bjutsecurity22
/
mindspore2022
forked from huawei/mindspore2022
0
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
8,346 Commits 18 Branches 22 Tags 3.6 GiB
Commit Graph

54 Commits

Author SHA1 Message Date
zhousiyi d0e58dd765 remove ccsrc/common.h 
replace frontend/operator/ops.h in backend with base/core_ops.h as
backend should not use any frontend-only primitive
 2020-08-20 09:34:07 +00:00
gukecai 66e7b02b4b independent stream parallel 2020-08-19 14:16:26 +08:00
zhoufeng 663278112f optimize code compile performance 
Signed-off-by: zhoufeng <zhoufeng54@huawei.com>
 2020-08-18 14:33:24 +08:00
Wei Luning c1c30a44f1 rename param_value -> param_info 2020-08-18 11:16:28 +08:00
gukecai 6362e954df Revert "independent stream parallel" 
This reverts commit adb6ff6c78.
 2020-08-17 20:30:36 +08:00
mindspore-ci-bot e2daa61072 !4514 refactor CreateNewCNode 
Merge pull request !4514 from Margaret_wangrui/refactor_CreateNewCNode
 2020-08-17 15:31:58 +08:00
mindspore-ci-bot dda4c26e7c !4423 add reshape type of tensor 
Merge pull request !4423 from lianliguang/add-reshape-type-to-tensor
 2020-08-17 11:34:11 +08:00
WilliamLian 6760d9976d add reshape type to tensor 2020-08-17 09:34:38 +08:00
Margaret_wangrui 614258dc26 refactor CreateNewCNode 2020-08-15 16:46:08 +08:00
mindspore-ci-bot 162a356fc8 !4345 new ctrl for parallel and iteration num 
Merge pull request !4345 from gukecai/parallel-genmask
 2020-08-14 09:50:36 +08:00
wenchunjiang b24943d496 adapte to remove inline 
merge me commit for remove inline

deal witch multiple cases of switch in ConstructKernelGraph

deal with switch and call cases in ConstructKernelGraph

fix bug and rebase master

ConstructKernelGraph adapte to remove inline

fix InsertMultipleAssignToGraph bug

add graph input to new graph which is created for switch input

replace CreateNewParameterFromCNode to NewParameter in order to set new
parameter's abstract and kernel_info

avoids create a new switch repeatedly when the cnode is a call switch without real input

null pointer check

update frontend code

Revert "update frontend code"

This reverts commit ce1f600d1e9b4b47d9b81122f981bbbe505dd250.

update frontend code PR_2948

fix bug of CheckLabalIndex

handle switch_layer in ConstructKernelGraph

add attr for assign node to avoid erasing by cse pass

cherry-pick ms commit[59b35f690ddcc94ff35a4f4eaf3816121b32235b]:temporary avoid list getitem problem

rebase master

Revert "cherry-pick ms commit[59b35f690ddcc94ff35a4f4eaf3816121b32235b]:temporary avoid list getitem problem"

This reverts commit 74c258f94260ca0769a1ef69c6ef8e831c301dbf.

Revert "handle switch_layer in ConstructKernelGraph"

This reverts commit cb5367f02d69facbca8d39e9234c501608aee27f.

Revert "update frontend code PR_2948"

This reverts commit 234ac583400a96a8ddd641f7a722e1ccd5e056c6.

Revert "merge me commit for remove inline"

This reverts commit 55c0ebd42b6699c7686f5ce585e745f87dd42280.

fix diff after rebase master

doing remove inline in me

overwrite FindNodePrimitive

Revert "doing remove inline in me"

