strisunshinewentingwang/pytorch

Pull Requests Commits

[Modes] refactor modes to only use a stack in cpp (#86458)

samdow committed 1 year ago

169ec120

[BE] Remove pip and conda installation in Linux build workflow (#87256)

huydhn committed 1 year ago

13cad7e1

Slowly introduce ops to be tested by test_numpy_ref on MPS backend (#87342)

aslvrstn committed 1 year ago

620dbc43

[1/N][C10D] Add a customized ScubaLogHandler implementation for internal FB use (#86699) (#87123)

wz337 committed 1 year ago

7bd04fb0

Only label checks against pull requests (#87488)

ZainRizvi committed 1 year ago

100beb20

fix for dynamo xml reporting (#87378)

clee2000 committed 1 year ago

2a6079d5

ci: Allow nvidia-smi to continue with non-0 exit (#87464)

seemethere committed 1 year ago

6e1764d8

functionalization: make view_copy outputs always contiguous (#85747)

bdhirsh committed 1 year ago

9ad1659b

Create workflow to make sure PRs have valid labels (#86829)

neelrpatel committed 1 year ago

294bfb8e

Skip auto request review on forked PR (#87482)

huydhn committed 1 year ago

fbcd4fe2

Remove redundant zeroing in col2im/im2col (#87375)

peterbell10 committed 1 year ago

5b7f027d

Grammatical update of the tech docs. (#87357)

chuksmbaka committed 1 year ago

4fc72b0f

Add dynamo smoke test (#87400)

williamwen42 committed 1 year ago

6efdcb07

[inductor] force 'fork' method for processes, cleanup (#87411)

zdevito committed 1 year ago

db83a057

[dynamo] Unify raise_on_* config to suppress_errors and raise by default (#87440)

ezyang committed 1 year ago

96691865

[FSDP][2/N] Fix grad zero vs. `None` edge case (#87308)

awgu committed 1 year ago

1133682c

[FSDP][1/N] Update `summon_full_params(with_grads)` `None` gradient (#87314)

awgu committed 1 year ago

4ee13a59

[quant][api] Add assert for backend in get_default_qconfig related apis (#86259) (#87331)

jerryzh168 committed 1 year ago

4caddac5

[FSDP][6/N] Remove FPW! (#87114)

awgu committed 1 year ago

4cc5d664

[FSDP][5/N] Update `FlatParamHandle` after FPW deprecation (#87113)

awgu committed 1 year ago

f8dd2742

[FSDP][4/N] Rework FPW test to not use FPW (#87112)

awgu committed 1 year ago

214d5175

[FSDP][3/N] Register `flat_param` to wrapped module (#87086)

awgu committed 1 year ago

277e37f9

[FSDP][2/N] Remove `_fsdp_wrapped_module.flat_param` (#86122)

awgu committed 1 year ago

9f8ef8ea

Reland "add an API for external backends to register custom device names (#86992)" (#87453)

bdhirsh committed 1 year ago

ce0c6e82

Fix input dimension issue in RNN, LSTM, GRU error message (#87442)

jyx-su committed 1 year ago

70c46d32

Revert "Back out "Revert D40198461: [pytorch][PR] Backport currently dont work with some models if:" (#87124)"

pytorchmergebot committed 1 year ago

0c1dec37

Audit for error prone isinstance int/float and add lint (#87345)

ezyang committed 1 year ago

d73d4aa7

OpInfo: Add test that sample_inputs_func returns a generator (#84567)

peterbell10 committed 1 year ago

1285542f

Reenable `isinstance` with `torch.distributed.ReduceOp` (#87303)

crcrpar committed 1 year ago

aa8248cc

Make LazyGraphExecutor extensible (#87218)

antoniojkim committed 1 year ago

d37dc6f6

Newer Older