strisunshinewentingwang/pytorch

Pull Requests Commits

[LTC] Remove view (#88445)

alanwaketan committed 1 year ago

c29502dd

Do not use double for single-prec upsample (#88277)

malfet committed 1 year ago

f2000842

Fix typos used in documents under torch directory (#88483)

kiszk committed 1 year ago

4ea2310f

[Reland] Use sudo when reset NVIDIA devices (#88605)

huydhn committed 1 year ago

d25be63c

Implement a constructor for nested_tensor that is similar to torch.tensor() (#88213)

Antoni Viros i Martin committed 1 year ago

c77368d4

Pin linux ninja dep to 1.10.2 (#88548)

huydhn committed 1 year ago

72a73519

Use test/test-reports for inductor (#88533)

huydhn committed 1 year ago

fdf28651

Fix segfault in has_torch_function (#88559)

peterbell10 committed 1 year ago

eb3f975c

Fix pull docs build running with a schedule and increase cpp doc timeout to 4h (#88589)

huydhn committed 1 year ago

4796e23b

Add a note on the stability of linalg functions. (#88313)

lezcano committed 1 year ago

d453b3c4

Revert "fallback for scatter_(scalar) (#88210)"

pytorchmergebot committed 1 year ago

b00c43b3

Dynamo Dashboard Improvements (#88516)

williamwen42 committed 1 year ago

0e67b2f7

(fix): Add some missing std::moves to C10 (#88512)

Skylion007 committed 1 year ago

b14e0650

Generalize gesvdjBatched to run whith full_matrices==false (#88502)

lezcano committed 1 year ago

d8506ff4

[DataPipes] Add group support to the sharding_filter (#88424)

VitalyFedyunin committed 1 year ago

9dadf8fc

SymIntify _copy functionalization kernels (and _copy_out too) (#88572)

ezyang committed 1 year ago

23a3eb37

fallback for scatter_(scalar) (#88210)

Krovatkin committed 1 year ago

896fa8c5

Publicly expose _LRScheduler to LRScheduler (#88503)

janeyx99 committed 1 year ago

0a69c50a

Upload test stats for inductor workflow (#88535)

huydhn committed 1 year ago

05b9e8ec

[torchdynamo] support torch.autograd._profiler_enabled (#88378)

yuguo68 committed 1 year ago

a3752408

Handle pin_memory in refs.randn (#88473)

SherlockNoMad committed 1 year ago

95d57b54

[nvfuser] skip extremal tests on rocm (#88587)

suo committed 1 year ago

bf49dada

Revert "Use sudo when reset NVIDIA devices (#88531)"

pytorchmergebot committed 1 year ago

7bf9db81

Revert "[fix] allow saving python attr on Tensor and Parameter via torch.save (#81616)"

pytorchmergebot committed 1 year ago

78a0ca29

[exir][fx] PassManager error handling (#88520)

angelayi committed 1 year ago

91a40398

[Dynamo] Fix bug: GradMode doesn't carry grad state correctly after graph break (#88537)

yanboliang committed 1 year ago

bd1ffc65

[2/n] Thread PG: add class _World to distributed_c10d.py (#781) (#88471)

Rodrigo Kumpera committed 1 year ago

6663ae55

Clarify rules for which commit is used in CI (#88425)

ZainRizvi committed 1 year ago

fc8f2f66

Upgrade Linux NVIDIA driver to the latest prod version (#88517)

huydhn committed 1 year ago

c407a7b2

Use sudo when reset NVIDIA devices (#88531)

huydhn committed 1 year ago

505486ce

Newer Older