zyf1234
/
mindspore2022
forked from huawei/mindspore2022
0
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
26,922 Commits 18 Branches 22 Tags 3.6 GiB
Commit Graph

151 Commits

Author SHA1 Message Date
i-robot a7dfad423e !19693 fix switch layer call to call 
Merge pull request !19693 from gaoyong10/runtime1.3
 2021-07-09 03:18:09 +00:00
lizhenyu 1d81102fd0 Remove nop node optimizer for cpu 2021-07-08 17:59:42 +08:00
gaoyong10 e51ddced63 Fix switch layer call to call 2021-07-08 16:10:21 +08:00
i-robot bf1ee2db87 !19632 [bugfix] custom bprop got unexpected input 
Merge pull request !19632 from zyli2020/r1.3_mindrt
 2021-07-08 02:03:41 +00:00
i-robot e9beb7978f !19544 Fit Gpu LoopCount for profiler module (add patch) 
Merge pull request !19544 from gzhcv/r1.3Profiler
 2021-07-08 01:19:45 +00:00
lizhenyu 293267ea48 [bugfix] custom bprop got unexpected input 2021-07-07 23:27:38 +08:00
gzhcv 94f28a2da3 Fit Gpu LoopCount for profiler module v2 2021-07-07 09:36:34 +08:00
lizhenyu dadfe54ced [bugfix] Heterogenous scenario in a ms function of PyNative Mode occur core dump 2021-07-06 21:30:33 +08:00
i-robot 88664ea17a !19370 Adjust CPU optimize graph pass 
Merge pull request !19370 from zyli2020/mindrt_debug
 2021-07-05 21:09:43 +00:00
i-robot ede0139b06 !19414 fix the control ring bug 
Merge pull request !19414 from limingqi107/bug_fix2
 2021-07-05 17:55:44 +00:00
limingqi107 dbf2030d90 fix the control ring bug 2021-07-05 21:51:32 +08:00
lizhenyu efcf910b80 Adjust CPU optimize graph pass 2021-07-05 17:52:46 +08:00
limingqi107 85222572f3 fix bug of host and device share weight 2021-07-05 17:08:39 +08:00
i-robot a09f38399f !19344 fix bug of host device from different graph 
Merge pull request !19344 from limingqi107/bug_fix2
 2021-07-04 16:58:51 +00:00
i-robot 4bb2474235 !19340 optimize the multi device execution performance 
Merge pull request !19340 from limingqi107/bug_fix
 2021-07-04 15:02:50 +00:00
limingqi107 633e18df4d fix bug of host device from different graph 2021-07-04 22:26:02 +08:00
limingqi107 05be63887c optimize the multi device execution performance and communication node running order 2021-07-04 20:42:58 +08:00
limingqi107 e761655a42 actor runtime support CPU dump 2021-07-04 11:19:09 +08:00
i-robot 8008103050 !19338 fixed sponge net 
Merge pull request !19338 from gaoyong10/new_runtime19
 2021-07-03 16:08:48 +00:00
i-robot 427e1077a6 !19329 fix bug of actor runtime host device multi-graphs 
Merge pull request !19329 from limingqi107/bug_fix
 2021-07-03 14:07:12 +00:00
gaoyong10 b3f0a29fb1 Fix Sponge net. 2021-07-03 21:44:11 +08:00
limingqi107 8f851f0c98 send output result must be before output data 2021-07-03 20:15:57 +08:00
limingqi107 c198c47beb fix bug of actor runtime host device multi-graphs 2021-07-03 18:59:44 +08:00
gaoyong10 03021379d4 Fix ocean and sponge net. 2021-07-03 16:22:00 +08:00
i-robot 46377a98b5 !19274 add the device id when memory not enough 
Merge pull request !19274 from limingqi107/bug_fix
 2021-07-02 21:28:12 +00:00
i-robot d0bab79fde !19252 add dump pb and nccl protected 
Merge pull request !19252 from limingqi107/bug_fix3
 2021-07-02 12:00:45 +00:00
limingqi107 04d45861b5 add the device id when memory not enough 2021-07-02 13:07:29 +08:00
limingqi107 70d156562a add dump pb and nccl protected 2021-07-02 11:25:43 +08:00
limingqi107 823fb50be5 ignore the internal parameter output in actor runtime 2021-07-01 21:56:14 +08:00
i-robot ea82e018e8 !19076 Optimize PyNative performance 
Merge pull request !19076 from zyli2020/fix_issue_defect
 2021-07-01 13:42:16 +00:00
lizhenyu a487f8c897 Optimize PyNative Performance 2021-07-01 16:04:08 +08:00
gaoyong10 e56fe7933b Fix weight input of gather actor 2021-07-01 10:10:36 +08:00
i-robot 26d1157d00 !19137 link auto monad for gather actor. 
Merge pull request !19137 from gaoyong10/new_runtime17
 2021-06-30 18:57:09 +00:00
i-robot d3ebec95a5 !19147 use the thread spin to optimize actor runtime performance 
Merge pull request !19147 from limingqi107/bug_fix
 2021-06-30 13:14:39 +00:00
i-robot bde38a582c !19117 SetDebugger for MindRTBackend and clean debug_actor code 
Merge pull request !19117 from parastooashtari/new_unified_gpu
 2021-06-30 09:25:54 +00:00
gaoyong10 7481779e95 link auto monad for gather actor 2021-06-30 15:14:05 +08:00
limingqi107 1bd99336af use the thread spin to optimize actor runtime performance 2021-06-30 14:55:13 +08:00
i-robot 86e45f6106 !19112 fix actor thread pool 
Merge pull request !19112 from yangjie159/master
 2021-06-30 00:41:52 +00:00
Parastoo Ashtari 4e4be17648 SetDebugger for MindRTBackend and clean debug_actor code 2021-06-29 12:25:11 -04:00
yangjie159 6b3e10efe0 fix actor thread pool 2021-06-30 00:01:58 +08:00
i-robot 442d73b1ac !19071 enable mindRT 
Merge pull request !19071 from limingqi107/actor_runtime
 2021-06-29 14:37:10 +00:00
limingqi107 406c252834 enable mindRT 2021-06-29 17:44:05 +08:00
gaoyong10 8d1b5781e7 Fetch backend node for switch actor. 2021-06-29 16:24:55 +08:00
i-robot 29e7da4c3e !18941 Fix mulit call in control flow 
Merge pull request !18941 from gaoyong10/new_runtime13
 2021-06-29 06:42:01 +00:00
i-robot 4fa7da3de5 !18990 fix the bug of actor runtime in the scene of kLoadRealInput is not parameter 
Merge pull request !18990 from limingqi107/actor_runtime2
 2021-06-29 06:37:14 +00:00
Parastoo Ashtari e2a4172560 make gpu dump truly async 2021-06-28 16:44:35 -04:00
limingqi107 5d1c8dc4f5 fix the bug of actor runtime in the scene of kLoadRealInput is not parameter 2021-06-28 20:42:25 +08:00
gaoyong10 a9eaebc598 Fix control flow recursion 2021-06-28 14:24:07 +08:00
i-robot 04445d4902 !18669 actor runtime message running performance optimizer 
Merge pull request !18669 from limingqi107/actor_runtime2
 2021-06-28 03:29:52 +00:00
limingqi107 ce68aff167 actor runtime message running performance optimizer 2021-06-28 09:14:33 +08:00