sfqcyy
/
mindspore
forked from OSSInnovation/mindspore
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity
7,962 Commits 7 Branches 0 Tags 311 MiB
Commit Graph

63 Commits

Author SHA1 Message Date
chenzupeng 3ec333e851 fix bug in softmax 2020-08-20 17:19:09 +08:00
chenzupeng 605c2b0b16 change opencl code path 2020-08-20 14:51:24 +08:00
wsc 6fe3627878 clean cmake building errors 2020-08-19 19:29:55 +08:00
mindspore-ci-bot 0feb98ae8e !4693 Modify CMakeList file of lite to generate packages. 
Merge pull request !4693 from wangshaocong/lite_cmake
 2020-08-19 17:03:19 +08:00
wsc 27ce4515f2 Modify CMakeList file of lite to generate packages. 2020-08-19 16:03:18 +08:00
xuyongfei a948fe05ab serving: update build.sh to add minddata support for acl 2020-08-19 11:43:58 +08:00
zhangxuetong 9f51871c66 optimize arithmetic 2020-08-14 15:26:26 +08:00
xuyongfei 3554cad325 update serving cpp_client compiler; update serving acl build option 2020-08-13 11:32:27 +08:00
hangq f8d5eb0e93 fix bug that same input tensor will repeat some times in graphIndex 2020-08-11 11:55:29 +08:00
yoni baehr 43d7d3af55 move grad ops to seperate folders 2020-08-09 15:40:32 +03:00
hangq ca6c84b806 modify return type of Model::Import from std::shared_ptr<Model> to Model 
*
 2020-08-08 16:36:01 +08:00
Eric 966438b367 Added dependency packaging 
Fixed issues
 2020-08-07 11:27:26 -04:00
hangq 7ebc7ca801 remove securec 2020-08-07 11:24:40 +08:00
hangq 320d331550 add -e gpu & add -DBUILD_MINDDATA 2020-08-05 10:16:53 +08:00
ervinzhang a53678894a fixed shell format 2020-08-04 14:02:34 -04:00
ervinzhang aec85d299b applied minor fixes 2020-08-04 10:36:11 -04:00
ervinzhang 719823291b rebased again 2020-08-04 09:17:06 -04:00
ervinzhang 50dcb79bdf added MindData lite 2020-08-04 09:16:30 -04:00
pengyongrong d831f8df5e add judgement for CL 2020-08-04 16:35:47 +08:00
wandongdong cfb58a2c8c fix tests compile errors 2020-08-04 10:01:54 +08:00
wandongdong 190fc3cc5a fix tests compile errors 2020-08-03 20:37:18 +08:00
mindspore-ci-bot 1354cf098b !3642 use akg to compile gpu ops rather than _akg 
Merge pull request !3642 from lingyunli63/use_full_akg_to_support_gpu
 2020-08-03 11:53:40 +08:00
lingyunli63 b38d8ee837 use akg.ms instead of _akg 2020-08-03 10:48:11 +08:00
changzherui 106ece221d modify version to r0.6 for master 2020-08-02 17:28:09 +08:00
hangq ac05742619 add run_test.sh & fix bug in type_id.h build.sh gtest 2020-07-31 22:23:10 +08:00
hangq e69f95a854 add increment building & remove redundant code & fix bug in benchmark 2020-07-30 19:00:52 +08:00
hangq 73cab6ec60 fix bug in build.sh for output 2020-07-30 19:00:52 +08:00
hangq b562a6cbf8 remove unused dir & fix bug in taring output .tar & fix bug in build.sh 2020-07-30 15:26:09 +08:00
hangq 174d308cfa add mindspore lite 2020-07-29 03:06:17 +08:00
xsmq 5087945479 回退 'Pull Request !3236 : use akg to compile gpu ops rather than _akg' 2020-07-28 22:41:33 +08:00
mindspore-ci-bot d469a347f8 !3236 use akg to compile gpu ops rather than _akg 
Merge pull request !3236 from lingyunli63/use_full_akg_to_support_gpu
 2020-07-28 15:29:26 +08:00
lingyunli63 42017a19e6 enable compile gpu in akg 2020-07-27 19:20:29 +08:00
John Tzanakakis b3c0eb61d5 GPU debugger - milestone 1 and GPU dump 
Additonal Authors: Adel Shafiei, Harshvardhan Gupta
 2020-07-22 15:58:36 -04:00
xuyongfei 314208633b serving add acl support, extract common inference interface 2020-07-21 17:16:38 +08:00
hexia 20213ee416 add serving option 2020-07-16 22:05:32 +08:00
caifubi cb8b5dbd76 Remove build option -S 2020-07-16 14:51:48 +08:00
caifubi c577952c9a Async Data Dump 2020-07-13 09:42:37 +08:00
ervinzhang bd5a777f81 introducing new C++ API 2020-07-10 16:32:38 -04:00
ZPaC a057639abc Add pslite dependency 2020-07-01 16:26:24 +08:00
mindspore-ci-bot 8c99f35241 !2669 Change comment for akg compilation option 
Merge pull request !2669 from TronZhang/akg_default
 2020-06-29 09:37:20 +08:00
yanghaoran 16976b5f68 update version to 0.5 2020-06-28 21:56:16 +08:00
tronzhang db2e924064 change comment for akg option 2020-06-28 17:52:23 +08:00
zhoufeng b7a50b24d6 Disable cuda9.2, use cuda10.1 as default 
Signed-off-by: zhoufeng <zhoufeng54@huawei.com>
 2020-06-23 22:35:23 +08:00
Shida He 4c056855e0 Implementation for mindspore debugger 2020-06-22 17:15:30 -04:00
mindspore-ci-bot f1a69de0b6 !2405 change some comment name in the whole project 
Merge pull request !2405 from chenzhongming/master
 2020-06-22 14:40:25 +08:00
chenzomi a834a6308e change some comment name in the whole project 2020-06-22 12:59:21 +08:00
Tron Zhang c01569f0e0 Change akg as on default for D backend 2020-06-22 10:26:26 +08:00
dinghao 8d76c708df init serving 2020-06-21 18:36:53 +08:00
gong chen a6dfa281ea Init GraphKernel. 
- It provides a unified style to express graph and kernel for user.
- It provides a unified IR to represent graph and kernel for developer.
- It breaks the boundary between graph and kernel.
- It provides more opportunities to do compile optimization.
 2020-06-20 22:31:54 +08:00
chenjianping 6034f9c1e2 support host reduce scatter and mpi config 2020-06-15 16:33:40 +08:00