strisunshinewentingwang/pytorch

Pull Requests Commits

[composite compliance] add test for fwd AD

kshitij12345 committed 2 years ago

c1ced8ff

ROCm: Enable test_distributed_spawn

jaglinux committed 2 years ago

81586a6a

ROCm: unskip c10 gloo tests

jaglinux committed 2 years ago

60e2ee39

[DataPipe] Add funtional API docstring and fix typo in test

ejguan committed 2 years ago

0ff05b1e

Make -h work with run_test.py

albanD committed 2 years ago

3d7abc0e

Improve more the error message with explicit recommendation

albanD committed 2 years ago

438cc79f

[shard] Sharder and ShardingPlan prototype (#73873)

wanchaol committed 2 years ago

78ea86a4

[FSDP] Fix exec order validation (static variable issue)

awgu committed 2 years ago

f7e4b77a

Initial prims, references, and test architecture for them (#75095)

Mike Ruberry committed 2 years ago

28c3e0f7

[PyTorch] Add native fast path for transformer encoder inference

swolchok committed 2 years ago

b369b89f

Rename tools/codegen to torchgen (#76275)

ezyang committed 2 years ago

36420b5e

[ONNX] Support restricted quantized range for activation.

BowenBao committed 2 years ago

8d31706b

[ONNX] Support per channel quantization

BowenBao committed 2 years ago

cada2cd3

[Distributed tests] Add skip for odd world_size condition

jaglinux committed 2 years ago

6ca8272d

Improving typing and typing-related performance in rendezvous.py

ishaan-mehta committed 2 years ago

d65ab9a6

Fixed typo in documentation of `LazyModuleMixin`

sugatoray committed 2 years ago

f980c3c1

Fix minor typos

meyavuz committed 2 years ago

89370e00

bypassed cublasLtMatMul bug after slicing (#76205)

xiaosun86 committed 2 years ago

a5ffdaf0

[Model Averaging] Make the error message more informative in hierarchical_model_averager.py

wayi1 committed 2 years ago

e9058039

Revert "stft: remove non-center overload and python functional wrapper"

pytorchmergebot committed 2 years ago

77f23d64

Emit ATen ops when symbolics raise + minor fixes

Thiago Crepaldi committed 2 years ago

90d31cb3

[nnc] Strides to Tensor (#72962)

IvanKobzarev committed 2 years ago

93906092

Remove incorrect xfail

peterbell10 committed 2 years ago

1a7e43be

Enable simple indexing into CSR tensor, add torch.select for CSR

IvanYashchuk committed 2 years ago

407e8eba

[Pytorch][3/4 Static dispatch] Move static dispatch logic to Operators.cpp (#76058)

priyaramani committed 2 years ago

5622c8b4

Make LTC codegen customizable enough for XLA migration (#76180)

wconstab committed 2 years ago

9d8ff02e

Remove cuda 11.5 builds since we have 11.6

atalman committed 2 years ago

31a6e6ca

stft: remove non-center overload and python functional wrapper

peterbell10 committed 2 years ago

6b7d89c4

Remove F.pad python implementation

peterbell10 committed 2 years ago

cb37e7a0

[fx] Don't use __module__ to test if a function is bound from C++

peterbell10 committed 2 years ago

7a80fc2c

Newer Older