51 Commits (6fcd6cab684bd2d2ae2da5613efa3aa4e2cd7d5a)

Author SHA1 Message Date
  huangbingjian 2a85af5d83 1. if Switch/SwitchLayer, do not replace Load or remove UpdateState; 2. add control flow testcases; 3. fix codedex problem 4 years ago
  huangbingjian 302746d6a3 fix code check 4 years ago
  huangbingjian e5d32c9ff8 add switch_defer_inline 4 years ago
  dayschan 771e3f61f3 Clean GraphKernel's codes from frontend 5 years ago
  zhangzhaoju a1db9c4959 Pybind11::object && PrimitivePy recycle optimize 5 years ago
  huangbingjian 1c2430b2bd update FindPrimalJPair 5 years ago
  zhangzhaoju ef4481c4a3 Memory leak fix, clear using cnt map of graph while funcgraph manager 5 years ago
  huangbingjian c8f9d93c2d modify findPrimalJPair 5 years ago
  zhousiyi 74d500a756 If a fv cnode is not mapped but belong to primal_graph_, then don't propagate this fv to caller, it should mapped instead 5 years ago
  He Wei 7d9a783993 [auto-monad] Support side-effects by auto-monad 5 years ago
  Zhang Qinghua c621a48533 Refactor grad operation implementation. 5 years ago
  lvliang 1490947ff0 adjust-the-location-of-cleaning-unuseless-memory-in-value-node 5 years ago
  zjun 67866474f9 Add pynative second derivative function 5 years ago
  tanghuikang 82450afa9e Optimize memory using in pynative mode 5 years ago
  Zhang Qinghua 829ab4492b Optimize the performance after Grad due to mapping Primitive to K. 5 years ago
  mindspore-ci-bot 0e6d06c463 !9230 Set trace info of Primitive CNode only for equiv out node, not the whole fprop function. 5 years ago
  Zhang Qinghua 8310236ff1 Set trace info of Primitive CNode only for equiv out node, not the whole fprop function. 5 years ago
  mindspore-ci-bot ea191ed5da !8463 Run ops one by one in pynative bp graph 5 years ago
  HulkTang c36b477568 Run ops one by one in pynative bp graph 5 years ago
  huanghui bda74495f3 Add trace mangager around front opt 5 years ago
  mindspore-ci-bot 56a731e48a !8406 [AutoParallel]Add auto parallel pipeline 5 years ago
  lichenever 2e1c43483e add auto parallel pipeline 5 years ago
  Zhang Qinghua bcfa1f72b1 Add debug and trace info for grad operation. 5 years ago
  Yi Huaijie d7faa77b5e support int64 shape 5 years ago
  Zhang Qinghua e3d62123e5 Bind the fvs with single fg bprop for switch layer. 5 years ago
  Zhang Qinghua 2aeff5ba95 Replace primal graph with fprop part of k graph after grad operation. 5 years ago
  es chow 93cd019855 add env_item cache for fv_node 5 years ago
  root 4e85071055 redundant codes clean 5 years ago
  kpy 4338dd266e optimizer pynative graph memory 5 years ago
  buxue 3fd73f9d08 raise exception when use HookBackward in graph mode 5 years ago
  mindspore-ci-bot 88f5cbe5db !5692 Add requires_grad option for python pass 5 years ago
  kingfo cfda024336 fix op id issue in pynative 5 years ago
  BowenK 1bdb26f9e8 Warming up python pass by adding inline passes before it 5 years ago
  mindspore-ci-bot 65819fb50c !5488 heck arg is tensor when graph output is a const or parameter 5 years ago
  buxue 4168c51f0d add check tensor when graph output is a const or parameter 5 years ago
  fary86 fcbb3e0edc Refactor ms_context implementation 5 years ago
  kingfo 1de829aeee resolve output twice out of memory 5 years ago
  zhoufeng 22ec976ef7 fix compilation warnings 5 years ago
  zhoufeng 663278112f optimize code compile performance 5 years ago
  mindspore-ci-bot f875bf21bc !2948 fix control flow 5 years ago
  huangdongrun 2a6d346d2f support if by if grad parameter 5 years ago
  panyifeng 1c296b96c9 fix switch layer sigle prim cell 5 years ago
  liubuyu 45b6600001 mv cc file that end with py to pybind 5 years ago
  liubuyu d81862a916 decoupling core and context 5 years ago
  liubuyu a499d4e47c decoupling core and debug 5 years ago
  liubuyu f4bc0bc9fe move the dependency of utils to core 5 years ago
  kpy 570da089a8 set output value for dynamic graph 5 years ago
  panyifeng 4d4e23fd9e Add bprop for sparse_tensor 5 years ago
  liubuyu 76dc80e7b7 Unified code style 5 years ago
  panyifeng 5a10383cc3 add coo_tensor 5 years ago