200 Commits (release-1.4)

Author SHA1 Message Date
  Megvii Engine Team 052a600f03 feat(mge/module): add python wrapper for unfold 4 years ago
  Megvii Engine Team 4fa9096d20 feat(megbrain): add correlation opr 4 years ago
  Megvii Engine Team b3aae4bb16 feat(dnn/cuda): add correlation kernel 4 years ago
  Megvii Engine Team 747b53c018 feat(dnn): add conv_bwd_data and conv_bwd_filter accuracy shake check 4 years ago
  Megvii Engine Team 3f5238fb38 feat(mgb/dnn): add accuracy shake checker 4 years ago
  Megvii Engine Team 07ab8cb6b6 feat(dnn): add param_pack for rocm 4 years ago
  Megvii Engine Team ff755451d2 refactor(mgb): move algo's name from info to desc and delete some algo's unnecessary param() method 4 years ago
  Megvii Engine Team 756c1eb7f2 fix(mgb/dnn): add cuda float naive matmul algo 4 years ago
  Megvii Engine Team a49f4a66b7 feat(dnn): add indexing_one_hot and indexing_set_one_hot opr 4 years ago
  Megvii Engine Team 2fd3fa8834 feat(cmake): update for enflame cmake compile 4 years ago
  Megvii Engine Team 68f2e59763 fix(mgb(ci)): fix tx1 ci testcase 4 years ago
  Megvii Engine Team 55974e8cf9 feat(log): opt log 4 years ago
  Megvii Engine Team 1c01128ff5 fix SASS cu111 4 years ago
  Megvii Engine Team ba2ad46e54 feat(gopt): add deconv nchw4 int8 opt pass, add deconv nchw int8 4 years ago
  Megvii Engine Team 5d350fc843 feat(dnn/cuda): add deconv int8 and fix cutlass conv wrapper base on modify cutlass 2.4 4 years ago
  Megvii Engine Team c82d88751a fix(dnn/cuda): add cuda nchw int8 conv impl with nchw4 to fix cu111 compatibility 4 years ago
  Megvii Engine Team f2b42bf09e chore(dotprod): add arm dotprod attribute for easy use 4 years ago
  Megvii Engine Team 97beae2fd8 fix(megdnn): fix megdnn benchmark testcase 4 years ago
  Megvii Engine Team 2de2222e46 feat(dnn/cuda): add cutlass batched gemv kernel for matmul operator 4 years ago
  Megvii Engine Team 973d2a0ac2 feat(dnn/cuda): add cutlass matmul using split k parallel 4 years ago
  Megvii Engine Team 03c921f7c4 feat(dnn/cuda): add cutlass matmul impls 4 years ago
  Megvii Engine Team 5b62acfa01 feat(dnn/armv7): add new matmul strategy k8x8x4 4 years ago
  Megvii Engine Team cf27dd642c fix(cuda): use cudnn8.0.4 as cu111 default libs 4 years ago
  Megvii Engine Team 649e4dd750 test(cuda): fix test for cu111 4 years ago
  Megvii Engine Team c69359d00d fix(dnn/cuda): disable cudnn conv_bias kernels for NCHW4_NCHW tensor format 4 years ago
  Megvii Engine Team 0e3a6329ff build(cuda): support cu111 build 4 years ago
  Megvii Engine Team cd7090acbb fix(opencl): enable image on mali(cl2.1) 4 years ago
  Megvii Engine Team c51a687cef chore(mge): update copyright years 4 years ago
  Megvii Engine Team af42ce7e69 fix(megdnn): some fixes of execution policy 4 years ago
  Megvii Engine Team 821656aa4b refactor(megdnn): refactor brute force algo in batched matmul 4 years ago
  Megvii Engine Team 08ff62deb6 refactor(megdnn): refactor batched matmul algo in conv bias 4 years ago
  Megvii Engine Team 8773926ef8 refactor(megdnn): refactor matmul algo in conv bias 4 years ago
  Megvii Engine Team e4b71bdf64 refactor(megdnn): remove unnessary 1x1 algo 4 years ago
  Megvii Engine Team b04ad06f84 refactor(megdnn): refactor matmul algo in conv backward filter 4 years ago
  Megvii Engine Team 25089e520e refactor(megdnn): refactor matmul algo in conv backward data 4 years ago
  Megvii Engine Team 0d720653ac refactor(megdnn): add default algo for convolution forward 4 years ago
  Megvii Engine Team 659217acd2 refactor(megdnn): refactor bfloat16 convbias to recursive inteface 4 years ago
  Megvii Engine Team 4a1d52c9c6 refactor(megdnn): refactor bfloat16 matmul to recursive inteface 4 years ago
  Megvii Engine Team b8febaf91f refactor(megdnn): refactor bfloat16 convolutionbackwardfilter to recursive inteface 4 years ago
  Megvii Engine Team f14e0c17e7 feat(mgb): add recursive for fastrun and megdnn test 4 years ago
  Megvii Engine Team 0e8b81c20e fix(dnn/opencl): fix elemwise negative stride support 4 years ago
  Megvii Engine Team 364afec033 chore(mge): update copyright years 4 years ago
  Megvii Engine Team ae8b38f634 fix(cmake/whl): reduce wheel size 4 years ago
  Megvii Engine Team 3bda334798 fix(dnn/fallback): fix segmentfault caused by im2col/conv1x1 using 4 years ago
  Megvii Engine Team 409a877267 feat(dnn): add algo interface for rocm&fallback matmul and batched matrix mul 4 years ago
  Megvii Engine Team a85531dd0f feat(mgb/opr): add tqt opr 4 years ago
  Megvii Engine Team c3a4b2225d feat(dnn/cuda): add cutlass impls for fused convolution reformat operation 4 years ago
  Megvii Engine Team 5f44203d7b feat(dnn/cuda): add a cutlass impl for fusing convolution and dimshuffle 4 years ago
  Megvii Engine Team 61f917fb8e feat(dnn/cuda): add impl for fusing warp perspective and dimshuffle 4 years ago
  Megvii Engine Team fc0fcd2f7f chore(winograd): remove winograd transform code 4 years ago