This reverts commit b42e893125bc624d323e855ac6ae615333c06e65.
 2020-08-13 11:51:44 +08:00
gukecai adb6ff6c78 independent stream parallel 2020-08-12 15:58:37 +08:00
mindspore-ci-bot 4142ee010c !4305 fix a sync stream problem in pynative mode 
Merge pull request !4305 from lvchangquan/master
 2020-08-12 10:37:03 +08:00
yujianfeng 6e99eab088 Erase the nodes without primitive value node input from the internal outputs 2020-08-11 21:18:22 +08:00
lvchangquan 9fcba1ef36 fix a sync stream problem in pynative mode 2020-08-11 20:30:44 +08:00
lvchangquan f0e24c9718 fix a bug with wrong call SyncDeviceToHost in pynative mode 2020-08-10 09:40:11 +08:00
mindspore-ci-bot 52689a7dcf !3938 decoupling core and context 
Merge pull request !3938 from liubuyu/master
 2020-08-05 14:16:37 +08:00
liubuyu d81862a916 decoupling core and context 2020-08-05 08:15:14 +08:00
yujianfeng 986408babf ignore create parameter from control depend inputs 2020-08-04 19:00:54 +08:00
mindspore-ci-bot a11e0e35f4 !3872 add internal output tensor 
Merge pull request !3872 from kisnwang/cache-internal-tensor
 2020-08-04 10:46:12 +08:00
ZPaC 4281f38075 Delete hard code in pull kernel. 2020-08-04 09:13:43 +08:00
mindspore-ci-bot be507b96e0 !3473 [refine]change base class of parameter 
Merge pull request !3473 from vlne-v1/change_base_class_of_parameter
 2020-08-03 16:05:46 +08:00
kswang 3ce0f33b27 add internal output tensor 2020-08-03 15:49:27 +08:00
mindspore-ci-bot fa96dfd161 !3774 Sync device data type to return tensor 
Merge pull request !3774 from JoyLvliang/sync-device-data-type-to-return-tensor
 2020-08-03 14:48:09 +08:00
Wei Luning a05c38bb63 make python Parameter inherit from Tensor 2020-08-03 09:12:24 +08:00
lvliang 45745c527e sync-device-data-type-to-return-tensor 2020-08-03 08:56:56 +08:00
kswang e6b0ac9b41 fix nopnode internal output error 2020-07-31 18:57:43 +08:00
liubuyu a499d4e47c decoupling core and debug 2020-07-31 10:39:31 +08:00
kswang 76733ce816 fix cpu multi graph mem error 2020-07-30 09:15:25 +08:00
mindspore-ci-bot 2d9133d773 !3534 Expand make tuple node of graph input 
Merge pull request !3534 from chenfei_mindspore/release-make-tuple-of-load-input-data
 2020-07-27 19:40:29 +08:00
chenfei 69f4c45bb8 get real parameters if graph input is a virtual cnode 2020-07-27 16:29:22 +08:00
liubuyu f4bc0bc9fe move the dependency of utils to core 2020-07-27 10:28:46 +08:00
mindspore-ci-bot f745ae0799 !3496 Fix getting output address of internal output 
Merge pull request !3496 from YuJianfeng/internal_output
 2020-07-26 18:03:54 +08:00
yujianfeng 2409140178 Fix getting output address of internal output 2020-07-25 17:00:06 +08:00
ZPaC 8943cdbc13 Delete parameter name hard code. 2020-07-25 15:54:22 +08:00
mindspore-ci-bot c20cd12216 !3110 Delete deprecated codes of ascend control flow 
Merge pull request !3110 from zhoufeng/delete-deprecated-codes
 2020-07-24 10:38:50 +08:00
zhoufeng 2943cb1c39 delete deprecated codes of ascend control flow 
Signed-off-by: zhoufeng <zhoufeng54@huawei.com>
 2020-07-23 14:07:50 +08:00
mindspore-ci-bot ef18c5f879 !3007 GPU debugger and GPU dump - milestone 1 
Merge pull request !3007 from john_tzanakakis/master_ms1
 2020-07-23 11:52:24 +08:00
yujianfeng 4d18e9ec35 Fix internal multiple outputs check 2020-07-23 08:56:42 +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
mindspore-ci-bot 6d6d2954f0 !3274 Use white list to control pslite compilation 
Merge pull request !3274 from ZPaC/fix-compile-error-with-cpu
 2020-07-21 22:02:04 +08:00
mindspore-ci-bot 69a10fcb4b !3219 serving add acl support, extract common inference interface 
Merge pull request !3219 from 徐永飞/master
 2020-07-21 21:31:35 +08:00
ZPaC 8ff1a9ab47 Use white list to control pslite compilation 2020-07-21 20:29:22 +08:00
xuyongfei 314208633b serving add acl support, extract common inference interface 2020-07-21 17:16:38 +08:00
huanghui 4c34867486 control optimizie for heterogeneous excutor 2020-07-21 14:46:29 +08:00
mindspore-ci-bot e810a68e5a !3173 fix gpu while 
Merge pull request !3173 from kisnwang/fix-gpu-while
 2020-07-18 09:41:35 +08:00
kswang 11f78fb8ad check invalid internal output 2020-07-17 19:35:34 +08:00
yujianfeng 188d74f15e Remove transdata and cast for internal outputs 2020-07-17 16:55:47 +08:00
ZPaC 52022c8013 Enable to train in parameter server mode 2020-07-16 21:39:48 +08:00
mindspore-ci-bot 130cc29603 !2931 Ascend control flow not split graphs 
Merge pull request !2931 from zhoufeng/liantiao1
 2020-07-15 20:57:37 +08:00