summaryrefslogtreecommitdiff
path: root/torch/lib
AgeCommit message (Expand)AuthorFilesLines
2018-04-19Let Gloo close socket, destroy() not needed for non-NCCL backend (#6787)Teng Li3-11/+13
2018-04-16Fixed NCCL build in fbcode (#6643)Teng Li1-1/+1
2018-04-10Fixed Clang Compilation Warnings for THD by removing outdated C linking (#6448)Teng Li1-15/+15
2018-04-09Using a function registry for THD init_methods for easy extension (#6334)Teng Li8-61/+127
2018-04-08[pytorch] add static linkage support for CuDNN and NCCL (#6410)Soumith Chintala1-1/+8
2018-04-05[ready] Implement log2 and log10 in PyTorch (#6272)Vishwak Srinivasan5-0/+10
2018-03-27Reorganize third-party libraries into top-level third_party directory (#6025)Edward Z. Yang64-10069/+0
2018-03-17Cleaner solution to the undefined references in RPC (#5817)Peter Goldsborough2-26/+31
2018-03-15Fix compilation with CUDA < 8.0 (#5621)Joan Puigcerver1-1/+6
2018-03-15Enable additional tensor types in Gloo backend (#5483)Myle Ott2-0/+4
2018-03-14Define RPC types out of source (#5794)Peter Goldsborough2-14/+28
2018-03-12Implement torch.reshape and Tensor.reshape (#5575)Sam Gross1-4/+1
2018-03-04Export MAX_JOBS for build_libs on WIndows (#5550)peterjc1231-1/+5
2018-02-27Fix ASAN alloc-dealloc-mismatch in TestMultiprocessing (#5428)Vedanuj Goswami1-1/+1
2018-02-20setup.py and cmake improvements (#5269)Edward Z. Yang1-19/+47
2018-02-12#4990, Makes Window build fail quicker (#5175)peterjc1231-1/+3
2018-01-28add compress flags to NCCLSoumith Chintala1-1/+1
2018-01-25add cuda9 options to ncclSoumith Chintala1-1/+10
2018-01-25Update pybind11, fix #4809 (#4811)Edward Z. Yang1-0/+0
2018-01-24Enabling Infiniband support for Gloo data channel with auto IB detection (#4795)Teng Li5-36/+112
2018-01-23Removing NCCL clear_group_cache workaround with one more check in new_group (...Teng Li1-0/+15
2018-01-12[ATen] Have any()/all() return a Tensor in preparation for dim/keepdim parame...gchanan1-2/+2
2018-01-11updating gloo to latest master (#4608)Soumith Chintala2-1/+1
2018-01-11Fixed setup.py to handle CUDNN_LIBRARY envvar with aten (#4597)Jon Crall2-0/+2
2018-01-10removing Local.cwrap entry in build_libs (#4595)Soumith Chintala1-1/+0
2018-01-06Fixes #4475, Add debug flag for Windows (#4508)peterjc1231-4/+10
2018-01-02Fixes for Windows build on master (#4432)Will Feng1-0/+9
2018-01-01Pick up NO_NNPACK for ATen build (#4423)Peter Goldsborough1-0/+1
2017-12-29Support NO_NNPACK environment variable (#4401)Edward Z. Yang1-0/+7
2017-12-28Adding torch.expm1() and its inplace function (#4350)Vishwak Srinivasan5-0/+5
2017-12-22Fixed unused result warnings in THDAdam Paszke2-2/+2
2017-12-20Fix compilation warnings (#4248)albanD1-1/+1
2017-12-15Allowing usage of GPU Direct within PyTorch for the Broadcast operation (#4183)Jorghi121-7/+24
2017-12-13Enable half communication for distributed (#4091)Christian Sarofeen4-0/+7
2017-12-04add missing CMAKE_GENERATORZach DeVito1-0/+1
2017-12-04Use ninja as the cmake backend as well.Zachary DeVito1-3/+5
2017-11-30Add support to emit compile_commands.json from CMake/ninjaZachary DeVito1-2/+4
2017-11-30Fix dependency build when pwd contains spaces (#3950)alcinos1-14/+15
2017-11-29Implemented NCCL Distributed Backend for PyTorch with new dist APIs (#3435)Teng Li17-14/+1364
2017-11-22Fix CharType min and max values (#3843)Simeon Monov1-1/+1
2017-11-22Fix errors in previous DataChannelMPI refactor (#3831)Adam Paszke1-28/+38
2017-11-21Improve DataChannelMPI (#3817)Adam Paszke2-174/+92
2017-11-20Reflect renaming of OS X to macOS (#3795)Scott Stevenson2-2/+2
2017-11-20Add Tensor.slice() (#3750)Sam Gross1-1/+1
2017-11-15[ATen] Rename isCuda -> is_cuda.Gregory Chanan1-1/+1
2017-11-15update gloo submodule (#3728)Soumith Chintala1-0/+0
2017-11-14set CC and CXX only when it's emptypeter1-2/+2
2017-11-12Fix CUDA builds for Windows (#3650)peterjc1231-2/+0
2017-11-10Enable EXPORT_ALL_SYMBOLS for CMAKE (#3617)peterjc1231-3/+5
2017-11-09Enable the build for MSVC 2017 and Ninja (#3595)peterjc1231-4/+27
generated by cgit v1.2.3 (git 2.25.1) at 2024-11-27 23:18:23 +0000