strisunshinewentingwang/pytorch

Pull Requests Commits

Extend nn.Transformer to support BERT (gelu) (#24181)

Meteorix committed 5 years ago

0cc92de4

implement bool_tensor.bernoulli_ (#25076)

ssnl committed 5 years ago

6100de9b

Use C10_DEPRECATED_MESSAGE instead of TORCH_WARN_ONCE for Tensor.data<T>() (#25319)

Will Feng committed 5 years ago

5248dd1a

Move new_criterion_tests from test_nn.py to common_nn.py (#25333)

Will Feng committed 5 years ago

80974dde

Remove unused THTensor_(add) and similar functions code. (#24864)

VitalyFedyunin committed 5 years ago

d0a525b5

Fix typo "takes takes" -> "takes"

zasdfgbnm committed 5 years ago

509abd9a

Disable cuda_distributions_test and converter_nomigraph_test on Windows. (#25305)

ezyang committed 5 years ago

53ac931a

Support all_reduce a list of same-device tensors #21640 (#24949)

jfc4050 committed 5 years ago

590619ab

Migrate erfinv and erfinv_ from the TH to Aten (CUDA) (#24943)

ifedan committed 5 years ago

afb7a162

Port `pow` operator from the TH code to Aten (#23492)

pbelevich committed 5 years ago

112f2494

vainaixr committed 5 years ago

d7cce323

Migrate digamma\digamma_\polygamma\polygamma_ from the TH to Aten (CPU) (#25048)

ifedan committed 5 years ago

9b109795

Revert D17052534: [pytorch][PR] Creates Torch-friendly Event class and adds Stream tracking to autograd

ezyang committed 5 years ago

529bb859

Implementation of cpu_serial_kernel for TensorIterator (#25125)

ifedan committed 5 years ago

e123e24e

Added documentation for nn.functional.bilinear (#24951)

Patrick Donnelly committed 5 years ago

883628cb

Align AT_FORALL macros with DISPATCH macros wrt Half. (#25268)

gchanan committed 5 years ago

fe541aab

Fix infer np scalar dtype mem leak (#24267)

ssnl committed 5 years ago

6c9410ff

Disable the copy constructor and = operator of DispatchStub (#24932)

xuhdev committed 5 years ago

dfa48f99

Remove some unused plugins.

gchanan committed 5 years ago

45943bd6

Mirwaisse committed 5 years ago

687aa781

upgrade MKL-DNN to v0.20.3 (#22910)

gujinghui committed 5 years ago

718feb6d

Work around for bias quantization for conv and linear operators (#25212)

raghuramank100 committed 5 years ago

9945c0ce

Return a message instead of void from rpc udf (#25283)

satgera committed 5 years ago

a74d702e

Revert D17059087: [quant] Reducing the test size for adaptive avg pool

ezyang committed 5 years ago

98beb9ec

int8 static quantization in the numerical debugger

csummersea committed 5 years ago

febcb3b7

Upgrade the deprecated data to data_ptr APIs (#25295)

jianyuh committed 5 years ago

c7ef50bd

Add the sparse feature information during logging in sparse lookup layer (#24863)

Xing Wang committed 5 years ago

8a8844dc

Revert D17063240: [fix] Specify width for st.floats in hypothesis_utils.tensor

Zafar Takhirov committed 5 years ago

b1f7e13d

improve interface error messages (#25228)

zdevito committed 5 years ago

ca4bc9fc

add serialization of interface (#25227)

zdevito committed 5 years ago

fba107f1

Newer Older