1. 22 May, 2018 1 commit
  2. 19 May, 2018 2 commits
  3. 17 May, 2018 1 commit
  4. 14 May, 2018 1 commit
  5. 12 May, 2018 1 commit
  6. 08 May, 2018 1 commit
  7. 01 May, 2018 1 commit
  8. 28 Apr, 2018 1 commit
  9. 27 Apr, 2018 1 commit
  10. 25 Apr, 2018 2 commits
  11. 23 Apr, 2018 1 commit
  12. 19 Apr, 2018 1 commit
  13. 15 Apr, 2018 1 commit
  14. 14 Apr, 2018 1 commit
  15. 13 Apr, 2018 1 commit
  16. 01 Apr, 2018 1 commit
  17. 31 Mar, 2018 2 commits
  18. 27 Mar, 2018 2 commits
  19. 19 Mar, 2018 1 commit
  20. 08 Mar, 2018 1 commit
  21. 01 Mar, 2018 2 commits
  22. 15 Feb, 2018 1 commit
  23. 13 Feb, 2018 1 commit
  24. 12 Feb, 2018 1 commit
  25. 10 Feb, 2018 1 commit
    • [TOPI] Initial NHWC layout support (#882) · 54d4fe4b
      * add 4 dim softmax
      
      * update for NHWC layout
      
      * remove layout param from softmax
      
      * fix typo
      
      * minor fix to pool
      
      support axis=1 ndims=5 softmax.
      
      add softmax axis
      
      * few fix for softmax
      
      * fix typo
      
      * add more doc
      
      * minor doc fix
      
      * fix upsampling output shape
      
      * fix lint
      
      * cleanup softmax
      
      * minor fix
      
      * raise exception instead of assert, handles negative axis
      
      * check axis after axis transformation
      masahi committed
  26. 09 Feb, 2018 1 commit
  27. 08 Feb, 2018 1 commit
  28. 07 Feb, 2018 1 commit
  29. 04 Feb, 2018 1 commit
  30. 02 Feb, 2018 1 commit
  31. 01 Feb, 2018 1 commit
  32. 28 Jan, 2018 1 commit
    • Porting schedules (except convolutions) to C++ (#763) · f280f23a
      * Ported injective schedules to C++. Added some elementwise ops.
      
      * Fix lint errors
      
      * Added reduction ops and schedules
      
      * Fix lint errors
      
      * Fix lint errors
      
      * Fix lint errors
      
      * Added transform ops
      
      * Fix lint errors
      
      * Fix lint errors
      
      * Added softmax, log_softmax, leaky_relu and flatten ops.
      Fixed issue where TVM_DECLARE_INTRIN_UNARY used the PureExtern flag
      instead of PureIntrinsic.
      Added softmax CUDA schedule.
      
      * Fix lint
      
      * Fix lint
      
      * Added binary_dense, batch_norm_inference, dense, dilate, scale_shift_*,
      global_pool and pool ops.
      Extended pad to allow specifying pad_value.
      Fixed issue where pad would throw if padding was zero in all dimensions.
      
      * Fix lint
      
      * Fix lint
      
      * Added CUDA schedules for dense, pool and global_pool
      
      * Added extern schedules for generic and CUDA
      
      * Fix lint
      
      * Added x86 binary schedules
      
      * Fix lint
      
      * Added rocm dense schedule. Added rocBLAS and cuBLAS support to dense ops
      
      * Added pow ops. Added x86 default and injective schedules
      
      * Fix lint
      
      * Fix lint
      
      * Fix lint
      
      * Fix lint
      
      * Fix lint
      
      * Fix indent
      
      * Removed schedules directory
      
      * Changed left_shift, right_shift to operators. Changed pad_value in pad() to remove pointer usage
      
      * Fixed usage of pad in nn/pooling.h. Fixed declaration of operator>>
      
      * Fixed comments for shift operators
      
      * Added comments to utility functions
      
      * Added TOPI C++ library, exporting broadcast_add op
      
      * Fix lint
      
      * Share libinfo.py with TVM
      
      * Fix lint
      
      * Add other broadcast ops
      
      * Fix lint
      
      * Fix imports in topi
      
      * Fix lib names
      
      * Fixed build issue where windows builds don't apply correct definitions
      
      * Removed TVM_EXPORTS from topi library
      
      * Attempted CI build fix
      
      * Add topi lib to tvm_multilib
      
      * Fix Jenkinsfile
      
      * Added TOPI build target to Makefile
      
      * Fix nn op namespaces.
      
      * Fix lint
      
      * Renamed TOPI lib to libtvm_topi
      
      * Removed _ffi/base.py
      
      * Remove _ffi from topi, now shared with tvm.
      
      * Make libtvm_topi loading optional
      
      * Fix compiler warnings
      
      * Fix lint
      
      * Fix lint
      
      * Fix lint
      
      * Fix build error by making new libs argument to Target optional
      
      * Added C++ Target type interop. Added registration of remaining C++ ops and schedules. Added test of broadcast ops
      
      * Fix lint
      
      * Fix lint
      
      * Fix compile error
      
      * Fix compiler warnings
      
      * Fix compiler warnings
      
      * Fixed int vector interop. Fixed argmin incorrectly invoking argmax. Fixed corner case in default schedules of attempting to fuse 0 length axes. Added tests for reduce ops.
      
      * Refactored reduce builders
      
      * Fixed typos in topi.cc. Added basic test.
      
      * Fixed padding size error. Added dense, dilate, pooling tests
      
      * Fixed issue where clip would output a different dtype to the input. Added split_sections op to cover the other mode of the python split op. Added tests.
      
      * Changed extension type numbers to avoid clash with NNVM
      
      * Fix lint
      
      * Fix compiler warnings
      
      * Removed use of std::vector from the public TOPI API
      
      * Fix lint
      
      * Add TOPI C++ tests to CI
      
      * Fixed detail namespacing. Improved comments.
      alex-weaver committed
  33. 19 Jan, 2018 1 commit
  34. 16 Jan, 2018 2 commits
    • [TOPI] Basic x86 schedules (#775) · 3df42cd7
      * add basic x86 schedules
      
      * parallelize & vectorize batchnorm + relu
      
      * fuse conv into bn + relu
      
      * move rc loop to outer
      
      * add nhwc conv
      
      * change weight layout to hwcf
      
      * conv + bn + relu fusion for nhwc conv
      
      * fix conv_nhwc schedule when no fusion
      
      * clean up default parallel schedules
      
      * simplify elemwise parallel
      
      * fix elemwise parallel for batch == 1
      
      * update nhwc conv test
      
      * fix and add comment
      
      * fix lint
      
      * remove redundant import
      
      * remove default multithreading for some ops
      
      * remove default multithreading for global pool
      masahi committed