strisunshinewentingwang/pytorch

Pull Requests Commits

Add error repro test for FSDP ignored modules with mixed precision (#89971)

mrshenli committed 1 year ago

5a36d998

add vjp test with non-contig inputs (#89375)

kshitij12345 committed 1 year ago

dfb533ca

Type torch._dynamo.guards (#89919)

ezyang committed 1 year ago

99dac4dd

Guarantee symbol allocation for all sizes/strides/storage offset (#89879)

ezyang committed 1 year ago

e03cde07

Add definitely_not_01 set to ShapeEnv. (#89871)

ezyang committed 1 year ago

74bcf2b6

When dealing with dupe arguments, prefer leafifying if possible (#89896)

ezyang committed 1 year ago

8d333761

[FSDP][Dynamo] Define annotation attributes as globals (#89913)

awgu committed 1 year ago

808cb2e8

remove torch.equal usages (#89527)

pmeier committed 1 year ago

4095ef8b

fix assert_close docstring (#89620)

pmeier committed 1 year ago

0acbcef4

document torch.testing.assert_allclose (#89526)

pmeier committed 1 year ago

d72cd4c4

enable ufmt for torch/testing/*.py (#89525)

pmeier committed 1 year ago

4baa78bb

ad more error info for cublasLtMatmul (#89983)

zhaojuanmao committed 1 year ago

850b53bb

Add manual meta implementations to quantize_per_tensor.tensor and co (#89958)

ezyang committed 1 year ago

a7473264

add mixed data type support for LayerNorm (#81851)

mingfeima committed 1 year ago

f1978b18

[vision hash update] update the pinned vision hash (#89749)

pytorchmergebot committed 1 year ago

b6d6c693

[codemod][llvm15] LLVM-15 fixes for caffe2/caffe2/video/video_decoder.cc (#89937)

r-barnes committed 1 year ago

b399acd2

[codemod][llvm15] LLVM-15 fixes for caffe2/caffe2/video/video_decoder.h (#89940)

r-barnes committed 1 year ago

2f5532a9

[codemod][llvm15] LLVM-15 fixes for caffe2/test/cpp/jit/test_graph_executor.cpp (#89936)

r-barnes committed 1 year ago

cc016141

[inductor] Disable parallel compilation inside fbcode (#89926)

bertmaher committed 1 year ago

6317311e

[Vulkan][TCC] Helper functions for vulkan quantized tests (#89922)

manuelcandales committed 1 year ago

8d8a215d

Minor fix for dynamo xla integration test (#89891)

JackCaoG committed 1 year ago

a6145072

quantization: make x86 as default backend (#88799)

XiaobingSuper committed 1 year ago

4bae8608

fix mkldnn quantization issue for weight reorder error (#86876)

XiaobingSuper committed 1 year ago

0e7918b9

RowwiseMoments: use float as acc type for bfloat16 inputs (#84405)

mingfeima committed 1 year ago

6372f11d

[Checkpoint] Minor update to checkpoint utils (#89964)

wz337 committed 1 year ago

ad1585b4

Implement gamma cdf (#89955)

ezyang committed 1 year ago

a43e09c0

Add device note to the docs of sparse tensor factory functions (#89910)

pearu committed 1 year ago

5167108c

Issue 68576 prefetch factor docstring changes (#89874)

blankstatic committed 1 year ago

11db12bd

Correct the label for quantization PRs (#89888)

jerryzh168 committed 1 year ago

7cf09139

[EASY] Replace direct use of Guard ctor with make_guard (#89945)

ezyang committed 1 year ago

1ccaa2a5

Newer Older