strisunshinewentingwang/pytorch

Pull Requests Commits

janeyx99 committed 2 years ago

15cb7f15

Remove unused impl

janeyx99 committed 2 years ago

7256e4cb

[utils][foreach] Add group tensor by device and dtype util

janeyx99 committed 2 years ago

8c005740

Remove DL_RUNTIME_BUG (#91960)

peterbell10 committed 2 years ago

4806a9e7

[static-runtime] clamp fast_sigmoid result into (0,1) range (#91993)

tenpercent committed 2 years ago

6287bb78

[modes] make python arg parser also check for python key (#91573)

samdow committed 2 years ago

d8e795ec

[Modes] Add assert that the mode isn't already on the stack (#90770)

samdow committed 2 years ago

70283863

[stateless] add weight tying support (#90477)

samdow committed 2 years ago

8b3c4bc4

Add bf16 and change header file include path (#91838)

guangyey committed 2 years ago

e03ac0ee

s/INDCUTOR/INDUCTOR/ (#91885)

ezyang committed 2 years ago

d24324bf

Preventing crashing incase of no network by loading from cache (#91569)

MohammedZ666 committed 2 years ago

84b819d0

enable `conj()` for sparse compressed tensors (#91695)

nikitaved committed 2 years ago

850cf894

hrnet_w18, tts_angular works with dynamic shapes (#91891)

ezyang committed 2 years ago

56ed976e

Support zero dimensions in softmax decompositions (#91322)

davidberard98 committed 2 years ago

d7dc1c2f

replace vec::vec_scalar_t with at::opmath_type (#91086)

jiayisunx committed 2 years ago

afd8dd08

inductor: fix .to(memort_format) issue which doesn't generate right stride (#91948)

XiaobingSuper committed 2 years ago

3790b505

[SDPA] Guard mem efficient attention in deterministic mode (#91979)

drisspg committed 2 years ago

92855a21

[ONNX] Fix 'prim::PackPadded' shape inference (#91829)

BowenBao committed 2 years ago

d540442e

Easy: add instructions for testing pytorch/builder (#91923)

dzhulgakov committed 2 years ago

812d774c

optimize scatter_add performance for gnn usage on CPU (#82703)

mingfeima committed 2 years ago

8f5f15a6

[Inductor] assert generator for random, dropout (#91833)

min-jean-cho committed 2 years ago

364f526b

Implement `torch._foreach_lerp` (#87562)

crcrpar committed 2 years ago

554a796a

Minor doc updates for S3 update procedure (#91978)

ezyang committed 2 years ago

7c907bd8

[CUBLAS][TF32] Change cuBLAS TF32 environment variable to be initialization only (#85859)

eqy committed 2 years ago

19723d75

[mergebot] Fix mergebot allow revert of codev diff (#91975)

clee2000 committed 2 years ago

de4e4c78

Make job names match BUILD_ENVIRONMENT (#91512)

clee2000 committed 2 years ago

6b542147

Revert "[Inductor] Added aten.uniform_ decomp (#90869)"

pytorchmergebot committed 2 years ago

43050b83

Add missing clang-tidy fixes for modernize-use-equals-(default|delete) (#91857)

Skylion007 committed 2 years ago

4dcb10e0

Add check-sparse-tensor-invariants flag to Context. (#90849)

pearu committed 2 years ago

b9a035c1

[vmap] all, any : batching rule (#91966)

kshitij12345 committed 2 years ago

949f25be

Older