strisunshinewentingwang/pytorch

Pull Requests Commits

Implement tensor.refine_names (#25842)

zou3519 committed 5 years ago

e544f885

Add fusion for quantized linear (#25624)

jerryzh168 committed 5 years ago

94964a9b

Automatic update of fbcode/onnx to 95252c2adec185e305e34486c6756ece9aa8f57f (#26137)

houseroad committed 5 years ago

e9e7e9d4

Create TensorBoard test classes in all cases (#26005)

orionr committed 5 years ago

ff7921e8

Add device check before accessing data_ptr in PackLayer (#26056)

Ailing Zhang committed 5 years ago

3acab233

Port fuse_linear from pytorch/tvm (#25623)

jerryzh168 committed 5 years ago

be82239c

C++ unregister_module function for Module (#26088)

ShahriarSS committed 5 years ago

18a0040f

Support quantizing any methods called (#25505)

jerryzh168 committed 5 years ago

1d870900

Kill kwarg_only declarations in Declarations.cwrap. (#25609)

gchanan committed 5 years ago

5fce7696

Fix build warning in vec256_qint.h

James Reed committed 5 years ago

e2e1f5ef

Implementation of ConstantThenLinearWarmupLRPolicy and CompositeCyclicalLRPolicy (#25970)

Taiqing Wang committed 5 years ago

784c4a91

Skip inserting duplicate observers (#25504)

jerryzh168 committed 5 years ago

f559c1d8

fix base_lr overridden in cyclic lr (#26105)

vincentqb committed 5 years ago

135bbc26

Stop reordering TH random function arguments.

gchanan committed 5 years ago

f9a8b8ad

remove "build_deps" arg from setup.py command in (#26113)

rohan-varma committed 5 years ago

369064fa

change gradle build to use static libtorch + gc-sections (#25984)

ljk53 committed 5 years ago

ffee507d

simplify build_android_gradle.sh (#25897)

ljk53 committed 5 years ago

fbc038ab

Kill TH(C)Blas kwarg_only declarations. (#25607)

gchanan committed 5 years ago

771cb628

Stop re-ordering TH(C)Blas arguments. (#25606)

gchanan committed 5 years ago

ad91d028

tracing with an opt-in by file name (#25895)

Krovatkin committed 5 years ago

1eae6355

make sure all out stringstreams start out empty in jit_log.hpp

Krovatkin committed 5 years ago

f9289949

Exposing Fused8BitRowwiseQuantizedToFloat in PyTorch (#26080)

Qi Zhou committed 5 years ago

076eaf4c

Skip test_triangular_solve_batched (#26108)

vishwakftw committed 5 years ago

f91fbf90

Automatic update of fbcode/onnx to 7988d8360b11e6003560076e9b1d4aa426db3244 (#25959)

houseroad committed 5 years ago

7e4ac8b8

TorchScript Serialization for dynamic LSTM

James Reed committed 5 years ago

bdc656da

Disable test_cuda.test_stream_event_nogil on ROCm (#26087)

bddppq committed 5 years ago

827d71d7

print source code when a function is executed (#25868)

Krovatkin committed 5 years ago

f3fdbba6

Experimental warning for named tensors (#26050)

zou3519 committed 5 years ago

4fb5a7c5

Move NamedTensorMetaInterface definitions to TensorImpl.h (#26030)

zou3519 committed 5 years ago

03bb7969

Make regular softmax warp size aware (#25956)

iotamudelta committed 5 years ago

a996b1d6

Newer Older