Nigel
/
mindspore
forked from mindspore-Ecosystem/mindspore
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity
49,878 Commits 24 Branches 0 Tags 3.9 GiB
Commit Graph

49878 Commits

Author SHA1 Message Date
yefeng c3a594021e shared thread pool for ModelparallelRunner 2022-12-15 15:30:00 +08:00
i-robot 5844b6924e
!46226 [lite]optimize big-matmul for arm64 
Merge pull request !46226 from 徐安越/r1.8_4
 2022-12-14 02:38:37 +00:00
i-robot 3f9939c009
!46520 [lite]fix threadpool-manager for non-mindrt 
Merge pull request !46520 from 徐安越/r1.8_3
 2022-12-14 02:36:19 +00:00
xuanyue 708cacbe03 optimize big-matmul for arm64 2022-12-13 10:01:01 +08:00
xuanyue 31c8757899 fix threadpool-manager for non-mindrt 2022-12-13 09:55:24 +08:00
i-robot f71cde2125
!46084 [MSLITE][CPU] matmul avx512 mask instruction opt 
Merge pull request !46084 from Greatpan/matmul_avx512_opt_r1.8
 2022-12-12 11:08:21 +00:00
i-robot 3e086209db
!46457 [MSLITE] Fix resize precision problems 
Merge pull request !46457 from zhangyongxian/dev_zhangyongxian_rs
 2022-12-06 07:08:29 +00:00
zhangyongxian 6ab0310b73 [MSLITE] Fix resize problems 2022-12-06 11:15:32 +08:00
i-robot c337b69c55
!46365 [MS][LITE]Add Sigmod ut 
Merge pull request !46365 from gongdaguo1/add_sigmod_ut
 2022-12-06 01:18:10 +00:00
gongdaguo1 047805959a add sigmod ut 2022-12-05 10:04:16 +08:00
i-robot 762730048b
!46396 [MSLITE] Fix coredump in multidevice inference 
Merge pull request !46396 from zhangyongxian/dev_zhangyongxian_coredump
 2022-12-04 10:05:07 +00:00
zhangyongxian e0113814d6 [MSLITE] Fix bug for multi deivce coredump 2022-12-04 03:55:43 +08:00
i-robot b3a2be2601
!46207 [MS][LITE]optimized kernel actor 
Merge pull request !46207 from gongdaguo1/optimized_kernel_actor_33
 2022-12-02 08:30:50 +00:00
i-robot b02dbda393
!46342 [MS][LITE][parallel predict]bug fix 
Merge pull request !46342 from yefeng/471-bug_fix_b330
 2022-12-02 02:36:50 +00:00
i-robot 7cf8c46e31
!46304 [MSLITE] Support Constant input for GPU shuffle op 
Merge pull request !46304 from zhangyongxian/dev_zhangyongxian_fixshuff
 2022-12-01 13:55:51 +00:00
yefeng d515f3b563 bug fix 2022-12-01 21:03:49 +08:00
i-robot f3f98e042f
!46316 [lite]restore tensor default attr 
Merge pull request !46316 from 徐安越/r1.8_3
 2022-12-01 11:48:00 +00:00
gongdaguo1 9c1d4b2cac optimzed kernel actor 2022-12-01 19:36:24 +08:00
xuanyue 9763d6f0cc restore tensor default attr 2022-12-01 15:31:17 +08:00
i-robot 02512c0539
!45750 [lite]add st examples 
Merge pull request !45750 from 徐安越/r1.8_3
 2022-11-29 14:01:20 +00:00
i-robot 331b35fabc
!46182 [lite]Complete initialization for tensor to avoid unexpected problems 
Merge pull request !46182 from 徐安越/r1.8_1
 2022-11-29 13:55:33 +00:00
xuanyue 6ef7ea2354 Complete initialization for tensor to avoid unexpected problems 2022-11-29 12:37:25 +08:00
i-robot a56dbf8e89
!46116 [MS][LITE][STABLE]change alloc unit to 256M 
Merge pull request !46116 from chenjianping/r1.8_dev2
