strisunshinewentingwang/pytorch

Pull Requests Commits

[Inductor] Provenance tracking for wrapper code (#105717)

SherlockNoMad committed 1 year ago

a44f8894

Revert "Correct dynamo logging docs (#105658)"

pytorchmergebot committed 1 year ago

050d3de0

[pytorch] Disable new autograd fallback for mobile builds (#105750)

zou3519 committed 1 year ago

4d5d4d8b

[FSDP][Easy] nit follow-ups to handle refactor (#105738)

awgu committed 1 year ago

221853af

Correct dynamo logging docs (#105658)

davidradl committed 1 year ago

f3a261e0

[C10D] Remove watchKey functionality from the Store. (#105014)

Rodrigo Kumpera committed 1 year ago

174b0c22

Bump pygments from 2.12.0 to 2.15.0 in /.ci/docker (#105654)

dependabot[bot] committed 1 year ago

9d2f56fd

Improve fake mode support by adding fake_context to ExportOutput (#105247)

Thiago Crepaldi committed 1 year ago

999ca07e

add lerp cpu support for half (#105607)

janeyx99 committed 1 year ago

803d42e4

[ONNX] Refactor AvgPool to support dynamic shapes (#105683)

titaiwangms committed 1 year ago

d5d6eb2d

[BE] f-stringify torch/ and scripts (#105538)

justinchuby committed 1 year ago

4cc1745b

[Dynamo] Enable torch._dynamo.config.suppress_errors by default (#105307)

yanboliang committed 1 year ago

4c73016f

[BE] Enable ruff's UP rules in pyproject.toml (#105437)

justinchuby committed 1 year ago

de8bd108

[8/n][FSDP] make use_dtensor=True work with offload_to_cpu=True for optim.load_state_dict() (#105690)

wz337 committed 1 year ago

6b2d48e7

[Inductor] Optimize read write merging in FusedSchedulerNode ctor (#105693)

mlazos committed 1 year ago

72b223cd

Allow (temporarily?) non-fake input during ONNX export with fake mode (#105246)

Thiago Crepaldi committed 1 year ago

842616bc

Improve basicAutogradNotImplementedFallback + new tests (#105591)

zou3519 committed 1 year ago

04da0c76

Fix Tensor::register_hook behavior on undefined tensors (#105587)

zou3519 committed 1 year ago

ed6de455

Kill process in `wait_for_process` if `SIGINT` fails to terminate it (#105625)

eqy committed 1 year ago

29f856e3

[Inductor] Replace functools.reduce union calls with set unions (#105720)

mlazos committed 1 year ago

ec26947c

[BE] Enable ruff's UP rules and autoformat nn/ mps/ and torch/ (#105436)

justinchuby committed 1 year ago

79c5e333

Skip test_cudnn_rnn when cudnn not available (#105701)

ezyang committed 1 year ago

322dff47

Simplify handle indexing (#105006)

voznesenskym committed 1 year ago

429d45f9

[dynamo] Fix import if numpy is not installed (#105711)

angelayi committed 1 year ago

b0a04331

[inductor][fx passes] batch layernom (#105492)

jackiexu1992 committed 1 year ago

e40f8ace

Rename some sizevars methods for clarity (#105585)

ezyang committed 1 year ago

a01a7329

[dynamo][numpy] Add support for builtin len() on numpy ndarray (#105691)

larryliu0820 committed 1 year ago

cce2b7e3

Update core aten decomp table (#105673)

angelayi committed 1 year ago

fed8d360

Skip deterministic mode for SAM (#105615)

msaroufim committed 1 year ago

c759a570

Revert "Add torch.utils to the docs page, remove dead code and fix docstrings (#105142)"

pytorchmergebot committed 1 year ago

11732586

Newer Older