strisunshinewentingwang/pytorch

Pull Requests Commits

[BE] unified run_process_no_exception code (#49774)

Rong Rong (AI Infra) committed 3 years ago

71766d89

torch.xlogy: Use wrapped_scalar_tensor / gpu_with_scalars to speed up GPU kernel. (#49926)

gchanan committed 3 years ago

74dcb6d3

[pytorch] add threshold_backward batching for vmap (#49881)

kaiwenw committed 3 years ago

483670ff

Add trace batching forward/backward rule (#49979)

ejguan committed 3 years ago

da790eca

Make use_c10_dispatcher: full mandatory for structured kernels (#49490)

ezyang committed 3 years ago

0216366f

Move default or no default logic into native.argument (#49489)

ezyang committed 3 years ago

6c833efd

codegen: Resolve overload ambiguities created by defaulted arguments (#49348)

ezyang committed 3 years ago

8eee8460

Tighten up error checking on manual_kernel_registration (#49341)

ezyang committed 3 years ago

7202c0ec

Construct CppSignatureGroup from NativeFunction (#49245)

ezyang committed 3 years ago

8e20594b

.circleci: Ignore unbound variables for conda (#50053)

seemethere committed 3 years ago

f0945537

Enforce c10-fullness for all ops (#49619)

smessmer committed 3 years ago

69ca5e13

move to non-legacy magma v2 headers (#49978)

t-vi committed 3 years ago

6e84a018

Improve `torch.flatten` docs and add tests to test_view_ops (#49501)

soulitzer committed 3 years ago

fdb81c53

Update update_s3_htmls.yml (#49934)

ppwwyyxx committed 3 years ago

b76822eb

Run test_type_hints first (#49748)

malfet committed 3 years ago

22bd2778

Add type annotations to _tensorboard_vis.py and hipify_python.py (#49834)

guilhermeleobas committed 3 years ago

211f3563

Making ops c10-full: list of optional tensors (#49138)

smessmer committed 3 years ago

c7e9abb6

Back out "[pytorch][PR] Preserve memory format in qconv op" (#49994)

Akshit Khurana committed 3 years ago

e44b2b72

[c10/**] Fix typos (#49815)

SamuelMarks committed 3 years ago

8aad66a7

Remove flops warnings from the default profiler use case (#49896)

Ilia Cherniavskii committed 3 years ago

749f8b78

Revert D25734450: [pytorch][PR] Improve `torch.flatten` docs and add tests to test_view_ops

soulitzer committed 3 years ago

de3d8f8c

Fix inf norm grad (reland) (#48611)

soulitzer committed 3 years ago

4677fc69

Improve `torch.flatten` docs and add tests to test_view_ops (#49501)

soulitzer committed 3 years ago

730965c2

Revert D25719980: [pytorch][PR] Accept input tensor with 0-dim batch size for MultiLabelMarginLoss

ngimel committed 3 years ago

cd608fe5

[PyTorch] Decouple version numbers from c10 and caffe2 targets (#49905)

iseeyuan committed 3 years ago

46afd7fc

[quant] Quantizable LSTM (#49671)

z-a-f committed 3 years ago

04a8412b

quant docs: add common errors section (#49902)

vkuzo committed 3 years ago

ffbb68af

Remove incorrect usage of layout(std430) on uniform buffers, correctly now treated as error in the latest release of Vulkan SDK. (#49572)

Ashkan Aliabadi committed 3 years ago

a7e1f4f3

Fix a KaTeX crash and many docstring issues (#49684)

rgommers committed 3 years ago

6a951a6f

Accept input tensor with 0-dim batch size for MultiLabelMarginLoss (#46975)

v0dro committed 3 years ago

6b56b71e

Newer Older