207 Commits (91efd67dbfc6be18527154397ade2ef0f899777c)

Author SHA1 Message Date
  Megvii Engine Team 539aac4482 fix(dnn/cuda): fix sass loader 5 years ago
  Megvii Engine Team 023d60aee4 fix(mgb/atlas): use aclmdlGetOutputSizeByIndex to create DataBuffer 5 years ago
  Megvii Engine Team 116eee5231 build(third_party): update megray 5 years ago
  Megvii Engine Team 23437864f9 fix(mgb/jit): mlir doesn't support broadcast 5 years ago
  Megvii Engine Team f87bba6806 feat(mgb/jit): add scalar support for mlir 5 years ago
  Megvii Engine Team 11b121a7d2 fix(mgb/jit): link libdevice.bc when generate nvvm ir 5 years ago
  Megvii Engine Team aa7f28b835 fix(mgb/jit): fix gpu kernel args order 5 years ago
  Megvii Engine Team 9767ca8f19 feat(mgb/jit): refactor code and add more elemwise mode 5 years ago
  Megvii Engine Team f9109bed8b feat(mge): add mlir cmake 5 years ago
  Megvii Engine Team a51d5b4c31 feat(mgb/jit): add mlir backend for cpu and cuda 5 years ago
  Megvii Engine Team dd39265e95 fix(mgb/dtype): enable TypeCvt for bool when trace(symbolic=True) 5 years ago
  Megvii Engine Team 6f581906a1 refactor(mge/profiler): refactor profiler 5 years ago
  Megvii Engine Team cc952b2b92 fix(rocm): fix rocm megdnntest sleep and a cut code 5 years ago
  Megvii Engine Team a66d4b8bb8 fix(mge/parampacksplit): fix param pack split mem forward 5 years ago
  Megvii Engine Team 76fa71573b feat(dnn/cuda): add cutlass nchw4 convolution 5 years ago
  Megvii Engine Team a558d4a253 fix(mgb/atlas): remove unnessary setdevice 5 years ago
  Megvii Engine Team 1fe8a21299 fix(mge): fix sublinear memory in jit.trace 5 years ago
  Megvii Engine Team 651920c7ad fix(dnn): fix nchw88 winograd weight preprocess 5 years ago
  Megvii Engine Team d06f248d90 fix(whl/mgb/imperative): fix symbols conflict runtime crash 5 years ago
  Megvii Engine Team 0e82b959a1 feat(mge/imperative): add sublinear options 5 years ago
  Megvii Engine Team 50d5421aa7 fix(mgb/zmq): fix unused-result when compiling with c++17 5 years ago
  Megvii Engine Team 16324e3076 feat(dnn/cuda): add remap backward 5 years ago
  Megvii Engine Team 968f74ce88 chore(mgb): add no_force_inplace option to ComputingGraph 5 years ago
  Megvii Engine Team 6e882c1a86 feat(whl/imperative): compat for build python whl imperative and legacy runtime 5 years ago
  Megvii Engine Team 7f857bd471 feat(mgb/rocm): add cmake for rocm and fix compile errors and bn 5 years ago
  Megvii Engine Team 9510136223 fix(mgb/rocm): remove begin-internal of rocm 5 years ago
  Megvii Engine Team 6b380e8965 feat(mge/imperative): run oss test and restore cmake list build items 5 years ago
  Megvii Engine Team 2dbe8194ad fix(mge/opr): fix reduction static infer value 5 years ago
  Megvii Engine Team c20d4cc6dc feat(dnn): fix opt pass nchw44 can not dump resnet 5 years ago
  Megvii Engine Team 00ef677249 fix(mgb): remove internal for cambricon and atlas 5 years ago
  Megvii Engine Team bca00f2e22 fix(dnn): midout at where neccessary in megdnn 5 years ago
  Megvii Engine Team a1e6720756 feat(dnn): enable bool comparison 5 years ago
  Megvii Engine Team 4a178a8dba feat(windows/cuda/cmake): support cmake cuda build on windows 5 years ago
  Megvii Engine Team 1915593b6c fix(version_depend): add a fake version on dev 5 years ago
  Megvii Engine Team 56381f808b fix(dnn/arm): use vcvtq_f32_s32 for all arm code 5 years ago
  Megvii Engine Team 1173205726 fix(gopt): nchw_nchwxx useable and opt pass use nchw_nchwxx_valid 5 years ago
  Megvii Engine Team eb18eba87d fix(gopt): fix nchw44 nchw44_dot gopt test 5 years ago
  Megvii Engine Team eab7ab0530 fix(gopt): gen nchw_nchw44 when kernel is optimized 5 years ago
  Megvii Engine Team 777f3ea970 refactor(gopt): format code 5 years ago
  Megvii Engine Team 30ce3c60bd Revert "fix(mgb/opr): change EQ opr's backward_graph to nullptr instead of InvalidGrad" 5 years ago
  Megvii Engine Team 14e71b551b feat(imperative): add helper for dnn opr caller 5 years ago
  Megvii Engine Team 230ab45a1e fix(mgb/naive): fix naive convolution no dispatch kernel in handle 5 years ago
  Megvii Engine Team 1bce857cb8 fix(mgb/opr-mm): use comp_node of config as default in CollectiveComm 5 years ago
  Megvii Engine Team 27205461ae feat(mgb/opr-mm): add register info cache for multi-machine oprs 5 years ago
  Megvii Engine Team 96ec586d28 fix(dnn): fix bool cvt 5 years ago
  Megvii Engine Team f829f836b9 test(mgb/index): add empty index desc tests 5 years ago
  Megvii Engine Team e73f2799d0 fix(mgb/index): enable index desc empty 5 years ago
  Megvii Engine Team ff60fdb82d feat(dnn): add bool type cvt on gpu 5 years ago
  Megvii Engine Team e8571cca51 fix(mgb/cuda): fix cuda host alloc set device 5 years ago
  Megvii Engine Team f7b5eced23 refactor(mgb/opr-mm): set False as default value of local_grad 5 years ago