strisunshinewentingwang/pytorch

Pull Requests Commits

Re-do D48544397: [TGIF Inplace] [xlv2][1/n] Expose a couple APIs from inline_container that will be used for chunk read" (#109183)

sayitmemory committed 2 years ago

a6fadf64

AOTInductor dynamic shape (#109012)

Yang Chen committed 2 years ago

9cd4548f

[export] Preserve shape dynamism for unused inputs. (#109239)

zhxchen17 committed 2 years ago

f4e96df6

[FSDP][Wrap] ModuleWrapPolicy callable (#109117)

rohan-varma committed 2 years ago

25bf1a49

[FSDP] continue if param not exist in sharded load (#109116)

rohan-varma committed 2 years ago

f558e86f

[ONNX] bump ort-nightly==1.16.0.dev20230908001 (#109212)

abock committed 2 years ago

68987544

Fix CUDA-12 wheel loading on AmazonLinux (#109244)

malfet committed 2 years ago

90068ab3

Revert "Support SymBool input to torch.compile (#107850)"

pytorchmergebot committed 2 years ago

47f79e9a

Revert "Remove expected test failures for cond (#108709)"

pytorchmergebot committed 2 years ago

de76c88d

Reformat line of code header to put co_name after (#109233)

ezyang committed 2 years ago

05170b0b

[quant][be] Add TestPT2ERepresentation test case (#108923)

jerryzh168 committed 2 years ago

c914ca75

Support register_hook on input tensors (#108903)

voznesenskym committed 2 years ago

064ae9ff

[inductor][easy] Enable mypy checking for all inductor files that already pass (#109238)

masnesral committed 2 years ago

50a08407

Disable cutlass tests in fbcode (#109241)

ipiszy committed 2 years ago

acad84ba

[ez][inductor][fx passes] quick fix for invalid nodes (#109234)

jackiexu1992 committed 2 years ago

62732bdc

[export] Normalize nn_module_stack paths. (#109231)

zhxchen17 committed 2 years ago

5edbee94

Support str() on user defined functions (#108973)

nakulcr7 committed 2 years ago

109ab6a0

Remove expected test failures for cond (#108709)

ydwu4 committed 2 years ago

a08e1370

Support SymBool input to torch.compile (#107850)

ydwu4 committed 2 years ago

9f6d70b2

[export] Separate out exported_program.py (#109147)

angelayi committed 2 years ago

025d1a18

[inductor] Parallelize Max Autotune step 2: Use multiple GPUs (#109127)

masnesral committed 2 years ago

4a09ed54

[inductor] Parallelize Max Autotune step 1: refactor autotune_process (#109126)

masnesral committed 2 years ago

ce428393

Remove include_0d from sample_inputs_gather (#109125)

guilhermeleobas committed 2 years ago

dbddf181

Update take_along_dim docs to include `dim=None` case (#109120)

guilhermeleobas committed 2 years ago

61f05787

Dispatch `numpy.take_along_axis` to `torch.take_along_dim` (#108880)

guilhermeleobas committed 2 years ago

d046376c

Add `SymInt` support to `torch.take_along_dim` (#108879)

guilhermeleobas committed 2 years ago

49e3d766

Fix failing inductor test (#109220)

drisspg committed 2 years ago

aca3bd44

Better error handling for cond (#108817)

ydwu4 committed 2 years ago

33c94b8b

Revert "add Half support for BatchNorm on CPU (#102070)"

pytorchmergebot committed 2 years ago

04a765f9

Disable tests mentioned in 109213 (#109232)

clee2000 committed 2 years ago

c44f8169

Newer Older