strisunshinewentingwang/pytorch

Pull Requests Commits

[AOTInductor] ProxyExecutor support Dynamic Shape (#110526)

SherlockNoMad committed 1 year ago

f1b94461

Add 9.0a to cpp_extension supported compute archs (#110587)

dzhulgakov committed 1 year ago

a0cea517

Bump pillow from 9.5.0 to 10.0.1 in /.ci/docker (#110494)

dependabot[bot] committed 1 year ago

Verified c89d35ad

Add requirement for input to AllGatherIntoTensor to be contiguous (#109561)

ani300 committed 1 year ago

efdf1553

Fix typo in BatchLinearAlgebraLibBlas.cpp (#110608)

eltociear committed 1 year ago

f21c322e

Quieter logs in CI (#110033)

clee2000 committed 1 year ago

d6e5898e

pin_memory support for NT (#110404)

jbschlosser committed 1 year ago

3597325b

[HigherOrderOp] fallthrough some keys by default. (#110478)

ydwu4 committed 1 year ago

cc1de493

Enable aarch64 for fixing undefined symbol error. (#110542)

jasonjk-park committed 1 year ago

26f634ee

[ROCm] conditionally enable hipsparse const descriptors for version >= 2.4.0 (#110317)

jeffdaily committed 1 year ago

a94b6f39

Made pattern-matcher diagnostics lazily reported + added TORCH_COMPILE_CPROFILE (#110504)

Chillee committed 1 year ago

f767a6c5

Revert "Made pattern-matcher diagnostics lazily reported + added TORCH_COMPILE_CPROFILE (#110504)"

pytorchmergebot committed 1 year ago

1e4c0641

[FSDP][optim_state_dict] Make the new optimizer allgather fusion work with fine-tuning models (#110540)

fegin committed 1 year ago

1a729618

Multiprocessing support for NT (#110292)

jbschlosser committed 1 year ago

f17fe89e

Back out "Enable pickling model prepared with QAT qconfig" (#110392)

andrewor14 committed 1 year ago

7c72238e

[AOTInductor] Store loaded kernels in the model (#110554)

khabinov committed 1 year ago

cf1b494a

`torch::nn::AdaptiveLogSoftmaxWithLoss`: check length of `cutoffs` (#106777)

Sehun0819 committed 1 year ago

c36b31d5

[vision hash update] update the pinned vision hash (#110571)

pytorchupdatebot committed 1 year ago

00b9afa4

export db links for user errors (#110555)

avikchaudhuri committed 1 year ago

416eca97

Revert "[Dynamo] SizeVariable can be indexed by symint (#110349)"

pytorchmergebot committed 1 year ago

21019620

[quant][pt2] Support cudnn_batch_norm in QAT fusion (#109908)

andrewor14 committed 1 year ago

62cad5b5

[dynamo] [easy]Remove InstructionTranslator from within Set (#110521)

lezcano committed 1 year ago

4b1e1381

[export] Add ir spec (#110394)

angelayi committed 1 year ago

a93337ed

One more small Perf Tweak to fill_ (#110294)

drisspg committed 1 year ago

a8653f35

Fix typo under torch/_inductor directory (#110530)

kiszk committed 1 year ago

434a996c

Made pattern-matcher diagnostics lazily reported + added TORCH_COMPILE_CPROFILE (#110504)

Chillee committed 1 year ago

9648df1a

Consider that ops can be fused into cat in the min-cut partitioner (#110501)

Chillee committed 1 year ago

e686341f

Include `onnx` and `onnxscript` information in collect_env.py (#110560)

justinchuby committed 1 year ago

d24e7be2

Fix type promotion of float8_e5m2 and float8_e4m3fn (#110279)

australopitek committed 1 year ago

653f966d

[aotinductor] Enable test_non_default_cuda_device on CI (#110509)

desertfire committed 1 year ago

c121f957

Newer Older