strisunshinewentingwang/pytorch

Pull Requests Commits

Support neg calls to dyn shapes (#94068)

voznesenskym committed 2 years ago

ab95b7a0

fix grain size setting for baddbmm_cpu_kernel (#98297)

mingfeima committed 2 years ago

11890156

Add PrivateUse1 for dispatching PyTorch Distributed Collectives. (#98137)

shaoyf42 committed 2 years ago

cc5f6495

Clean up CUDA 11.6 Docker images (#98395)

huydhn committed 2 years ago

d3adbbf4

[BE] Fix `collect_env` for python-path-with-space (#98415)

malfet committed 2 years ago

bd785320

[EASY] Fix some more places where we incorrectly assume only Tensor (#98310)

ezyang committed 2 years ago

680bf14a

Disable persistent reductions with dynamic shapes (#98405)

ezyang committed 2 years ago

478df47f

[CI] Update update_expected.py to make it generate a combined csv file (#98407)

desertfire committed 2 years ago

007587aa

[quant][pt2e][fix] Fix the internal test failures caused by refactor (#98378)

jerryzh168 committed 2 years ago

a7611483

Add check for 0 to 1 inclusive for elements of target tensor in BCE loss (#97814)

kiersten-stokes committed 2 years ago

2a48f43f

Export function symbols to enable Windows build of Intel Extension for PyTorch (#98054)

z71258847 committed 2 years ago

3112d2a2

[inductor] Move `tl.broadcast` call out codegen.common (#98304)

peterbell10 committed 2 years ago

013c7f5b

[inductor] Enable CSE on masked loads (#98303)

peterbell10 committed 2 years ago

bb4174d2

rewrite at::vec::*::convert_to_int_of_same_size (#98429)

mikey dagitses committed 2 years ago

aa7850c2

Forward fix for DataLoader to accept custom Sharding DataPipe (#97287)

ejguan committed 2 years ago

29d2e4b7

Add detect_fake_mode (#98321)

ezyang committed 2 years ago

d01ee10b

Extract ExtraMeta symbolic shape fields into a dedicated SymbolicShap… (#98399)

ppiskorski committed 2 years ago

5854923c

Add slow workflow to upload test stats workflow (#98447)

clee2000 committed 2 years ago

5af47dbb

[Easy] Fix minor errors in DTensor examples (#98430)

mrshenli committed 2 years ago

d0eafed7

[Dynamo] Support typing.Union and typing.Optional (#98384)

yanboliang committed 2 years ago

b1c29254

Add HPU to the storage tensor backends (#98404)

sujoysaraswati committed 2 years ago

846415f6

[MPS] Add `random_` overload (#98333)

malfet committed 2 years ago

29cde007

[c10d] delete lengths offset checks (#98368)

wanchaol committed 2 years ago

c9b1e099

[Dynamo] Fix bug of torch.is_floating_point & is_complex (#98393)

yanboliang committed 2 years ago

9c7b03d5

Add hacky example inputs to dynamo produced graph (#96561)

tugsbayasgalan committed 2 years ago

ebeaf8ad

Fix default argument of `parse_ir` stub (#98397)

take-cheeze committed 2 years ago

3d36f6f1

[pt2][inductor] retry add `triton.__verison__` as cache key, update cache layout (#98369)

nmacchioni committed 2 years ago

3c8e9e38

Python Dispatcher should respect FuncTorchBatchedDecomposition key (#98328)

zou3519 committed 2 years ago

f21a176c

Patch release process description (#98425)

atalman committed 2 years ago

78e991e5

Require sequence length in huggingface to be dynamic (#98335)

ezyang committed 2 years ago

37b91432

Newer Older