283 Commits (8d00a8d803dba83322c19fc7aa60d81452f7b699)

Author SHA1 Message Date
  ms_yan 36a8886ca2 Revert "[feat] [assistant] [I3T96T] add new Dataset operator CMUARCTICDataset" 4 years ago
  djc b077aa1cab [feat] [assistant] [I3T96T] add new Dataset operator CMUARCTICDataset 4 years ago
  djc 4e6f7dc97d [feat] [assistant] [I3T96X] add new Dataset operator LibriSpeechDataset 4 years ago
  yelihua 72e6058265 get rank id when set hccl env for single card train 4 years ago
  yelihua a6dc9a0a07 get rank id when set hccl env for single card train 4 years ago
  hwjiaorui 23b1fb7f3f opt code and add childgraph manager 4 years ago
  zhousiyi 94e5fe6242 keep tuple_getitem as much as possible to reduce the number of func graphs 4 years ago
  hwjiaorui 1fa30a665f tensor device address setting 4 years ago
  kswang 4a456ca954 clean code 4 years ago
  i-robot 4c3c3ceee6 !21157 add AdamWeightDecay to optSet 4 years ago
  i-robot eb5b2c478e !21230 fix python kernel execute lock 4 years ago
  kswang 8fa85cac34 add adamweightdecay to opset and update lock input tensors 4 years ago
  i-robot 2a5052ddb6 !21068 do not put Push operator into UpdateState 4 years ago
  wilfChen 126bfff431 fix graph execution acquire gil. 4 years ago
  limingqi107 fe36a158af unified runtime pclint fixed 4 years ago
  Margaret_wangrui ee93310023 do not put Push operator into UpdateState 4 years ago
  Margaret_wangrui 779d5f9dbf fix ps find pull node bug 4 years ago
  limingqi107 0b8937cd6c unified runtime bug of auto moand and host device of multi graphs 4 years ago
  Margaret_wangrui f9a064e464 Add ref user to UpdateState to ensure the order 4 years ago
  lizhenyu 209dc5f584 fix codedex warnings 4 years ago
  ZPaC a9a0f590e6 Fix master static check 4 years ago
  lizhenyu 5535e80414 [bugfix] custom bprop got unexpected input 4 years ago
  lizhenyu 6c467a3815 [bugfix] Heterogenous scenario in a ms function of PyNative Mode occur core dump 4 years ago
  caifubi c25de59f77 fix allreduce fusion index bug 4 years ago
  lizhenyu 2d2bcef88d fix adam op accu bug 4 years ago
  i-robot 70910a6fbc !19110 Ignore monad input when GetOpInputTensor 4 years ago
  tanghuikang 16a392fcb3 Ignore monad input when GetOpInputTensor 4 years ago
  i-robot cc8d760c8b !18980 Add NewCNodeWithInfos for kernel graph 4 years ago
  l00591931 c7bc867ae4 Add CloneCNodeWithInfos for kernel graph 4 years ago
  kswang 9610b48019 visit nop node for output addr exist check 4 years ago
  chujinjin ac3670ef91 modify log level for pynative ascend 4 years ago
  i-robot 32a54eb4eb !15330 add updated paramter for graph 4 years ago
  kswang 42155e7e5c add updated parameter for graph 4 years ago
  i-robot 117a93000b !18799 fix get output addr error 4 years ago
  kswang 6afd18a4d8 fix get outputaddress error 4 years ago
  kswang 3247c00555 optimize heter memcpy 4 years ago
  i-robot 19b87fe35e !18172 unify runtime for PyNative supports bp hook and custom bp operator 4 years ago
  lizhenyu 887d96063b unify runtime support pynative hook 4 years ago
  caifubi 02d68cb877 Add check for allreduce fusion split 4 years ago
  limingqi107 70ff24cb15 fix the same output of actor runtime 4 years ago
  i-robot 50304d82fb !17661 add permission for save files 4 years ago
  buxue 0049109d11 add permission for save files 4 years ago
  lizhenyu f3e5d67512 fix core dump when destroy device context in PyNative mode 4 years ago
  i-robot f5ceaff1d3 !17583 modify log level for pynative 4 years ago
  gaoyong10 bf6528645c Unified runtime support mixed precision. 4 years ago
  chujinjin 1dd950218b modify log level for pynative 4 years ago
  mindspore-ci-bot 177ead9b8b !17751 actor runtime support GraphKrenel 4 years ago
  limingqi107 7f634d12f0 actor runtime support GraphKrenel 4 years ago
  TinaMengtingZhang 2fa05b66a1 change device_id to rank_id in dump path 4 years ago
  mindspore-ci-bot 5616837107 !17587 code format 4 years ago