summaryrefslogtreecommitdiff
path: root/aten
AgeCommit message (Expand)AuthorFilesLines
2018-02-13Additional sparse Variable fixes (#5203)Sam Gross3-13/+38
2018-02-13Update aten docs (#5197)Zeming Lin3-1398/+2670
2018-02-13 add reduce=True arg to SoftMarginLoss (#5071)li-roy6-25/+103
2018-02-12Improve Variable interface (#5127)Peter Goldsborough1-1/+2
2018-02-12Move EmbeddingBag into ATen (#4856)cpuhrsch4-1/+538
2018-02-12Fix memory corruption in im2col/vol2col based convolution kernels. (#5173)Edward Z. Yang15-32/+117
2018-02-12Expose sparse variable sspaddmm (#5017)Richard Zou4-0/+115
2018-02-09Implement sparse tensor and variable norm(value) (#4882)Richard Zou6-1/+43
2018-02-09Enable scalars. (#5158)gchanan1-1/+1
2018-02-09add reduce=True arg to HingeEmbeddingLoss (#5130)li-roy2-5/+7
2018-02-08Improves Conv*d(Transposed) docs to have correct newline and formatting (#5139)Tongzhou Wang1-4/+14
2018-02-08Initial type hints for function_wrapper (#4947)Richard Zou1-19/+220
2018-02-08Ensure Tensors have storages in resizeNd (#5115)Sam Gross2-2/+2
2018-02-08remove some warning introduced by #2764 (#5104)Lynn1-100/+105
2018-02-07Restore torch.mm behavior for sparse variables (#5077)Richard Zou3-10/+20
2018-02-07Add scalar module tests for common_nn. (#5095)gchanan2-0/+7
2018-02-07Bind Tensor.random_ in ATen for CUDA (#5111)Sam Gross1-3/+4
2018-02-06Check shape instead of number of elements for some losses (#5085)Richard Zou8-23/+23
2018-02-06Fix CPU torch.multinomial with noncontiguous prob tensor (#5093)Tongzhou Wang6-46/+55
2018-02-06change long to int64_t (#5094)Tongzhou Wang3-9/+9
2018-02-06Check that indices and values are on the same device (#5089)Sam Gross1-1/+4
2018-02-06Fixed double memory accesses of several pointwise operations. (#5068)Yongjik Kim1-15/+21
2018-02-06Fix TH compile warnings (#5065)Tongzhou Wang34-182/+70
2018-02-06Implement hinge_embedding_loss as a native function. (#5080)gchanan2-0/+23
2018-02-06Cuda 9.1 is cuda version 9010 not 9100 (#4861)Christian Sarofeen3-3/+3
2018-02-06parallelize vol2col and col2vol of Conv3D with CPU backend (#4824)Lynn1-121/+229
2018-02-06Fix reduction functions not respecting the strides of output when output is c...Richard Zou8-8/+106
2018-02-05Fix topk work size computation (#5053)albanD1-7/+9
2018-02-05Fix maxpool3d / avgpool3d crashs (#5052)Richard Zou5-90/+131
2018-02-05Bring back Tensor::data<__half>() and remove base Tensor::data() template (#5...Peter Goldsborough6-6/+17
2018-02-05fix #5047 (#5048)Tongzhou Wang1-4/+4
2018-02-05Port cuDNN RNN bindings to ATen (#4881)Edward Z. Yang10-37/+1361
2018-02-05add reduce=True argument to MultiLabelMarginLoss (#4924)li-roy6-33/+150
2018-02-05Expose sparse variable addmm, addmm_ (#5016)Richard Zou2-7/+11
2018-02-05Don't allow scalars where vectors are required in mv, addmv, ger, addr. (#5003)gchanan5-8/+104
2018-02-04Move the cudnn include path before system include path (#5026)Simeon Monov1-1/+1
2018-02-02Initial GraphExecutor Implementation. (#4982)Zachary DeVito3-10/+16
2018-02-02Don't allow scalars in torch.dot for Variables. (#4972)gchanan4-4/+16
2018-02-02Fix C FFI extension after moving TH to C++ (#5005)Will Feng4-24/+36
2018-02-02Fix output_nr not incremented correctly (#4812)Tongzhou Wang18-741/+994
2018-02-02make torch.set_num_threads also set MKL threads (take 2) (#5002)Soumith Chintala1-5/+12
2018-02-02Replace async with non_blocking for Python 3.7 (#4999)Peter Goldsborough8-18/+18
2018-02-02Improve CUDA softmax performanceAdam Paszke1-7/+18
2018-02-01Add size checks for sparse tensor constructor (#4113)Richard Zou1-0/+21
2018-02-01Add README.md for ATen/cudnn (#4998)Tongzhou Wang1-0/+4
2018-02-01Rewrite ATen native docs. (#4816)Edward Z. Yang2-33/+259
2018-02-01Make cat/cat_out native function that rejects scalar inputs. (#4992)gchanan4-2/+27
2018-02-01Support stack_out as a native function. (#4977)gchanan5-11/+30
2018-01-31Revert "torch.set_num_threads sets MKL option too" (#4967)Tongzhou Wang1-9/+4
2018-01-31fix triu and tril for zero-strided inputs on gpu (#4962)albanD2-27/+18
generated by cgit v1.2.3 (git 2.25.1) at 2024-11-30 20:57:30 +0000