strisunshinewentingwang/pytorch

Pull Requests Commits

[FSDP2] Added mixed precision (#118223)

awgu committed 222 days ago

8c1480f5

[FSDP2] Added autograd/memory/overlap/frozen/2D/AC tests (#118136)

awgu committed 222 days ago

3956ce01

[dynamo] Capture untyped_storage().resize_() (#119647)

jansel committed 222 days ago

39c68efd

[DDP] Change the --no-optimize-ddp flag to reflect the latest usage (#119437)

fegin committed 222 days ago

c0e5cca4

Prevent DCE'ing unbacked SymInt for view outputs (#119552)

ezyang committed 222 days ago

c2522554

Avoid performing replacements when it would unrefine ranges (#117356)

ezyang committed 222 days ago

52de407b

fix torch.cumsum docs (#117944)

jeejeelee committed 222 days ago

0fd371c8

[inductor] Refactor device guard Python codegen to allow nested indentation (#119673)

aakhundov committed 222 days ago

c2a835d7

Fix typo in private attr of `inference_mode` (#119167)

toslunar committed 222 days ago

f4b5f710

Implement analysis for for-loops (#119730)

oulgen committed 222 days ago

36292871

caffe2: remove support for specifically running "flaky tests" (#112007)

asm89 committed 222 days ago

2ae655b4

[EZ] Set maximum supported version of Python as 3.12 (#119743)

malfet committed 223 days ago

60148f17

improve cuda graph symint logging msg (#119739)

eellison committed 223 days ago

beb00413

fix compile DTensor.from_local in trace_rule_look up (#119659)

wanchaol committed 223 days ago

bfb9ea1a

Skip log line if no tensors were dedupped (#119742)

mvpatel2000 committed 223 days ago

379183a0

[BE] Use more GTest primitives in XPU unit tests (#119527)

guangyey committed 223 days ago

a4c476a0

Fix C++20 build (#112333)

cyyever committed 223 days ago

47a2e6b6

[DTensor] Only wait on AsyncCollectiveTensor after DTensor-based state dict loading (#119716)

yoyoyocmu committed 223 days ago

2bda6b4c

Linear-BN Fusion: add precondition check (#119264)

min-jean-cho committed 223 days ago

2502a011

[MPS] Enable `conj` and `conj_physical` (#119669)

malfet committed 223 days ago

15ef52a0

Revert "Add Accelerator device and shell hooks (#119329)"

pytorchmergebot committed 223 days ago

214f06ae

Revert "[BE] Properly mark destructor overrides (#119656)"

pytorchmergebot committed 223 days ago

7d4b6668

[mypy] refactor mkldnn_fusion._is_valid_binary to avoid [union-attr] has no attribute (#119085)

ColinPeppler committed 223 days ago

2921c2b3

[Lint] replace [assigment] with [method-assign] for methods (#119706)

SandishKumarHN committed 223 days ago

db228f1e

Revert "Add missing include for internal build (#119721)"

pytorchmergebot committed 223 days ago

9f8c84a3

Support FunctoolsPartialVariable::get_function, fix NamedTupleVariable::as_proxy and handle call_function in get_fake_values_from_nodes (#119435)

laithsakka committed 223 days ago

ea8e4fd5

[dynamo] Support torch.distributed.fsdp._flat_param._same_storage_size (#119627)

jansel committed 223 days ago

74d55b0e

Revert "Avoid performing replacements when it would unrefine ranges (#117356)"

pytorchmergebot committed 223 days ago

472500e3

Revert "Improve TORCHDYNAMO_EXTENDED_DEBUG for GuardOnDataDependentSymNode (#119412)"

pytorchmergebot committed 223 days ago

2492f874

[quant][pt2] Fix _disallow_eval_train error message (#119694)

andrewor14 committed 223 days ago

830ed6d9

Newer Older