strisunshinewentingwang/pytorch

Pull Requests Commits

Change watchdog timeout logging from INFO to ERROR. (#50455)

pritamdamania committed 3 years ago

4e248eb3

[StaticRuntime][ATen] Add out variant for narrow_copy (#49502)

Hao Lu committed 3 years ago

4e766167

Caffe2 Concat operator benchmark (#50449)

Marat Subkhankulov committed 3 years ago

49896c48

[Pytorch Mobile] Remove caching (in code) of interned strings (#50390)

dhruvbird committed 3 years ago

af968cd6

Type annotations in test/jit (#50293)

r-barnes committed 3 years ago

8c25b970

Create subgraph rewriter (#49540)

Ansley Ussery committed 3 years ago

4c97ef8d

Add type annotations to torch.nn.modules.padding (#49494)

guilhermeleobas committed 3 years ago

374951d1

[te] Create TargetMachine only once with correct options to fix perf (#50406)

bertmaher committed 3 years ago

cb37709b

[quant][refactor] Minor refactor of some typos (#50304)

z-a-f committed 3 years ago

7d28f1c8

[quant][bug] Fixing the mapping getter to return a copy (#50297)

z-a-f committed 3 years ago

39aac654

Automated submodule update: tensorpipe (#50441)

facebook-github-bot committed 3 years ago

412e3f46

[package] better error message when unpickling a mocked obj (#50159)

suo committed 3 years ago

50744cd0

fixing autodiff to support Optional[Tensor] on inputs (#49430)

jjsjann123 committed 3 years ago

6d947067

Stop moving scalars to GPU for one computation in leaky_rrelu_backward. (#50115)

gchanan committed 3 years ago

c198e6c6

Clean up some type annotations in test/jit/...../test_class_type.py (#50156)

r-barnes committed 3 years ago

cf45d65f

Check CUDA kernel launches in caffe2/caffe2/utils/math (#50238)

jessijzhao committed 3 years ago

725640ed

[vmap] Add batching rules for comparisons ops (#50364)

RockingJavaBean committed 3 years ago

5cdc32bf

Fix MultiheadAttention docstring latex (#50430)

jankrepl committed 3 years ago

b2f7ff7d

Add Post Freezing Optimizations, turn on by default in torch.jit.freeze (#50222)

Elias Ellison committed 3 years ago

a389b30b

Peephole Optimize out conv(x).dim(), which prevents BN fusion (#50221)

Elias Ellison committed 3 years ago

30aeed7c

[JIT] Factor out peephole to own test file (#50220)

Elias Ellison committed 3 years ago

a69f008c

[JIT] Add Frozen Conv-> Add/Sub/Mul/Div fusion (#50075)

Elias Ellison committed 3 years ago

69711493

[JIT] Frozen Graph Conv-BN fusion (#50074)

Elias Ellison committed 3 years ago

035229c9

[PyTorch] Devirtualize TensorImpl::sizes() with macro (#50176)

swolchok committed 3 years ago

b5d38269

Add new patterns for ConcatAddMulReplaceNaNClip (#50249)

ShijunK committed 3 years ago

158c98ae

Enable fast pass tensor_fill for single element complex tensors (#50383)

anjali411 committed 3 years ago

58344380

Disable complex dispatch on min/max functions (#50347)

imaginary-person committed 3 years ago

6420071b

add type annotations to torch.nn.modules.normalization (#49035)

guilhermeleobas committed 3 years ago

4411b5ac

Added linalg.pinv (#48399)

IvanYashchuk committed 3 years ago

9384d31a

Fix Error with torch.flip() for cuda tensors when dims=() (#50325)

dheerajgattupalli committed 3 years ago

314351d0

Newer Older