strisunshinewentingwang/pytorch

Pull Requests Commits

PCA and SVD for low-rank matrices, LOBPCG for positive-defined generalized eigenvalue problem - copy (#34721)

pearu committed 4 years ago

8bae1ed1

Revert D20251830: [TensorExpr] Add tensorexpr benchmarks.

Mikhail Zolotukhin committed 4 years ago

976d6aaa

caffe2::OperatorBase do not need to be aware of at::Tensor functions (#34810)

malfet committed 4 years ago

ef78fa86

[TensorExpr] Add tensorexpr benchmarks. (#34230)

Mikhail Zolotukhin committed 4 years ago

e93e7b27

[TensorExpr] Add LLVM codegen. (#34228)

Mikhail Zolotukhin committed 4 years ago

ea5c86c2

[TensorExpr] Add CUDA codegen. (#34227)

Mikhail Zolotukhin committed 4 years ago

35e7efeb

[TensorExpr] Add a fuser pass based on tensor expressions. (#34226)

Mikhail Zolotukhin committed 4 years ago

42b2c8c6

[TensorExpr] Pull changes to core classes for representing expressions and statements from the side branch. (#34224)

Mikhail Zolotukhin committed 4 years ago

e31d462e

[fix][tiny][caffe2] Avoid triggering errors when allow ratio is 100% (#34757)

Xinyi Zhang committed 4 years ago

99b91ee2

Enable JIT tests on Windows (#27029)

peterjc123 committed 4 years ago

24c9e61e

Initial implementation of NNPI Int8FC op

Yinghai Lu committed 4 years ago

1af60023

[jit] copy unused/ignored methods to ScriptModule during compilation (#33981)

suo committed 4 years ago

a57f92e4

[quant][graphmode] Add quantization pattern for quantized::add_relu (#33532)

jerryzh168 committed 4 years ago

cec9758a

Remove unused newWithSize declaration. (#34730)

gchanan committed 4 years ago

8eaafbd9

Remove unused newView declaration. (#34729)

gchanan committed 4 years ago

b94d6508

Migrate dirichlet_grad from CUDA_tensor_apply4 to TensorIterator (#33996)

zasdfgbnm committed 4 years ago

a66b837b

Migrate binary_cross_entropy_backward from CUDA_tensor_apply4 to (#33995)

zasdfgbnm committed 4 years ago

c3c0cf15

[C++ API Parity] [Optimizers] added closure to optimizers (#34790)

anjali411 committed 4 years ago

762be86e

[C++ API] RNN / GRU / LSTM layer refactoring (#34322)

Will Feng committed 4 years ago

bdd7dbfd

Add overloaded name to prim operators (#34280)

iseeyuan committed 4 years ago

d4f182d0

Removes unused THCTensor_(triu), THCTensor_(div) (#34712)

Mike Ruberry committed 4 years ago

c86d1361

solve conv3d backward get incorrect result problem (#34358)

XiaobingSuper committed 4 years ago

c258e473

Move min and max(reduce all) to Aten(CPU) (#33936)

XiaobingSuper committed 4 years ago

7848c229

Disallow sending CUDA tensors over RPC for current RPC agents. (#33604)

pritamdamania committed 4 years ago

f058c03b

CUDA Loops: move address computation into policy, make policy.load load all arguments (#33720)

zasdfgbnm committed 4 years ago

f404537c

Fix backward compatibility check test for schemas containing (#34782)

kimishpatel committed 4 years ago

528aabd3

[PyTorch BC] Clean up the BC whitelist (#34784)

houseroad committed 4 years ago

15c84c37

Remove unnecessary import (#34778)

tgrlnc committed 4 years ago

08bc3c6c

Export roi_align_gradient_op to c10 (#34776)

xiaomengy committed 4 years ago

1d81bd02

Fix missing header (#34762)

Yinghai Lu committed 4 years ago

373c80ee

Newer Older