strisunshinewentingwang/pytorch

Pull Requests Commits

[JIT] remove builtin interpolate functions (#34514)

Elias Ellison committed 4 years ago

514cba06

Fix _cat operator (#34591)

VitalyFedyunin committed 4 years ago

962e3624

Prohibit copying autograd engines (#34567)

malfet committed 4 years ago

a22008f9

Replace THPLayout with at::Layout in Python Argument Parser (#34543) (#34584)

terhuhf committed 4 years ago

3c76b2ae

fix the quantized batchnorm2d (#34579)

lly-zero-one committed 4 years ago

f70945b1

[jit] kill script namespace (#34515)

suo committed 4 years ago

c235be42

Delete OperatorOptions, absorb AliasAnalysisKind into FunctionSchema. (#34588)

ezyang committed 4 years ago

cf8b7282

Fix typo in data.rst (#34624)

samkellerhals committed 4 years ago

b039bca4

[PT] add macro to expose caffe2 ops to PyTorch mobile (#34578)

linbinyu committed 4 years ago

2fe7fc68

[net_transform] only skip ConstantFill for autogen_grad (#34628)

xianjiec committed 4 years ago

0dc0fffc

Remove cudaMemcpy on full memory overlap (#34548)

zasdfgbnm committed 4 years ago

86fb522a

Fix for handling batch size 0. (#34599)

kimishpatel committed 4 years ago

adb8e261

[C++ API] Update torch::nn layer docs (#34522)

Will Feng committed 4 years ago

9064fafb

[JIT] Add support for tolist for GPU-resident Tensors (#34554)

Meghan Lele committed 4 years ago

56832bf7

[ci] try to fix rocm builds (#34600)

suo committed 4 years ago

866505b1

Fix typo in documentation (#34581)

kvthr committed 4 years ago

2de4f245

[On-device Benchmark] speed_benchmark_torch switch to log latency from dataset level to row level (#34598)

Peng Xia committed 4 years ago

25e4e9eb

Fix SELECTED_OP_LIST file path issue (#33942)

gemfield committed 4 years ago

70f32986

[JIT] Torchbind error if python instantiate class that doesnt exist (#34568)

James Reed committed 4 years ago

1f834b5c

Disable ROCM when building mobile libtorch. (#34478)

syoyo committed 4 years ago

12fb8148

[distributed] quicker exit in the case of failed tests in distributed (#34150)

rohan-varma committed 4 years ago

b553e691

small typos (#34589)

ettiee committed 4 years ago

2cf576e9

Stop last usage of newWithSize. (#34386)

gchanan committed 4 years ago

82cdd3ab

Revert D20193196: [pytorch][PR] PCA and SVD for low-rank matrices, LOBPCG for positive-defined generalized eigenvalue problem

ezyang committed 4 years ago

4b929e54

Revert D20282846: Delete OperatorOptions, absorb AliasAnalysisKind into FunctionSchema.

ezyang committed 4 years ago

6f8a8e4e

Revert D20379910: [pytorch][PR] Set USE_RCCL cmake option (dependent on USE_NCCL)

ezyang committed 4 years ago

63964175

PCA and SVD for low-rank matrices, LOBPCG for positive-defined generalized eigenvalue problem (#29488)

pearu committed 4 years ago

2ec779d4

Stop using ctypes to interface with CUDA libraries. (#33678)

peterbell10 committed 4 years ago

5fc5cf65

Delete OperatorOptions, absorb AliasAnalysisKind into FunctionSchema. (#34160)

ezyang committed 4 years ago

9d42177a

Beef up documentation on Dispatcher.h, reorder methods for clarity. (#33838)

ezyang committed 4 years ago

b2344b70

Newer Older