Zhang Qinghua
a137fa1d0b
Optimize the Executors routines.
- Fix the key generating.
- Distinguish the executors.
4 years ago
i-robot
0531edfa67
!20918 Eliminate forward CNode in grad graph decorated by ms_function
Merge pull request !20918 from JoyLvliang/eliminate_forward_cnode_in_grad_graph_decorated_by_ms_function
4 years ago
i-robot
47fe617ac2
!22419 [AutoParallel]Fix AutoParallel predict
Merge pull request !22419 from lichen/fix_auto_parallel_predict
4 years ago
7347157+joylvliang@user.noreply.gitee.com
0fb07a6377
eliminate_forward_cnode_in_grad_graph_decorated_by_ms_function
4 years ago
i-robot
e6e1f37ae4
!22346 [Core] Fix the bug of scope setting when cloning nodes
Merge pull request !22346 from Xiaoda/86-fix-the-fullname-scope-bug
4 years ago
lichenever
c8da805eea
fix_auto_parallel_predict
4 years ago
i-robot
8c6d4a05fc
!22397 Add JVP for forward mode auto diff.
Merge pull request !22397 from LiangZhibo/fwd2
4 years ago
l00591931
c07c286fcd
Add forward mode auto diff
4 years ago
i-robot
07eaa1969b
!22344 fix compile error when weight is used in while condition
Merge pull request !22344 from xychow/fix-while-loop-with-weight-in-condition
4 years ago
Xiaoda Zhang
b2703879c6
fix the scope setting error when cloning nodes
4 years ago
i-robot
5e13353f27
!22368 [AutoParallel]Rectification_module_import
Merge pull request !22368 from lichen/rectification_module_import
4 years ago
i-robot
9b28bd6308
!22342 [Auto parallel] Adjust the device matrix of OneHot operator
Merge pull request !22342 from Xiaoda/85-adapt-onehot
4 years ago
lichenever
ac1cb63f3c
rectification_module_import
4 years ago
Xiaoda Zhang
ae2d3b4832
adjust the device matrix for onehot use in moe
4 years ago
i-robot
0adacd1b68
!22226 [AutoParallel]Refacor pipeline code
Merge pull request !22226 from lichen/Refactor_part_of_pipeline
4 years ago
zhousiyi
44bdcb101c
fix while loop compile error when weight is used in while condition
4 years ago
i-robot
3d5d3defc9
!22083 Fix auto_monad bug and ensure compilation performance is not degraded
Merge pull request !22083 from Margaret_wangrui/fix_auto_monad_bug
4 years ago
lichenever
5812076512
Refactor_part_of_pipeline
4 years ago
i-robot
6c7ffb7f1d
!22145 refactor updatestate_eliminater
Merge pull request !22145 from huangbingjian/refactor_updatestate
4 years ago
i-robot
1e47ff7bc3
!22108 modify check strategy for conv2d
Merge pull request !22108 from yangzhenzhang/modify-check-strategy-for-conv2d
4 years ago
Margaret_wangrui
79d3b3c55c
Fix auto_monad bug and ensure compilation performance is not degraded
4 years ago
yangzhenzhang
f1afaeac5a
modify check strategy for conv2d
4 years ago
huangbingjian
d7687994db
refactor updatestate_eliminater
4 years ago
yujianfeng
712b9bd013
convert some ops bprops to mindir
4 years ago
ms_yan
36a8886ca2
Revert "[feat] [assistant] [I3T96T] add new Dataset operator CMUARCTICDataset"
This reverts commit b077aa1cab .
Revert "[feat] [assistant] [I3T96X] add new Dataset operator LibriSpeechDataset"
This reverts commit 4e6f7dc97d .
delete pass_registry_test.cc
comment hiai_nlu_model_multi.pb related line
4 years ago
djc
b077aa1cab
[feat] [assistant] [I3T96T] add new Dataset operator CMUARCTICDataset
4 years ago
djc
4e6f7dc97d
[feat] [assistant] [I3T96X] add new Dataset operator LibriSpeechDataset
4 years ago
zjun
35aab6144d
Fix pynative memory leak
Signed-off-by: zjun <zhangjun0@huawei.com>
4 years ago
i-robot
a6fade32b1
!22072 Clean Code for compiler
Merge pull request !22072 from YuJianfeng/clean
4 years ago
i-robot
954864bfba
!22037 incoroporate getitem if indexed node is zeros_like
Merge pull request !22037 from xychow/always-incorporate-getitem-if-zeros-like
4 years ago
yujianfeng
e2091f09b4
Clean Code for compiler
4 years ago
huanghui
678c49a606
fix pylint and codecheck at 0819
4 years ago
zhousiyi
dcd374961e
incoroporate getitem if indexed node is zeros_like so AddN(MakeTuple(x, zeros_like)) can be optimized
4 years ago
i-robot
415275ae17
!21805 support adafactor model parallel
Merge pull request !21805 from yangzhenzhang/auto-parallel-support-adafactor-opt
4 years ago
i-robot
a77a0b968d
!21761 comm_recompute_interface.
Merge pull request !21761 from yao_yf/comm_recompute_interface
4 years ago
yangzhenzhang
7ca64d2235
auto parallel support adafactor opt
4 years ago
yao_yf
5277b229be
add cell comm recompute interface
4 years ago
huangbingjian
c3bb0b91b4
check call/switch/switchlayer in EliminateUpdateStateForPureNode
4 years ago
Margaret_wangrui
dc1b5cf1bb
Do not do replacement processing when the value of the parameter may be modified in the attach node of updatestate
4 years ago
i-robot
69e8603a2b
!21774 handle replace in updatestate_eliminate
Merge pull request !21774 from huangbingjian/replace_dev
4 years ago
i-robot
530da3e37e
!21527 fix_reshape_reshape_in_auto_parallel
Merge pull request !21527 from yao_yf/fix_reshape_reshape_in_auto_parallel
4 years ago
huangbingjian
88059b76ee
handle replace in updatestate_eliminate
4 years ago
zhousiyi
94e5fe6242
keep tuple_getitem as much as possible to reduce the number of func graphs
shrink output of func_graph other than set unused to dead node
4 years ago
yao_yf
e233880e41
fix reshape depend reshape in auto parallel
4 years ago
i-robot
6320bf6b7b
!21676 check strategy for conv2d
Merge pull request !21676 from yangzhenzhang/check-strategy-for-conv2d
4 years ago
i-robot
50d54a7482
!20787 Memory leak fix.
Merge pull request !20787 from zhangzhaoju/master_leak_fix
4 years ago
yangzhenzhang
d18c813ee4
check strategy for conv2d
4 years ago
zhangzhaoju
c7e6ad4f7d
Add swtich to controll cache or not cache graph for ad
4 years ago
i-robot
e2c8c8f385
!21583 [MS][parallel]Support user configured full batch dataset loading
Merge pull request !21583 from Chong/PanGu
4 years ago
i-robot
322bb37070
!21045 pangu_train_and_eval
Merge pull request !21045 from yao_yf/pangu_train_and_eval
4 years ago