strisunshinewentingwang/pytorch

Pull Requests Commits

[state_dict][3/N] Cleanup StateDictOptions, make it more readable (#111108)

fegin committed 1 year ago

b1db9590

Remove some patterns from PrimTorch merge rules (#111230)

nkaretnikov committed 1 year ago

625a3b1a

[dynamo] Register einops functions lazily (#110575)

peterbell10 committed 1 year ago

6aa91c8d

[dynamo] Add specialized variable tracker for sys.modules (#110990)

peterbell10 committed 1 year ago

8747e4c8

[CI] Add auto label rule for torch/_export (#111181)

desertfire committed 1 year ago

058cb70a

[sparse] Add padding for dense matrices in semi-structured sparse (#110583)

jcaip committed 1 year ago

8db72a43

Revert "Remove dead code (#111207)"

pytorchmergebot committed 1 year ago

2b6f281e

Revert "AOTAutograd: Go down inference path if no outputs require grad (#111011)"

pytorchmergebot committed 1 year ago

cf6b1cdf

[Doc] Fix typo in cpp/installing when wheel is used (#111143)

malfet committed 1 year ago

d84dcfb3

Remove dead code (#111207)

lezcano committed 1 year ago

c2ed714f

[state_dict][2/N] Let distributed.state_dict accepts single optimizer (#111107)

fegin committed 1 year ago

e99abaae

[dynamo] fix prim lowering validation logic for dynamic shape args (#111208)

jon-chuang committed 1 year ago

ac768333

[DCP] Improve with_temp_dir robustness (#111106)

fegin committed 1 year ago

247d5e16

[CUDA][cuFFT] Initialize CUDA context for cuFFT before execute is called (#110326)

eqy committed 1 year ago

5a2ab7dc

Revert "Move at::{Refcounted,}MapAllocator to c10 (#109881)"

pytorchmergebot committed 1 year ago

f68d6e81

Fix typo under c10 directory (#111155)

kiszk committed 1 year ago

8162f417

Fix typo under torchgen directory (#111154)

kiszk committed 1 year ago

ac48c11a

[BE] Enable Ruff's Flake8 PYI042 (#111114)

isdanni committed 1 year ago

b460c308

[pt][group_fusion] fix shape guarding in fusion candidate search (#111174)

chaekit committed 1 year ago

5db9f911

[PyTorch] AOTI: generate reused thread_locals when tensors provably have static shape (#110892)

swolchok committed 1 year ago

84975339

[PyTorch] AOTI: Add aoti_torch_assign_tensors to ABI (#110909)

swolchok committed 1 year ago

bf72a723

[dynamo] Forward fix a bunch of distributed collective allow fixes (#111171)

voznesenskym committed 1 year ago

cff71c47

Add `lazy_clone_storage` to create COW storages (#110192)

kurtamohler committed 1 year ago

33f15134

[export] Fix issue with internal model (#111140)

angelayi committed 1 year ago

35750bf9

[C10D] Introduce C++ side Collective Callbacks. (#110307)

wconstab committed 1 year ago

359336e3

Improve reflection_pad2d lowering for dynamic shapes (#110988)

ezyang committed 1 year ago

d24539ee

[inductor] Implement Fx graph caching to improve warm compilation time. (#103453)

masnesral committed 1 year ago

0dfa3545

[Dynamo]Expose bytecode hooks and add example usage for decompilation in docs (#110714)

youkaichao committed 1 year ago

69dcbc02

Fix mkldnn_matmul error on AArch64 (#110150)

imzhuhl committed 1 year ago

cdc8d709

Fixed description of run_on input for linux-binary-test workflow (#111191)

DanilBaibak committed 1 year ago

24bd3301

Newer Older