strisunshinewentingwang/pytorch

Pull Requests Commits

Add source location to class instantiation error (#24990)

davidriazati committed 5 years ago

43c4b9f2

Add OneCycleLR (#25324)

vincentqb committed 5 years ago

05f1fed6

change LBFGS's default tolerance_grad to 1e-7 (#25240)

lili committed 5 years ago

1b7f7aa1

Revert "Let logical_xor support non-bool tensors." (#25269)

gchanan committed 5 years ago

f362a5a0

don't throw in constant prop (#25270)

Elias Ellison committed 5 years ago

44bd63c7

@albanD's #15219 augmented with SavedVariable::weak_grad_fn_ (#23502)

skrah committed 5 years ago

5dd915cd

contrib-tensorboard: removed external tensorboardX dependency (#25259)

Alexander Melnikov committed 5 years ago

eb2c5930

Include the correct header for make_unique in named tensor headers (#25178)

zou3519 committed 5 years ago

df51cbe3

Implement name inference for torch.matmul (#25177)

zou3519 committed 5 years ago

6f5fe96c

Implement name inference for torch.bmm (#25123)

zou3519 committed 5 years ago

d2719b54

Fix possible deadlock in SharedCache inside a forked child proc (#25158)

ssnl committed 5 years ago

eb756746

Fixes test_equal

z-a-f committed 5 years ago

805cf983

Refactor MinMax observer (#23902)

Haixin Liu committed 5 years ago

06757acb

Fix named tensor test (#25313)

zou3519 committed 5 years ago

e335cc3a

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

Newer Older