strisunshinewentingwang/pytorch

Pull Requests Commits

Eliminate unnecessary std::any(all edges of a node) in an unrolled version of depth first search.

HC Zhu committed 4 years ago

262ae2d5

[quantization] Make FP16 RNN use new prepack op (#34339)

James Reed committed 4 years ago

8a17dc65

[JIT] Introduce BuiltinOpFunction and integrate into torchbind (#34098)

James Reed committed 4 years ago

45a504dd

[JIT] Virtualize Function (#33921)

James Reed committed 4 years ago

60e8615a

[JIT] Move stuff out of class_type.cpp (#33900)

James Reed committed 4 years ago

bb111425

Fixed typos in quantization docs / docstrings (#34182)

dsplabs committed 4 years ago

65bad41c

Back out "[jit] Add type tags to lists/dicts in pickle" (#34406)

Leah Dickstein committed 4 years ago

c5e822b7

Fix overlapping keywords (#34142)

Brandon Green committed 4 years ago

392afb9f

Add the 3d avg pool for video related model (#33339)

lly-zero-one committed 4 years ago

b0479506

[PyTorch BC] Clean up the BC whitelist (#34393)

houseroad committed 4 years ago

d9851602

Fix doc and type hints for "torch.add"; fix deprecated python calls in tests (#33935)

xwang233 committed 4 years ago

ccf6fab6

[Lite Trainer] Add necessary registrations for MNIST model (#33717)

iseeyuan committed 4 years ago

01edb745

Add nhwc memory format test for dropout (#34379)

zasdfgbnm committed 4 years ago

96ca06cf

Reenable large conv tests (#34259)

zasdfgbnm committed 4 years ago

37dfc6c4

Enhance reproducibility documentation (#33795)

duncanriach committed 4 years ago

516a5874

.circleci: Remove macOS builds related to CUDA (#34333)

seemethere committed 4 years ago

079de7f3

.circleci: Update default smoke tests from cuda 10.0 -> 10.2 (#34328)

seemethere committed 4 years ago

2d3f6cbf

[PyTorch] Remove const modifiers from passed by value integers in qbatch_norm_fn (#34378)

malfet committed 4 years ago

5608ffc4

Fix Conv.cpp, &&= is not a C++ operator (#34381)

zasdfgbnm committed 4 years ago

c6ea71b6

[aten] Don't deadlock in IValue::Future impl, tests. (#34099)

jjlilley committed 4 years ago

5f641f93

Add scripts to promote S3 artifacts from test channels to stable channels (#34274)

seemethere committed 4 years ago

0489b8da

[ModelLoading] Use byte encoding for uint8, fp16 etc. instead of int32 (#34343)

Rohith Menon committed 4 years ago

879a90b3

Remove unnecessary assert in autograd engine (#34307)

albanD committed 4 years ago

98afce3c

[Aten] Init container iterators to an unsigned type (#34159)

malfet committed 4 years ago

6d8a0f67

[AMD] Remove num_gpu check for remote execution (#34318)

xw285cornell committed 4 years ago

4c99351d

[aten] remove stmt unreachable, variable never used warnings (#34017)

jspark1105 committed 4 years ago

4872b126

[c10] remove warning attribute does not apply to any entity (#34018)

jspark1105 committed 4 years ago

82a177c0

[RPC] Create local RRef<ModuleInterface> remotely in Python, use it remotely in TorchScript (#34183)

xush6528 committed 4 years ago

17ceb694

Clean up some legacy scalar/empty handling. (#34217)

gchanan committed 4 years ago

a7da4490

Make sure Vec256 int32_t and int16_t loadu temprary arrays are properly initialized (#34281)

xuhdev committed 4 years ago

9c5578fd

Older