 2) (#101310)

malfet committed 1 year ago

47c99e3a

[inductor] Insert triton barrier before storing to inplace buffers (#100769)

peterbell10 committed 1 year ago

5fe834af

[PyTorch] Avoid extra refcounting in vector variant of VariableType::unpack (#95835)

swolchok committed 1 year ago

05077f2a

[CI] Change dashboard workflow inputs type to boolean (#101308)

desertfire committed 1 year ago

6afa9a4a

Fix typos in troubleshooting.rst (#101305)

FindHao committed 1 year ago

a12b640d

Cpp Reduce LR on plateau scheduler (#100311)

pawelKapl committed 1 year ago

6ac05427

Use 20.04 as base image (#101301)

malfet committed 1 year ago

c772d569

[CI] Add workflow_dispatch.inputs to control dashboard runs (#101279)

desertfire committed 1 year ago

066175d6

[PyTorch] add test for numel slow path affecting data_ptr (#100993)

swolchok committed 1 year ago

a8c32eb7

[BE][GHF] Add `retries_decorator` (#101227)

malfet committed 1 year ago

568bac79

Handle tail 0-size tensor appropriately in `MultiTensorApply` (#100811)

crcrpar committed 1 year ago

2fcc2002

[caffe2/tools/autograd] Fix non-determinism in code gen (#101287)

andrewjcg committed 1 year ago

52f526cf

[bazel] add python targets (#101003)

vors committed 1 year ago

630593d3

[quant][pt2] Handle constant conv args in prepare QAT fusion (#100525)

andrewor14 committed 1 year ago

4434b9af

Revert "Mark Windows CPU jobs as unstable (#100581)" (#100676)

huydhn committed 1 year ago

3f734c58

Fix cuda graphs & sdpa for dropout==0 (#101280)

eellison committed 1 year ago

7e333fe5

Disable conv cache emptying (#101038)

eellison committed 1 year ago

a8ff647e

[FSDP]Skip unshard call during checkpointing for NO_SHARD sharding strategy (#101095)

zhaojuanmao committed 1 year ago

5ac48eb3

implement a function to convert a storage to copy-on-write (#100819)

dagitses committed 1 year ago

aec11b8c

ASAN: fix use-after-free (#101064)

AlekseiNikiforovIBM committed 1 year ago

f0f700e8

Fixed a bug in interpolate uint8 AVX2 on non-contig input (#101136)

vfdev-5 committed 1 year ago

a3700571

[BE] super small comment update to gradcheck.py (#101103)

janeyx99 committed 1 year ago

4a7ee79b

[optim][BE] split test file into logical parts: SWA, LR, optim (#101100)

janeyx99 committed 1 year ago

a53cda1d

Revert "[dynamo 3.11] enable other torch 3.11 dynamo-related tests (#99180)"

pytorchmergebot committed 1 year ago

a64e97b6

[ROCM] Add build ROCM support to build-triton-wheel.yml (#95142)

dllehr-amd committed 1 year ago

8e542180

Fix deserialization for UpsamplingBilinear2d (#101248)

tfsingh committed 1 year ago

dfd822d7

Don't set_current_node in DDP. (#101046)

ezyang committed 1 year ago

fa40195f

[dynamo] Skip tests that are broken in fbcode (#101217)

bertmaher committed 1 year ago

d54fcd57

Newer Older