yujianfeng
|
266e960acb
|
Not do cse for the nodes set recomputed before recompute pass
|
5 years ago |
xsmq
|
73b7154e55
|
fix cmakelint error
|
5 years ago |
yujianfeng
|
08e00b20a9
|
Move the whole graph nodes firstly for the single used graph when inline
|
5 years ago |
chenfei
|
874f350eb9
|
add dump level
|
5 years ago |
mindspore-ci-bot
|
21addb331d
|
!11009 Fix continuous calls for recompute api
From: @ginfung
Reviewed-by: @zhunaipan,@zh_qh
Signed-off-by: @zh_qh
|
5 years ago |
yujianfeng
|
7cac3d3a47
|
Fix continuous calls for recompute api
|
5 years ago |
mindspore-ci-bot
|
1c942ce49f
|
!10221 support grad accumulation for auto parallel
From: @yangzhenzhang
Reviewed-by:
Signed-off-by:
|
5 years ago |
yujianfeng
|
ba311cf57f
|
Add cnode attr for recomputation
|
5 years ago |
yangzhenzhang
|
9da3f9bec9
|
mini step grad accumulation
|
5 years ago |
mindspore-ci-bot
|
275da50b32
|
!10474 Add recomputed pass
From: @ginfung
Reviewed-by:
Signed-off-by:
|
5 years ago |
yujianfeng
|
7b412d7cb2
|
add recompute nodes
|
5 years ago |
yujianfeng
|
728fac6c9f
|
Expand J for innermost graph first when the graph also contains J primitive
|
5 years ago |
mindspore-ci-bot
|
32dc10a735
|
!9864 Adjust the location of cleaning unuse memory in value node
From: @joylvliang
Reviewed-by:
Signed-off-by:
|
5 years ago |
lvliang
|
1490947ff0
|
adjust-the-location-of-cleaning-unuseless-memory-in-value-node
|
5 years ago |
yujianfeng
|
e1aa34a030
|
Ignore the graphs with defer_inline flag when specializing on graph arguments
|
5 years ago |
buxue
|
2530943a7f
|
optimize list getitem in bprop
|
5 years ago |
zjun
|
67866474f9
|
Add pynative second derivative function
Signed-off-by: zjun <zhangjun0@huawei.com>
|
5 years ago |
mindspore-ci-bot
|
afa931ae79
|
!9308 Clean useless tensor after run pynative bprop graph
From: @HulkTang
Reviewed-by:
Signed-off-by:
|
5 years ago |
mindspore-ci-bot
|
0aa7197ef8
|
!9694 ZerosLike dynamic shape support
From: @peilin-wang
Reviewed-by: @liangchenghui
Signed-off-by: @liangchenghui
|
5 years ago |
mindspore-ci-bot
|
c2ad16352f
|
!9532 build for mac
From: @xiaoyisd
Reviewed-by:
Signed-off-by:
|
5 years ago |
Peilin Wang
|
d9fb28b9fc
|
zeroslike dynamic shape no opt
fix ci
fix ci
fix ci
fix ci
|
5 years ago |
tanghuikang
|
82450afa9e
|
Optimize memory using in pynative mode
|
5 years ago |
zhupuxu
|
4f569677b7
|
redundant_codes
Signed-off-by: zhupuxu <zhupuxu@huawei.com>
|
5 years ago |
xiaoyisd
|
424999267a
|
mac build
|
5 years ago |
mindspore-ci-bot
|
f6a22cb455
|
!9591 Assignment enables negative index
From: @liangzhibo
Reviewed-by: @ginfung,@zh_qh
Signed-off-by: @zh_qh
|
5 years ago |
l00591931
|
3189868a15
|
Assignment enable index smaller than 0
|
5 years ago |
mindspore-ci-bot
|
3b7acc57c9
|
!9508 [Auto parallel] Set CNODE's fullname when cloning
From: @xiaoda_zh
Reviewed-by: @kisnwang,@stsuteng
Signed-off-by: @stsuteng
|
5 years ago |
Xiaoda Zhang
|
9a9e3a751e
|
set cnode's fullname when cloning
|
5 years ago |
Zhang Qinghua
|
25a2b8cd5b
|
Never inline middle after block for control flow.
|
5 years ago |
Zhang Qinghua
|
3ea67d4549
|
Refactor and simplify the inlining procedure.
|
5 years ago |
Zhang Qinghua
|
829ab4492b
|
Optimize the performance after Grad due to mapping Primitive to K.
|
5 years ago |
mindspore-ci-bot
|
0e6d06c463
|
!9230 Set trace info of Primitive CNode only for equiv out node, not the whole fprop function.
From: @zh_qh
Reviewed-by: @ginfung
Signed-off-by: @ginfung
|
5 years ago |
Zhang Qinghua
|
8310236ff1
|
Set trace info of Primitive CNode only for equiv out node, not the whole fprop function.
|
5 years ago |
mindspore-ci-bot
|
ea191ed5da
|
!8463 Run ops one by one in pynative bp graph
From: @HulkTang
Reviewed-by:
Signed-off-by:
|
5 years ago |
HulkTang
|
c36b477568
|
Run ops one by one in pynative bp graph
|
5 years ago |
huanghui
|
bda74495f3
|
Add trace mangager around front opt
|
5 years ago |
Xiaoda Zhang
|
c79e988b0d
|
set fullname for reshape after reshape-elimination
|
5 years ago |
yao_yf
|
9cda064716
|
auto parallel predict
|
5 years ago |
lichenever
|
ee34ae9259
|
add_receive_eliminate_pass
|
5 years ago |
mindspore-ci-bot
|
56a731e48a
|
!8406 [AutoParallel]Add auto parallel pipeline
From: @lichen666
Reviewed-by:
Signed-off-by:
|
5 years ago |
lichenever
|
2e1c43483e
|
add auto parallel pipeline
|
5 years ago |
Zhang Qinghua
|
bcfa1f72b1
|
Add debug and trace info for grad operation.
|
5 years ago |
mindspore-ci-bot
|
244b7034e8
|
!7926 [ME][OptPass]fix bug when eliminate unused parameter in 'inline' pass
From: @chenfei52
Reviewed-by:
Signed-off-by:
|
5 years ago |
chenfei
|
e41c304b3e
|
dump ir of subpass of every substitution
add cache for transformed graph in inline
|
5 years ago |
mindspore-ci-bot
|
dbcdda18ed
|
!8293 Resolve specialize error during transforming after block in PyNative mode.
Merge pull request !8293 from 张清华/grad_opt2
|
5 years ago |
Zhang Qinghua
|
4e6e68f187
|
Resolve specialize error during transforming after block in PyNative mode.
|
5 years ago |
Yi Huaijie
|
d7faa77b5e
|
support int64 shape
|
5 years ago |
mindspore-ci-bot
|
e62c116277
|
!8163 Add irpass related to PrimTile.
Merge pull request !8163 from huangbingjian/add_irpass_tile
|
5 years ago |
HuangBingjian
|
0951cd6230
|
add IR pass {PrimTile, X, Empty} -> X
|
5 years ago |
chenfei
|
5d0b3597ff
|
incorporate switch pass should handle mutiple getitem
|
5 years ago |