710 Commits (7838ba94c0344381492abd4278b47298c63107da)
 

Author SHA1 Message Date
  Megvii Engine Team 2272abe18d fix(mgb/fallback): disable nchw44 in conv1x1 and im2col in x86 4 years ago
  Megvii Engine Team 230ab45a1e fix(mgb/naive): fix naive convolution no dispatch kernel in handle 4 years ago
  Megvii Engine Team 22853fa20c feat(mge/quantization): add `mapping` parameter for custom modules 4 years ago
  Megvii Engine Team 6e70fa7a11 feat(dnn/arm): add fp32 asm gemm for a53 a55 and i8i8i16 gemm for a72 a53 4 years ago
  Megvii Engine Team dbaf84b0ef feat(imperative): add cond_take opr 4 years ago
  Megvii Engine Team df356635b7 fix(mgb/fallback): delete im2col duplicate code and fix nchw44 usable 4 years ago
  Megvii Engine Team 4a2270834f fix(mgb/fallback): fix conv1x1 and conv1x1_gemv nchw44 usable 4 years ago
  Megvii Engine Team b778d22523 feat(mgb/fallback): add conv1x1_gemv, conv1x1 and im2col 8x8x16/8x8x32 support bias 4 years ago
  Megvii Engine Team c357db0134 feat(mgb/arm_common): add 8x8x16 nchw44 max pooling 4 years ago
  Megvii Engine Team 7f5f375fda feat(dnn/arm): add armv7 nchw_nchw44 3x3s2 asm kernel 4 years ago
  Megvii Engine Team b7d5fa7e64 fix(sdk/load_and_run): fix misuse std::string::substr 4 years ago
  Megvii Engine Team 1bce857cb8 fix(mgb/opr-mm): use comp_node of config as default in CollectiveComm 4 years ago
  Megvii Engine Team 27205461ae feat(mgb/opr-mm): add register info cache for multi-machine oprs 4 years ago
  Megvii Engine Team a7ff580e54 feat(mge/utils): add net stats to calculate parameters and flops 4 years ago
  Megvii Engine Team 96ec586d28 fix(dnn): fix bool cvt 4 years ago
  Megvii Engine Team f26cd398e3 build(third_party): Update megray version 4 years ago
  Megvii Engine Team f829f836b9 test(mgb/index): add empty index desc tests 4 years ago
  Megvii Engine Team e73f2799d0 fix(mgb/index): enable index desc empty 4 years ago
  Megvii Engine Team b43f6a2602 fix(mge/quantization): handle empty Observer in QATModule 4 years ago
  Megvii Engine Team 13e8f00a37 feat(mge/module): add forward hook support 4 years ago
  Megvii Engine Team ab9fa48ee7 feat(mge/quantization): make `q_dict` a kwarg rather than an arg 4 years ago
  Megvii Engine Team f8810f733a feat(mge/imperative): prepare to make whl 4 years ago
  Megvii Engine Team ff60fdb82d feat(dnn): add bool type cvt on gpu 4 years ago
  Megvii Engine Team e8571cca51 fix(mgb/cuda): fix cuda host alloc set device 4 years ago
  Megvii Engine Team f7b5eced23 refactor(mgb/opr-mm): set False as default value of local_grad 4 years ago
  Megvii Engine Team 7a8183f4e0 fix(mge/quantization): fix enable observer bug 4 years ago
  Megvii Engine Team 555ecea9bc feat(mge/quantization): add bias fakequant support 4 years ago
  Megvii Engine Team 9440842e27 fix(mge/core): fix Tensor deepcopy issue 4 years ago
  Megvii Engine Team d4b86b844e feat(mge/dtype): add int2 lowbit support and example 4 years ago
  Megvii Engine Team 3931099ea7 fix(dnn/test): fix nchw_nchw44 i8i8i16 benchmark 4 years ago
  Megvii Engine Team bcf5691ddf feat(dnn/arm): add nchw_nchw44 i8i8i16 2x2 3x3 5x5 7x7 s1 s2 conv 4 years ago
  Megvii Engine Team c7b6ef35c1 feat(dnn/cuda): add warp perspective backward mat idx 5 years ago
  Megvii Engine Team a773d07678 feat(dnn/arm_common): add nchw44 8x8x16 channel wise conv 4 years ago
  Megvii Engine Team 09b5f3d434 fix(mgb/core): fix multi thread pool deactive and multi thread conflict 4 years ago
  Megvii Engine Team ef239f835f feat(windows/python_whl): make windows HAPPY for build megbrain python package 4 years ago
  Megvii Engine Team bf6cbc1df7 build(third_party): fix git apply issue 4 years ago
  Megvii Engine Team 5eb491c5af Merge pull request #74 from ChaiMind:master 4 years ago
  Megvii Engine Team b72f1e8258 chore(build): cleanup BUILD files 4 years ago
  Megvii Engine Team e258812f12 feat(dnn): add bool dtype 4 years ago
  Megvii Engine Team 734c498d27 perf(mgb/core): improve DevMemAlloc when it has single stream 4 years ago
  Megvii Engine Team 39bd66fc63 fix(mgb): fix TensorRT missing cudaSetDevice 4 years ago
  Megvii Engine Team ab9dfbcefc test(mgb): fix tensorrt tests missing cudaSetDevice 4 years ago
  Megvii Engine Team b43fb1a97c perf(mgb): add CUDA host memory allocator 4 years ago
  Megvii Engine Team 2afceb4187 fix(mgb/atlas): use dyn output alloc if enable dynamic batchsize 4 years ago
  Megvii Engine Team 6bcc6faec8 feat(mge/imperative/opr): modify batch_norm to support frozen BN 4 years ago
  Megvii Engine Team 7ca3d579db feat(dnn): make mk4 and mk8 matmul for winograd both on aarch64 and armv7 supports n=1 4 years ago
  Megvii Engine Team 54d18115b6 fix(imperative): fix grad of BatchNorm 4 years ago
  Megvii Engine Team 80c4705317 perf(mgb): use midout in megbrain to reduce binary size 4 years ago
  Megvii Engine Team 35c712767d fix(mge/quant): fix TQT epoch scale change bug 4 years ago
  Megvii Engine Team e6e41242c7 fix(mge/quant): fix zero grad warn in TQT train 4 years ago