strisunshinewentingwang/pytorch

Pull Requests Commits

Revert "PyTorch -> C++17 (#98209)"

Catherine Lee committed 2 years ago

39a62ea1

Revert "Migrate jobs from windows.4xlarge windows.4xlarge.nonephemeral instances (#100377)"

pytorchmergebot committed 2 years ago

543b7ebb

Migrate jobs from windows.4xlarge windows.4xlarge.nonephemeral instances (#100377)

jeanschmidt committed 2 years ago

7caac545

Move pattern match for foreach before bulky if-else in `save_variables` (#100445)

crcrpar committed 2 years ago

311c2bb7

Revert "Mount /etc/hosts into container (#100475)"

pytorchmergebot committed 2 years ago

e8a1d0be

[dynamo][moco] Disallow_in_graph distributed APIs (#100071)

anijain2305 committed 2 years ago

5fbb4066

[c10d] Add new Store methods: append, multi_get, multi_set. (#100379)

Rodrigo Kumpera committed 2 years ago

0dc671c2

PyTorch -> C++17 (#98209)

r-barnes committed 2 years ago

8f0c825d

ci: win cpu test -> trunk, cuda test -> periodic (#100478)

seemethere committed 2 years ago

50b0fff0

[BE] Update cutlass with NVIDIA upstream changes to 3.1 (#100333)

Skylion007 committed 2 years ago

0efab604

enable headdims > 64 for flash attention on sm90 (#99776)

Fuzzkatt committed 2 years ago

06bf5d4d

[pt2] add `SymInt` support for `dsplit`, `hsplit`, `vsplit` (#100352)

nkaretnikov committed 2 years ago

279f3cd0

Add size check before calling stack_.at(dict_pos) in unpickler.cpp (#94300)

m4drat committed 2 years ago

794e3971

Use yaml.SafeLoader instead of legacy yaml.Loader (#100443)

kit1980 committed 2 years ago

ab65bac3

Add error_inputs_sparse method to OpInfo (#100389)

pearu committed 2 years ago

02a0fb8d

Replace master with main in links and docs/conf.py (#100176)

Svetlana Karslioglu committed 2 years ago

d425da8b

Support expandable_segments:True in fbcode for caching allocator

zdevito committed 2 years ago

0aac2446

Mount /etc/hosts into container (#100475)

malfet committed 2 years ago

99ded8bb

Revert "[functorch] test for compiling functorch transforms (#100151)"

pytorchmergebot committed 2 years ago

af92fc1c

Initial version of Dynamo capture for HigherOrderOperator (#99988)

zou3519 committed 2 years ago

4c99f9cd

Refactor OutputGraph (#99987)

zou3519 committed 2 years ago

984a2397

Revert "[pytorch] Accelerate indexing_backward_kernel with duplicates (#99441)"

pytorchmergebot committed 2 years ago

1114673c

[inductor] Remove redundant model copy when running with cpp_wrapper (#100275)

desertfire committed 2 years ago

ec3c8abb

[export] Migrate internal verifier to subclass export/verifier

angelayi committed 2 years ago

Verified af62d098

[pt2] add `SymInt` support for `tensordot` and `inner` (#100356)

nkaretnikov committed 2 years ago

41361538

[distributed][sharded_tensor] Move local_shards check from ShardedTensorBase to ShardedTensor (#100197)

IvanKobzarev committed 2 years ago

4582ceb2

Make backend_accuracy_fails suppress errors in same_two_models (#100324)

ezyang committed 2 years ago

8556cf20

Run minifier tests same process when possible (#100416)

ezyang committed 2 years ago

054a254b

Prevent Triton from getting eagerly imported when importing torch._inductor (#100374)

ezyang committed 2 years ago

f093ee17

Speed up minifier tests by applying some configs that speed things up. (#100387)

ezyang committed 2 years ago

74cc3771

Older