 2022-11-29 01:53:57 +00:00
i-robot fb614756fe
!46034 [LITE] optimize gelu tensorrt op 
Merge pull request !46034 from WangWenzhe/r1.8_gelu
 2022-11-29 01:05:59 +00:00
jpc_chenjianping 13ab16045b change mem alloc unit to 256M 2022-11-28 14:38:34 +08:00
i-robot 6589360aa6
!46087 [lite]fix core_dump due to malloc(0) 
Merge pull request !46087 from 徐安越/r1.8_1
 2022-11-28 02:18:59 +00:00
xuanyue 42bdd385d6 add st examples 2022-11-28 10:18:24 +08:00
greatpanc a83e66293f matmul avx512 opt r1.8 2022-11-28 07:30:07 +08:00
i-robot a64b3728ff
!46090 [MS][LITE][STABLE]optimize scheduler performance 
Merge pull request !46090 from chenjianping/r1.8_dev2
 2022-11-27 03:21:35 +00:00
xuanyue f52620a4e9 fix core_dump due to malloc(0) 2022-11-27 09:55:11 +08:00
jpc_chenjianping 0b9a1d401b optimize scheduler performance 2022-11-26 17:54:55 +08:00
i-robot d97b2f4e1a
!46068 [MS][LITE]Fix kernel actor name 
Merge pull request !46068 from gongdaguo1/r18_add_parallel_use_kernels_message
 2022-11-26 07:15:33 +00:00
gongdaguo1 284393b061 code check 2022-11-26 11:46:41 +08:00
i-robot 335ab321bb
!46049 [MS][LITE][STABLE]optimize op performance 
Merge pull request !46049 from chenjianping/r1.8_dev2
 2022-11-26 01:43:36 +00:00
i-robot e55a5c113e
!45821 [MS][LITE]Add kernel parallel 
Merge pull request !45821 from gongdaguo1/r18_add_parallel_use_kernels_message
 2022-11-25 12:35:26 +00:00
jpc_chenjianping 495afd6d71 optimize op performace 2022-11-25 20:11:12 +08:00
wangwenzhe 769a78b223 optimize gelu tensorrt op 2022-11-25 15:12:25 +08:00
i-robot 255fe929e5
!45858 [MS][LITE][RUNNER] bugfix 
Merge pull request !45858 from yefeng/469-fix_bug
 2022-11-23 09:28:43 +00:00
gongdaguo1 6b36380b14 add parallel kernel 2022-11-22 19:35:49 +08:00
i-robot 18a402eda5
!45850 stride slice bug 
Merge pull request !45850 from lianliguang/r1.8
 2022-11-22 11:01:11 +00:00
yefeng 806ceab605 bug fix 2022-11-22 15:07:55 +08:00
i-robot 4b98342ca1
!45751 [lite]optimize inferShape 
Merge pull request !45751 from 徐安越/r1.8_1
 2022-11-22 06:11:52 +00:00
i-robot 3529d1bfcf
!45848 modify the rst file in 1.8 
Merge pull request !45848 from 宦晓玲/code_docs_1122.8
 2022-11-22 06:05:16 +00:00
lianliguang b4714ae736 fix some bug of load grad_nn_ops 2022-11-22 11:58:57 +08:00
huanxiaoling 04dfa6a80e modify the rst file in 1.8 2022-11-22 11:55:41 +08:00
i-robot ffe72b7586
!45818 [MS][LITE] code check 
Merge pull request !45818 from jianghui58/dev_r1.8
 2022-11-22 01:24:31 +00:00
i-robot dafee4a6a1
!45568 [MSLITE] Support dynamic stridedslice begin 
Merge pull request !45568 from zhangyongxian/dev_zhangyongxian_dynamicslicebegin
 2022-11-21 13:54:27 +00:00
i-robot 702f270751
!45712 [LITE] Fix OCR model accuracy bug 
Merge pull request !45712 from WangWenzhe/r1.8_voice
 2022-11-21 12:54:35 +00:00
jianghui58 f4ea446e0d code check 2022-11-21 20:09:00 +08:00
张勇贤 f6c844f27d [MSLITE] Support dynamic stride slice begin 2022-11-21 18:43:37 +08:00