wangwei
/
MegEngine
Not watched
1
0
Fork 0
Code
Releases 31 Wiki Activity
Issues 0 Pull Requests 0 Datasets Model Cloudbrain
1698 Commits
28 Branches
25 MB
2473 Download
Tree: cbf3bb18da
Branches Tags
${ item.name }
Create branch ${ searchTerm }
from 'cbf3bb18da'
${ noResults }
Commit Graph

114 Commits (cbf3bb18da08a8354e3378ce7178d768bf64151f)

Author SHA1 Message Date
  Megvii Engine Team b18feaab33 feat(dnn/cuda): use cutlass remove shared load imma conv kernel 5 years ago
  Megvii Engine Team 1af350c6d2 feat(dnn): add fill kernel 5 years ago
  Megvii Engine Team 3eb0505f9b feat(imperative): add support for quantized conv transpose2d 5 years ago
  Megvii Engine Team 3b452d8c16 feat(mgb): cuda conv support nhwc format and fp16 dtype 5 years ago
  Megvii Engine Team 2aba0378b9 refactor(mgb/dnn): fix group conv is_available 5 years ago
  Megvii Engine Team 4a92346b7a refactor(mgb): refactor group conv3d 5 years ago
  Megvii Engine Team 6ce212d2e0 refactor(mgb): refactor group conv 5 years ago
  Megvii Engine Team 869a03271b perf(mgb): disable FoldingConvBiasDimshufflePass in cuda10 for performance 5 years ago
  Megvii Engine Team 8d248a6a9a fix(dnn/cuda): fix testcase for fallback nchw qs8 conv 5 years ago
  Megvii Engine Team 894a2407c2 feat(dnn/cuda): add relayout format kernel for nchw <-> nhwc 5 years ago
  Megvii Engine Team f41a808694 feat(dnn/cuda): add nhwc int4 conv support 5 years ago
  Megvii Engine Team 633016a962 fix(dnn/cuda): fix AlgoFallbackNCHWQS8 to support Float32 dst 5 years ago
  Megvii Engine Team 43098fb8f1 feat(mge): add SlidingWindowTranspose opr 5 years ago
  Megvii Engine Team b078dda90b feat(mge/random): add some random op and remove random/distrbution.py 5 years ago
  Megvii Engine Team f30c0e06a6 feat(mgb/opr): add lsq opr 5 years ago
  Megvii Engine Team 1cfdbc565c feat(dnn): add deterministic max pooling 5 years ago
  Megvii Engine Team a5060a2bfe feat(mgb/opr): add check_has_inf kernel and opr 5 years ago
  Megvii Engine Team d915c5a3fd refactor(mgb): make convolution3D handle noncontiguous tensors 5 years ago
  Megvii Engine Team d04cd67faf refactor(mgb): make conv-backward-filter handle noncontiguous tensors 5 years ago
  Megvii Engine Team 44376f702a refactor(mgb): make conv-backward-data handle noncontiguous tensors 5 years ago
  Megvii Engine Team 7b2a76d1ee refactor(mgb): make conv handle noncontiguous tensors 5 years ago
  Megvii Engine Team b87af9f77f feat(dnn/cuda): topk support fp16 5 years ago
  Megvii Engine Team 606540bef4 feat(dnn/cuda): add nhwc 4bit warp perspective 5 years ago
  Megvii Engine Team 1e6019436c feat(dnn/cuda): add nhwc int4 pooling 5 years ago
  Megvii Engine Team 319436dd14 feat(dnn/cuda): add cutlass impls for uint4 x int4 conv bias 5 years ago
  Megvii Engine Team d28eba4ea5 feat(dnn/cuda): add cutlass impls for int4 conv bias 5 years ago
  Megvii Engine Team 2d4e62ef58 feat(dnn/cuda): add cuda uint4 pooling 5 years ago
  Megvii Engine Team 19919384fc feat(dnn/cuda): add cuda uint warp perspective 5 years ago
  Megvii Engine Team 4a802d21ca feat(dnn/cuda): add conv u4xs4 sass kernel 5 years ago
  Megvii Engine Team adf75a291d perf(dnn/cuda): add sass int4 128x128 5 years ago
  Megvii Engine Team 8da2f698a3 feat(dnn/cuda): support warp perspective/pooling op when channel not aligned to 64 5 years ago
  Megvii Engine Team 4fe68ac9ed feat(dnn/cuda): support transforming layout between nchw and nchw64 when channel not aligned to 64 5 years ago
  Megvii Engine Team 56e863b7d4 fix(dnn/cuda): fix int4 epilogue stg bug 5 years ago
  Megvii Engine Team 12a0e61542 feat(dnn/cuda): add cuda elemwise int4 5 years ago
  Megvii Engine Team df1af59b5c feat(dnn): warp perspective support int4 5 years ago
  Megvii Engine Team 2398df079c feat(dnn/cuda): add cuda int4 pooling 5 years ago
  Megvii Engine Team e250afb08f feat(dnn/cuda): support conv_bias for nchw64 and qint4 5 years ago
  Megvii Engine Team 8fef78d06d feat(dnn/cuda): add relayout format when width is an odd number 5 years ago
  Megvii Engine Team 19a554d674 test(dnn/cuda): add testcase for transforming tensor layout between nchw and nchw64 5 years ago
  Megvii Engine Team 23032f50f2 feat(dnn/cuda): support float16 for index_incr_multi_axis_vec 5 years ago
  Megvii Engine Team 938944027d fix(mgb/dnn): fix cudnn8 convbias 5 years ago
  Megvii Engine Team 1997b1a289 feat(dnn/cuda): add correlation kernel 5 years ago
  Megvii Engine Team c3f8cf04fa feat(dnn): add conv_bwd_data and conv_bwd_filter accuracy shake check 5 years ago
  Megvii Engine Team 1e6ef3771f feat(mgb/dnn): add accuracy shake checker 5 years ago
  Megvii Engine Team ff755451d2 refactor(mgb): move algo's name from info to desc and delete some algo's unnecessary param() method 5 years ago
  Megvii Engine Team 756c1eb7f2 fix(mgb/dnn): add cuda float naive matmul algo 5 years ago
  Megvii Engine Team 68f2e59763 fix(mgb(ci)): fix tx1 ci testcase 5 years ago
  Megvii Engine Team ba2ad46e54 feat(gopt): add deconv nchw4 int8 opt pass, add deconv nchw int8 5 years ago
  Megvii Engine Team 5d350fc843 feat(dnn/cuda): add deconv int8 and fix cutlass conv wrapper base on modify cutlass 2.4 5 years ago
  Megvii Engine Team c82d88751a fix(dnn/cuda): add cuda nchw int8 conv impl with nchw4 to fix cu111 compatibility 5 years ago