wangwei
/
MegEngine
Not watched
1
0
Fork 0
Code
Releases 31 Wiki Activity
Issues 0 Pull Requests 0 Datasets Model Cloudbrain
1155 Commits
28 Branches
25 MB
536 Download
Tree: 44c8d2d16f
Branches Tags
${ item.name }
Create branch ${ searchTerm }
from '44c8d2d16f'
${ noResults }
Commit Graph

254 Commits (44c8d2d16f7276a00a170eeb2f7cfe9d6277184f)

Author SHA1 Message Date
  Megvii Engine Team 44c8d2d16f refactor(megdnn): refactor matmul algo in deformable conv 5 years ago
  Megvii Engine Team b04ad06f84 refactor(megdnn): refactor matmul algo in conv backward filter 5 years ago
  Megvii Engine Team 25089e520e refactor(megdnn): refactor matmul algo in conv backward data 5 years ago
  Megvii Engine Team 0d720653ac refactor(megdnn): add default algo for convolution forward 5 years ago
  Megvii Engine Team 659217acd2 refactor(megdnn): refactor bfloat16 convbias to recursive inteface 5 years ago
  Megvii Engine Team 4a1d52c9c6 refactor(megdnn): refactor bfloat16 matmul to recursive inteface 5 years ago
  Megvii Engine Team b8febaf91f refactor(megdnn): refactor bfloat16 convolutionbackwardfilter to recursive inteface 5 years ago
  Megvii Engine Team f14e0c17e7 feat(mgb): add recursive for fastrun and megdnn test 5 years ago
  Megvii Engine Team 85fa988348 refactor(dnn): add get_algorithm_from_desc interface 5 years ago
  Megvii Engine Team 2b8150ab52 fix(dnn): fix bazel build issue for cambricon platform 5 years ago
  Megvii Engine Team 329306b031 fix(cmake/cuda): fix build at cuda `copy` env caused by b278a69e1 5 years ago
  Megvii Engine Team 364afec033 chore(mge): update copyright years 5 years ago
  Megvii Engine Team ae8b38f634 fix(cmake/whl): reduce wheel size 5 years ago
  Megvii Engine Team 3bda334798 fix(dnn/fallback): fix segmentfault caused by im2col/conv1x1 using 5 years ago
  Megvii Engine Team 87ff58f7fc fix(megdnn): add algo for matmul/batchedmatrixmul of naive and opencl 5 years ago
  Megvii Engine Team a3caa5d3b7 fix(mgb(dnn)): fix convbias cudnnConvBiasActivation 5 years ago
  Megvii Engine Team 55042195d4 chore(winograd): add Convolutionv2 param 5 years ago
  Megvii Engine Team 409a877267 feat(dnn): add algo interface for rocm&fallback matmul and batched matrix mul 5 years ago
  Megvii Engine Team 8f7f52ae4d feat(jit): add memfwd in jit executor opr 5 years ago
  Megvii Engine Team dfb2b2ce49 fix(dnn): change pooling window size smaller than padding constraint to log_error 5 years ago
  Megvii Engine Team a85531dd0f feat(mgb/opr): add tqt opr 5 years ago
  Megvii Engine Team c3a4b2225d feat(dnn/cuda): add cutlass impls for fused convolution reformat operation 5 years ago
  Megvii Engine Team 5f44203d7b feat(dnn/cuda): add a cutlass impl for fusing convolution and dimshuffle 5 years ago
  Megvii Engine Team 61f917fb8e feat(dnn/cuda): add impl for fusing warp perspective and dimshuffle 5 years ago
  Megvii Engine Team eb826422c4 fix(dnn): forbid pooling window size smaller than padding 5 years ago
  Megvii Engine Team fc0fcd2f7f chore(winograd): remove winograd transform code 5 years ago
  Megvii Engine Team d1adc9a22f fix(dnn): fix opencl algo search 5 years ago
  Megvii Engine Team 0398a7867f fix(build/windows/cuda/llvm): fix windows bazel build with cuda 5 years ago
  Megvii Engine Team 3bf73ff16f feat(dnn): add cuda preprocess fusion 5 years ago
  Megvii Engine Team 86cf7490ec feat(dnn/aarch64): add quantizeds4 matmul int4x4x16_k8x8x8 5 years ago
  Megvii Engine Team 142f31a875 perf(dnn/cuda): change conv_bias heu, prefer dnn chanwise impl, dislike dnn batch gemm conv1x1 5 years ago
  Megvii Engine Team f214e14695 refactor(mgb/cuda): use single implementation of get_device_prop from utils 5 years ago
  Megvii Engine Team 54e79dd1d9 perf(mgb/cuda): do not call cudaGetDeviceProperties to avoid io traffic 5 years ago
  Megvii Engine Team a1877ee0fa refactor(dnn): refactor algo interface, use algoinfo instead of global algorithm 5 years ago
  Megvii Engine Team 6f5d0febf1 perf(dnn/cuda): enhance performance for pooling forward 5 years ago
  Megvii Engine Team 6856ce9ce2 feat(dnn): support conv bias activation for nchw4 input tensor format and nchw output tensor format 5 years ago
  Megvii Engine Team 60c6d59fc9 feat(mbg/core): support bias preprocess in conv_bias 5 years ago
  Megvii Engine Team 1f75c7ade4 ci(midout): fix midout and reopen midout test 5 years ago
  Megvii Engine Team 1e71e0afe0 refactor(dnn): refactor deconv algo 5 years ago
  Megvii Engine Team 89ad33aeb3 feat(dnn/cuda): support weight preprocessing for cutlass algorithms 5 years ago
  Megvii Engine Team c03249c059 feat(dnn/opr): add megdnn fake quant opr 5 years ago
  Megvii Engine Team 739f927c4c feat(dnn/cuda): opt dp4a conv for small channel base on cutlass 5 years ago
  Megvii Engine Team 1f8e40753f fix(mkl): fix windows mkl LOG compute exception 5 years ago
  Megvii Engine Team 4aa277a203 refactor(dnn/cuda): misc 5 years ago
  Megvii Engine Team f7b2bdae1a refactor(dnn): refactor algorithm type interface 5 years ago
  Megvii Engine Team 18ec5341f2 refactor(dnn): remove unused costmodel in cuda 5 years ago
  Megvii Engine Team e39f938662 refactor(dnn): remove ProfileCache and matmul algo in x86 5 years ago
  Megvii Engine Team 89303cd829 feat(megdnn/rocm): add bn for rocm backend 5 years ago
  Megvii Engine Team aea829c9fa feat(megdnn/rocm): add average inclusive mode for pooling 5 years ago
  Megvii Engine Team 1217801133 perf(mge): add opdef for broadcast 5 years ago