Bessie
/
mindspore2022
forked from huawei/mindspore2022
0
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity
28,297 Commits 18 Branches 0 Tags 564 MiB
Commit Graph

133 Commits

Author SHA1 Message Date
zhaosida 00118a1467 code clean master 2021-08-03 11:07:10 +08:00
kswang 9e8156b020 clean code 2021-07-27 10:30:21 +08:00
ZPaC a9a0f590e6 Fix master static check 2021-07-15 20:42:42 +08:00
lvchangquan 09c54c7923 use op transdata to improve performance in print process 2021-07-06 15:09:18 +08:00
i-robot b2111fcdf5 !18909 modify log level for pynative ascend 
Merge pull request !18909 from chujinjin/modify_log_level_for_pynative_ascend
 2021-06-28 01:13:22 +00:00
chujinjin ac3670ef91 modify log level for pynative ascend 2021-06-26 14:37:20 +08:00
yuchaojie bf13a8a01e support group conv2d in pynative 2021-06-26 11:48:06 +08:00
i-robot 32a54eb4eb !15330 add updated paramter for graph 
Merge pull request !15330 from kisnwang/add-updated-paramter-for-graph
 2021-06-25 02:10:06 +00:00
kswang 42155e7e5c add updated parameter for graph 2021-06-24 17:19:42 +08:00
i-robot 0966595cfd !18538 support FracZ format conversion when conv2d group > 1 
Merge pull request !18538 from yuchaojie/op_select
 2021-06-23 07:14:19 +00:00
yuchaojie 689158f79b FracZ format conversion when conv2d group > 1 2021-06-23 09:32:16 +08:00
kswang 3247c00555 optimize heter memcpy 2021-06-21 21:34:53 +08:00
LaiYongqiang d4d6fb940d memory reuse code clean 2021-06-18 09:41:55 +08:00
chujinjin 1dd950218b modify log level for pynative 2021-06-08 11:16:05 +08:00
John Tzanakakis 7d460de8b4 adapt offline debuger to new dump, sync support for iteration ranges, GPU default dump mode correction 2021-06-04 08:38:50 -04:00
jiangshuqiang 6154f957fb fix static check 2021-05-31 16:47:21 +08:00
zhaosida d2be1b1bd7 code clean master 2021-05-29 09:32:48 +08:00
mindspore-ci-bot 41eaf3f58d !16708 Handle GraphKernel node in PSEmbeddingCache 
From: @dayschan
Reviewed-by: @limingqi107,@gaoxiong1
Signed-off-by: @gaoxiong1
 2021-05-24 09:04:22 +08:00
liuxiao93 2a3a787049 add host_format for device_info of MetaTensor. 2021-05-22 14:44:11 +08:00
dayschan 93937537b2 Bugfix in PSEmbeddingCache, handle GraphKernel nodes 2021-05-21 16:55:44 +08:00
mindspore-ci-bot 84607e3a51 !15890 fix an allreduce calculate bug in pynative mode 
From: @lvchangquan
Reviewed-by: 
Signed-off-by:
 2021-05-11 09:43:28 +08:00
lvchangquan 0b09fdf94c fix an allreduce bug with two streams sync problem 2021-05-10 10:00:58 +08:00
ZPaC 12f95b51f4 Add server code part2 2021-05-06 20:36:04 +08:00
Margaret_wangrui b39b3d7dd5 fix bot warnings 2021-04-26 16:23:55 +08:00
mindspore-ci-bot b7554d574f !15590 fix addr is null 
From: @jjfeing
Reviewed-by: @zhoufeng54,@kisnwang
Signed-off-by: @kisnwang
 2021-04-26 09:31:12 +08:00
jjfeing 77e9b0c913 fix addr is null 2021-04-25 10:37:11 +08:00
lvchangquan bbd4ff9346 fix some pclint warnings in master 2021-04-22 17:06:16 +08:00
mindspore-ci-bot 0fe493e205 !15161 support hbm perfermance test 
