strisunshinewentingwang/pytorch

Pull Requests Commits

[don't merge] Test doc previews

driazati committed 3 years ago

6088f145

Revert D30735341: Migrate uses of THCReduceApplyUtils to cuda_utils::BlockReduce

ngimel committed 3 years ago

a97015f2

[fx] make const fold code more pythonic (#64451)

Yinghai Lu committed 3 years ago

b1215060

[quant] Enable jit tracing on quantizable LSTM (resubmission) (#64638)

Zafar Takhirov committed 3 years ago

24e1315d

Factor out TensorBase that doesn't depend on native operators (#63612)

peterbell10 committed 3 years ago

d701357d

Make doc previews use its own S3 bucket (#64594)

driazati committed 3 years ago

92318a91

TST Adds inplace checks to module_info (#63739)

thomasjpfan committed 3 years ago

43c0f033

Migrate uses of THCReduceApplyUtils to cuda_utils::BlockReduce (#64442)

peterbell10 committed 3 years ago

a5ad08ec

.github: Run docker containers in detach mode (#64459)

seemethere committed 3 years ago

deb9775c

[NNC] Add Softplus operator (#64589)

anijain2305 committed 3 years ago

18d24bb5

Add `__matmul__` to the magic methods for FX tracing (#64512)

Chillee committed 3 years ago

35413a16

update scatter formula (#64546)

kshitij12345 committed 3 years ago

195cb4ef

fixing trapezoid() comments for clarity (#64592)

NivekT committed 3 years ago

1409492f

Add forward mode differentiation for torch.linalg.cholesky and transpose (#62159)

IvanYashchuk committed 3 years ago

dd8f6ac5

Fix typo embedding_renorm_cuda_ (#64542)

Leo7654 committed 3 years ago

a2934b38

[c10d] Provide failure reason from ProcessGroup when aborting NCCL comm (#64241)

rohan-varma committed 3 years ago

e0e832c2

Change MaxUnpool to accept tensors with 0-dim batch sizes. (#64082)

v0dro committed 3 years ago

7205ca02

Add Half conversion of bit cast for SYCL kernel (#64340)

chengjunlu committed 3 years ago

ba8c1fc6

[nnc] Provide helpful error messages about turning off the fuser (#64516)

bertmaher committed 3 years ago

7f0feafa

Allow disabling cache in autocast (automatic mixed precision) (#63552)

leslie-fang-intel committed 3 years ago

768014b3

Adding support for lowering 4Bit EmbeddingBag Operator (#5806)

Protonu Basu committed 3 years ago

b6161324

restore test_inplace_comparison_ops_require_inputs_have_same_dtype Expected behavior (#64267)

Freey0 committed 3 years ago

2223737d

[quant] AO migration of the `quantize.py` (resubmission) (#64445)

Zafar Takhirov committed 3 years ago

9cc44aad

[TensorExpr] Don't rely on exceptions in Vectorizer. (#64609)

Mikhail Zolotukhin committed 3 years ago

72274e2a

[fx_const_fold] Fix constant folding for attrs in submodule hierarchies (#64342)

jfix71 committed 3 years ago

2341ec9e

Add __ge__ to TorchVersion (#64565)

Rikorose committed 3 years ago

57212054

add out variant of linear (#61801)

Maksim Levental committed 3 years ago

81fe2c5e

Fix building docs instructions (#64508)

Stevenjin8 committed 3 years ago

71ba76b1

Fix quicklint (#64612)

malfet committed 3 years ago

4e98304e

Revert D29998114: [pytorch][PR] enable bf16 mkldnn path for gemm

ngimel committed 3 years ago

e777e1b0

Older