strisunshinewentingwang/pytorch

Pull Requests Commits

check sparse sizes (#47148)

ngimel committed 3 years ago

4a2fb340

[quant] Add _FusedModule type to capture all fused modules for quantization (#47484)

jerryzh168 committed 3 years ago

65e5bd23

Add complex autograd support for torch.mean (#47566)

anjali411 committed 3 years ago

8339f883

Make gen_op_registration flake8 compliant (#47604)

malfet committed 3 years ago

3d962430

Batching rule for Tensor.new_empty_strided (#47226)

zou3519 committed 3 years ago

b80da898

Implement Tensor.new_empty_strided(sizes, strides, *, dtype, device, requires_grad) (#47225)

zou3519 committed 3 years ago

59aca022

[caffe2] Fix duplicate name bug in Net.AddExternalInput (#47530)

dongyuzheng committed 3 years ago

4a58f35b

Revert D24801481: [pytorch][PR] Add AcceleratedGraphModule and serialzie GraphModule to JSON

malfet committed 3 years ago

6248e062

Add AcceleratedGraphModule and serialzie GraphModule to JSON (#47233)

gcatron committed 3 years ago

9e0102c1

[PyTorch Mobile] Don't use __ROOT__ for inference only ops

iseeyuan committed 3 years ago

8182558c

[pytorch] continue to rewrite gen_python_functions.py with typed models (#46978)

ljk53 committed 3 years ago

16c72a5a

Add docs on how to toggle TF32 flags on C++ (#47331)

zasdfgbnm committed 3 years ago

4a7de274

Support RRef.backward() for Owner RRefs. (#46641)

pritamdamania committed 3 years ago

781e0ed8

Support the strided tensor on input for torch.cat (#46859)

lly-zero-one committed 3 years ago

5a5258cb

[ONNX] Reimplement _var_mean to ensure non-negative (#47240)

jiafatom committed 3 years ago

6e69a24a

The dimension being reduced should not be coalesced by TensorIterator (#47237)

zasdfgbnm committed 3 years ago

f23a2a11

Correctly print out sign of near-zero double values (#47081)

tugsbayasgalan committed 3 years ago

29184f86

BoxWithNMSLimit support int `batch_splits` input (#47504)

842974287 committed 3 years ago

c19eb4ad

Implement Complex tensor support in all reduce and all gather (#47523)

gmagogsfm committed 3 years ago

9d0c6e94

Add complex support for torch.mean [CUDA] (#47048)

RockingJavaBean committed 3 years ago

f90da88d

Enable diag for bool Tensors (#47455)

H-Huang committed 3 years ago

451e7d3d

Add bool tensor support for where (#47454)

H-Huang committed 3 years ago

3253ccbd

Support extra files in _load_for_mobile (#47425)

Martin Yuan committed 3 years ago

a1fef453

Correctly compare Stream IValues (#47303)

tugsbayasgalan committed 3 years ago

3f9697b1

Build nightly binaries only for the latest ROCM (#47503)

malfet committed 3 years ago

25d1fb51

Update the error message for retain_grad (#47084)

iramazanli committed 3 years ago

e09ec8ee

Fix rounding error flakiness in quantized_test (#47468)

ajliu committed 3 years ago

7af9752f

[JIT] add support for torch.jit.Final in python 3.6 (#47393)

Lilyjjo committed 3 years ago

63778779

Back out "[c10] make intrusive_ptr available as a pybind holder type"

wanchaol committed 3 years ago

31d041c9

Add support for log_softmax (#47409)

navahgar committed 3 years ago

8eb228a7

Newer Older