strisunshinewentingwang/pytorch

Pull Requests Commits

FX Partitioner: use "all aten ops that decompose to prims" as supported NVFuser ops

davidberard98 committed 2 years ago

17659a7f

Remove flacky test and replace with Mocking test (#80159)

davidchencsl committed 2 years ago

8cfb8f94

Prevent out of bounds access to null LTC operands (#80060)

henrytwo committed 2 years ago

fc6b645f

[c10d] Make allgather as a custom op (#79669)

Jiewen Tan committed 2 years ago

3359af93

Remove obsolete code from build.sh (#80199)

kit1980 committed 2 years ago

bfc2a435

Revert "parse pysymints to IValues (#80066)"

pytorchmergebot committed 2 years ago

0322ecc3

[FakeTensor] Use the device of the meta tensor for fallback kernel (#80193)

eellison committed 2 years ago

c1d11f40

[MPS] Fix binary ops between int32 tensor with int64 scalar (#80220)

malfet committed 2 years ago

a6556efd

FX graph partitioner and fuser (#79439)

SherlockNoMad committed 2 years ago

752c06e0

[ONNX Export] Use `half_pixel` instead of `pytorch_half_pixel`. (#80003)

lazycal committed 2 years ago

86682caf

[GHA] make commit hash pins merge rules more specific (#80233)

janeyx99 committed 2 years ago

9dd16c18

remove unnecessary tuple check on tensor types (#79896)

migeed-z committed 2 years ago

a47cc182

Use generators with all/any in torch/optim (#78142)

kit1980 committed 2 years ago

de7219e8

[MPS] Add L1 loss test (#80010)

qqaatw committed 2 years ago

ff44bfa1

Revert "Add validation for mapper function in datapipes with `input_col` (#79344)"

pytorchmergebot committed 2 years ago

fcdaf351

Remove overly restrictive assert in adam (#80222)

albanD committed 2 years ago

375668cd

[quant] Quantizable documentation (#79957)

HDCharles committed 2 years ago

0308609b

[ao][sparsity] Implemented state_dict() and load_state_dict() functions (#79883)

macandro96 committed 2 years ago

524d1812

[quant] Add QuantizedMHA class (#79956)

HDCharles committed 2 years ago

3bcec850

[ao][sparsity] Implemented the step() function (#79822)

macandro96 committed 2 years ago

af4e2b2c

[ao][sparsity] Base scheduler class for Data Schedulers (#79817)

macandro96 committed 2 years ago

70b7bca4

[primTorch] asinh, atanh (#80210)

khushi-411 committed 2 years ago

b00448df

Cleanup buck_setup.sh a bit (#80198)

malfet committed 2 years ago

2d7d5a75

Revert "[FakeTensor] Use the device of the meta tensor for fallback kernel (#80193)"

pytorchmergebot committed 2 years ago

35268bdc

[ao] Adds framework for InputWeightEqualization Detector (#79916)

vspenubarthi committed 2 years ago

845021db

Support softmax(nested tensor) (#80179)

YifanShenSZ committed 2 years ago

54a1cc52

Revert "Raise warning for unpickable local function (#80140)"

pytorchmergebot committed 2 years ago

79ba65c0

Add forward AD for linalg.det and simplify its backward (#79487)

lezcano committed 2 years ago

42a23596

Raise warning for unpickable local function (#80140)

ejguan committed 2 years ago

4b75b7d3

[Static Runtime] Add Metadata to ProcessedNode depending upon the op type (#79961)

akshayParashar1995 committed 2 years ago

3afc802c

Older