strisunshinewentingwang/pytorch

Pull Requests Commits

Update sgd doc to insist on momentum buffer initial value (#92111)

albanD committed 1 year ago

60e37a6e

Improve type hints for Module forward hooks (#92061)

ringohoffman committed 1 year ago

a26e5e21

Add kwargs support to torch.export() API (#92013)

Thiago Crepaldi committed 1 year ago

890b6828

Add check-sparse-tensor-invariants flag to Context - 2nd try. (#92094)

pearu committed 1 year ago

b3e4f502

[dynamo] support comparing numpy ndarray (#91870)

chunyuan-w committed 1 year ago

a111dd90

[ONNX] Fix potential flaky test in test_verification.py (#92105)

BowenBao committed 1 year ago

fa3841ff

[quant][fx] Add support for GRU in fx graph mode quantization (#91976)

jerryzh168 committed 1 year ago

ec3941ad

[Quant][docs] Move parts of BackendConfig tutorial (#91999)

andrewor14 committed 1 year ago

0bd3fa3d

[Inductor Perf CI] Enable perf CI smoke test (#92051)

weiwangmeta committed 1 year ago

a617d031

unify reduction types from different operators: scatter, scatter_reduce, segment_reduce (#91499)

mingfeima committed 1 year ago

eb7b8977

[vision hash update] update the pinned vision hash (#92119)

pytorchmergebot committed 1 year ago

a70387f0

Update dynamic skips after #92076 (#92103)

ezyang committed 1 year ago

fbbb1959

Fix OneCycleLR error log (#92040)

milesial committed 1 year ago

9412778d

Switch Windows CI jobs to G5 runners (#91727)

huydhn committed 1 year ago

61cdae0c

[DTensor] require DeviceMesh size equals world size (#91801)

XilunWu committed 1 year ago

b7cad020

[DTensor] create default process group when absent (#91756)

XilunWu committed 1 year ago

3dd9dbd9

Revert "Make ModuleList derive from Sequence[T] and type it appropriately (#89135)"

pytorchmergebot committed 1 year ago

f8e641ba

[quant][pt2e] Add a test to confirm we can set qconfig according to module_name (#91977)

jerryzh168 committed 1 year ago

8fa66a63

Fixes to DSA infra (#91835)

r-barnes committed 1 year ago

6f749fd1

Limit the memory and CPU of Bazel build to avoid crashing the runner (#92056)

huydhn committed 1 year ago

4636fe70

Reland "AOT Autograd refactor + cleanup, handle intermediate views of bases, use view replay, fix non-tensor input handling" (#92076)

ezyang committed 1 year ago

7078ad5b

fix in-place geometric pmf (#92049)

min-jean-cho committed 1 year ago

da77b10b

Fixed output memory format mismatch for bicubic2d (#90470)

vfdev-5 committed 1 year ago

5f55335c

[JIT] Add tests for pow() with different dtype inputs (#91946)

davidberard98 committed 1 year ago

c4a6f21b

[functorch] move batch_norm_replacement to torch.func (#91412)

samdow committed 1 year ago

515dff78

Revert "[FSDP] Do not clean FQNs even for `use_orig_params=True` (#91767)"

pytorchmergebot committed 1 year ago

7bdcf6d4

sparse_mask: remove redundant mask.coalesce() in to_dense_backward (#92001)

nikitaved committed 1 year ago

91920ee6

Fixup torch jit with some initializers and moves (#92037)

Skylion007 committed 1 year ago

b9182cbb

generate set_device call to ensure context existence (#92055)

ngimel committed 1 year ago

5625f521

[Inductor] Support vectorized transpose in CPP backend (#91532)

jgong5 committed 1 year ago

7c641eaa

Newer Older