283 Commits (bada826b18f3e4dc8aac0f82af3457794de9c7bb)

Author SHA1 Message Date
  yuchaojie 61bf4b18a2 fix_consecutive_allreduce_bug 5 years ago
  mindspore-ci-bot 5686315199 !4557 change profiling log level 5 years ago
  mindspore-ci-bot e2a63790bd !4731 genmask parallel and fix iter num 5 years ago
  mindspore-ci-bot 55bd09c689 !4684 add log for gpu profiler 5 years ago
  gukecai 66e7b02b4b independent stream parallel 5 years ago
  askmiao 5a817d7444 add log and modify log level for gpu profiler 5 years ago
  caifubi 6eba89b1bc fix profiling log bug 5 years ago
  mindspore-ci-bot 58523a41fe !4392 use builtin float16 for arm 5 years ago
  mindspore-ci-bot 8f6ed032e5 !4428 Operation Overflow Watchpoint for D-Chip debugger 5 years ago
  mindspore-ci-bot 1ca715c7e7 !4656 fix redundancy print 5 years ago
  jjfeing 16e67ff143 fix redundancy print 5 years ago
  zhoufeng 663278112f optimize code compile performance 5 years ago
  mindspore-ci-bot 5f3e7aa6b1 !4652 block transdata to change format for now 5 years ago
  lvchangquan 0b63a1ffe4 block transdata to change format for now. 5 years ago
  mindspore-ci-bot 3fb58fcbe4 !4585 add gpu nccl broadcast 5 years ago
  mindspore-ci-bot 19192e75e3 !4609 change unsupport to unsupported 5 years ago
  Adel Shafiei 4834a3378b Op Overflow Watchpoint support for D-chip debugger 5 years ago
  chenzomi bb125cb309 change unsupport to unsupported 5 years ago
  gukecai 6362e954df Revert "independent stream parallel" 5 years ago
  baihuawei b9ebd9c280 add gpu nccl broadcast 5 years ago
  zhousiyi e1aa49a4b7 use built-in float16 in arm_neon.h for lite arm 5 years ago
  mindspore-ci-bot 9f635e52c7 !4463 change some wrong log about static memory 5 years ago
  liangzelang 1608c4d096 change-static-memory-size-log 5 years ago
  mindspore-ci-bot b48c1f45f0 !4236 fix gpu heterogeneous bug 5 years ago
  baihuawei 517fcc16ee fix gpu heterogeneous 5 years ago
  mindspore-ci-bot 21014fd624 !4235 add gpu profiler feature 5 years ago
  mindspore-ci-bot 15c533d481 !4277 Profiling Support Multi Graph 5 years ago
  askmiao 25cae1a2e7 add profiler featrue 5 years ago
  gukecai adb6ff6c78 independent stream parallel 5 years ago
  John Tzanakakis 3569513232 fix d-chip wacthpoints, latest value for GPU inputs 5 years ago
  caifubi 09946fcad5 Support Profiling With Graph Only Have Hccl Op 5 years ago
  mindspore-ci-bot 8040e8bf89 !4130 modify some bug and add test case for gpu dropout op 5 years ago
  hanhuifeng2020 ab6f7420b5 modify some bug and add test case for gpu dropout op 5 years ago
  kswang 69c096c2a6 dlopen cpu mpi adapter 5 years ago
  mindspore-ci-bot 21dfac0432 !4105 mem pools expands from high addr, dynamic mem expands from low addr 5 years ago
  mindspore-ci-bot 64721fa57e !4104 fix a bug with using dynamic memory 5 years ago
  mindspore-ci-bot 7280d3170a !3768 GPU debugger grpc implementation and smart kernel read 5 years ago
  liangzelang fe1f36ea5c mem pools expands from high addr, dynamic mem expands from low addr 5 years ago
  mindspore-ci-bot 1f28a7c097 !4063 Decouple ME and AKG for GPU. 5 years ago
  lvchangquan 87022fce3c fix a bug with using dynamic memory. 5 years ago
  Zhang Qinghua 22e0a0ba76 Decouple ME and AKG for GPU. 5 years ago
  mindspore-ci-bot 8908f6ef19 !4098 Graph compile performance optimization 5 years ago
  zhoufeng 2f5cbfc26f graph compile performance optimize 5 years ago
  lvliang 3a61d646d4 decoupling-the-interface-of-mallocing-mem 5 years ago
  mindspore-ci-bot ba3a2976dc !4038 support ops cholesky for resnet50 thor gpu 5 years ago
  mamba_ni 96642a76fd support cusolver AND OPS cholesky_solve 5 years ago
  kswang 51e9fbf973 format ompi 5 years ago
  mindspore-ci-bot 7b9478aae9 !3989 change parameter's device dtype to infer 5 years ago
  mindspore-ci-bot 52689a7dcf !3938 decoupling core and context 5 years ago
  zhoufeng ca7154a548 graph compile performance optimization 5 years ago