From: @laiyongqiang
Reviewed-by: @jjfeing,@majorzhang
Signed-off-by: @majorzhang
 2021-04-16 20:28:40 +08:00
laiyongqiang de0f6edf5a support hbm profile 2021-04-16 18:01:31 +08:00
mindspore-ci-bot d96420553b !14843 Memory reuse switch 
From: @ding_fei_fei
Reviewed-by: @kingxian,@kingxian
Signed-off-by: @kingxian,@kingxian
 2021-04-16 14:57:17 +08:00
dingpeifei 8d830f1e5c Memory reuse switch 2021-04-15 14:33:44 +08:00
limingqi107 b3a5ccebc3 fix codedex 2021-04-14 09:32:56 +08:00
tanghuikang c8a14ba016 Clean code 2021-04-09 09:28:53 +08:00
lizhenyu a87b141cf9 init device address for anf node 2021-04-07 17:33:22 +08:00
mindspore-ci-bot d4a0df8a94 !14483 Orederd value nodes before allocate memory 
From: @laiyongqiang
Reviewed-by: @jjfeing,@zhoufeng54
Signed-off-by: @zhoufeng54
 2021-04-01 11:40:03 +08:00
laiyongqiang 4526ce6845 fix reallocate memory bug for communication op 2021-03-31 17:24:39 +08:00
laiyongqiang 4bf386aa9f orederd value nodes before allocate memory 2021-03-31 16:14:24 +08:00
kswang bfc51f96ce support launch kernel for graph mode 2021-03-27 09:50:03 +08:00
tanghuikang e8382c6187 Ingnore transpose with nop_op attr in pynative infer 2021-03-23 09:30:49 +08:00
zhoufeng b7e5f956e5 workspace of comm op can be reused 
Signed-off-by: zhoufeng <zhoufeng54@huawei.com>
 2021-03-16 14:42:52 +08:00
mindspore-ci-bot fb31e724aa !13093 Enable tbe dynamic shape op for run op 
From: @HulkTang
Reviewed-by: @zhoufeng54,@jjfeing
Signed-off-by: @jjfeing
 2021-03-11 13:48:32 +08:00
mindspore-ci-bot c69142fdc1 !12968 update reshape type for 3d nodes 
From: @liubuyu
Reviewed-by: 
Signed-off-by:
 2021-03-10 20:33:04 +08:00
tanghuikang 7138bf66f6 Enable tbe dynamic shape op for run op 2021-03-10 15:17:01 +08:00
liubuyu 518818fbef reshape type for 3d nodes 2021-03-10 08:56:10 +08:00
tanghuikang dac64f30ee Support ms_function + heterogenous 2021-03-09 18:59:03 +08:00
caifubi 171b468bb3 PyNative AllReduce Bucket 2021-03-03 15:47:18 +08:00
He Wei 7d9a783993 [auto-monad] Support side-effects by auto-monad 
The basic idea is: exploits data dependency to control the execution order
of side-effect operations, and keep the semantics of ANF unchanged.

The ControlDepend primitive is removed and there are two primitives added:

1. UpdateState:
```
  a = Assign(para, value)
```
became:
```
  a = Assign(para, value, u)
  u = UpdateState(u, a)
```

2. Load:
```
  x = Add(para, value)
```
became:
```
  p = Load(para, u)
  x = Add(p, value)
  u = UpdateState(u, p)
```
 2021-02-08 09:01:15 +08:00
mindspore-ci-bot 025fec757c !12087 optimize the memory alloc error info 
From: @limingqi107
Reviewed-by: @kisnwang,@cristoval
Signed-off-by: @cristoval
 2021-02-04 20:23:55 +08:00
mindspore-ci-bot 0d24b45a71 !12017 add ps cache mode check 
From: @zyli2020
Reviewed-by: @cristoval,@limingqi107
Signed-off-by: @limingqi107
 2021-02-04 18:55:59 +08:00
limingqi107 366f3e668d optimize the memory alloc error info 2021-02-04 15:30:01 +08:00