strisunshinewentingwang/pytorch

Pull Requests Commits

[torch export] add source_fn_stack metadata in non-strict mode (#121909)

strisunshinewentingwang committed 312 days ago

2f04a852

Add privateuseone flags for c10::EventFlag (#121118)

chentianyi16 committed 312 days ago

0e68eb15

[aoti] Fix compilation bug for buffer mutations (#121688)

angelayi committed 312 days ago

9f314d4a

[AOTInductor] Include build cmds at the end of wrapper file (#121872)

SherlockNoMad committed 312 days ago

0636c118

[sigmoid] Use deserializer from oss. (#121839)

zhxchen17 committed 312 days ago

c4092921

[Inductor] Fix a dynamic shape problem when lowering diagonal (#121881)

desertfire committed 312 days ago

499136a4

[with_effects] Skip over profiler.record_function_exit (#121829)

angelayi committed 312 days ago

5b164251

[ez] Document for add_var_to_val (#121850)

SherlockNoMad committed 312 days ago

f1f7c5c3

[BE] Add S3 bucket argument to number of workflows (#121907)

jeanschmidt committed 312 days ago

4c3a052a

[FSDP2] Added 2D DCP save/load test (#121747)

awgu committed 312 days ago

38d7d366

[c10d] Add generic scuba logging capability into c10d (#121859)

shuqiangzhang committed 312 days ago

443444dc

Add CUTLASS kernel as choice for (u)int8/(b)float16 mixed MM autotuning (#119986)

alexsamardzic committed 312 days ago

83f8e514

relax tol for flaky nansum_out_dtype_cuda_float32 test (#121550)

jeffdaily committed 312 days ago

be0bdf11

Checkout release branch rather then commit_hash when building triton release (#115379) (#121901)

atalman committed 312 days ago

7e13b5ba

[Fix] Fixed behaviour for the conversion of complex tensors to bool (#121803)

andoorve committed 312 days ago

956059fa

Add CUTLASS kernel as choice for _int_mm() Inductor autotuning (#119685)

alexsamardzic committed 312 days ago

1251f0fa

Make PyTorch compilable against upcoming Numpy-2.0 (#121880)

malfet committed 313 days ago

38d9bb5a

Do not compile FP16 arith internally (#121844)

malfet committed 313 days ago

b4c53aa0

Handle transitive replacements in Triton kernel mutation analysis (#121867)

aakhundov committed 313 days ago

3eb322ff

Enable FX graph cache for a batch of inductor tests (#121696)

masnesral committed 313 days ago

4cd503c1

Graph break on step closure in optimizer (#121777)

mlazos committed 313 days ago

15abc56b

Fix quantized linear vulkan tests (#120960)

andre-meta committed 313 days ago

f85f58bf

[Quant][Inductor] Enable quantization linear pattern fusion with int8_mixed_bf16 for gelu (#116004)

Le-Zheng committed 313 days ago

a37caa6e

[Quant][Inductor] Enable quantization linear pattern fusion for gelu inside inductor (#114854)

Le-Zheng committed 313 days ago

43d68e9c

[Quant][PT2E] Enable linear and linear-unary post-op gelu quant recipe for x86 inductor quantizer (#114853)

Le-Zheng committed 313 days ago

25e00545

Use memcache versioning for autotune remote cache (#121748)

oulgen committed 313 days ago

a04e7fca

[C10D] Fix coalescedCollective op Flight Recording (#120430)

wconstab committed 313 days ago

7e076c75

Revert "[export] allow Dim(1,2) for export dynamic shapes (#121642)"

pytorchmergebot committed 313 days ago

bf7ac4dd

Only FA2 doesn't support attn-mask (#121825)

drisspg committed 313 days ago

3e02a7ef

[export] allow Dim(1,2) for export dynamic shapes (#121642)

pianpwk committed 313 days ago

a8dcbf27

Older