liuluobin
aa468ce193
Add nested usage of shard verification
2022-10-26 08:00:46 +08:00
yao_yf
bc5a7d8d1e
parallel func mode in sink mode
2022-10-20 16:54:00 +08:00
i-robot
0245490acb
!43657 remove usage of ScalarToArray
...
Merge pull request !43657 from zhujingxuan/replace
2022-10-19 01:23:45 +00:00
i-robot
7fab747289
!42506 mv ge pass from frontend to backend
...
Merge pull request !42506 from zhangyinxia/master
2022-10-17 07:57:50 +00:00
lichen
41af052460
remove_parallel_related_in_model
2022-10-17 10:01:17 +08:00
zhangyinxia
e3814181ed
mv ge pass from frontend to backend
2022-10-13 22:01:51 +08:00
i-robot
63416f83c7
!42768 fix moe issue
...
Merge pull request !42768 from bichaoyang/master
2022-10-13 08:47:55 +00:00
zhujingxuan
1fd6d59f90
remove usage of ScalarToArray
2022-10-13 14:34:06 +08:00
i-robot
bf1f4ddce2
!42226 Add the check of the return value of Assign, AssignAdd, AssignSub.
...
Merge pull request !42226 from Margaret_wangrui/check_assign_assignsub_assignadd
2022-10-13 02:33:02 +00:00
i-robot
e4c41701b0
!43570 Avoid subgraphs returning parameter Ref type data by inserting Load, preventing the performance of the bert network from deteriorating.
...
Merge pull request !43570 from Margaret_wangrui/master
2022-10-13 01:11:50 +00:00
bichaoyang
92b30828b7
fix issue of insert mirror;optimaze sharding propagation
2022-10-12 16:11:26 +08:00
Margaret_wangrui
de0f7f23b9
Avoid subgraphs returning parameter Ref type data by inserting Load, preventing run graph mode with kernelbykernel.
2022-10-11 18:56:16 +08:00
Margaret_wangrui
dd2c45a829
Add the check of the return value of Assign, AssignAdd, AssignSub.
2022-10-11 15:50:47 +08:00
lvyufeng
4724964e8a
Revert "add value print for Parameter"
...
This reverts commit eff301e3e5 .
2022-10-11 09:54:02 +08:00
xiongkun
b710580e64
add padlist check
...
tick
tick
tick
tick
remote &
fix con
fix con
modify test
remove conv ut
2022-10-08 17:22:48 +08:00
i-robot
dfcd89f1bf
!42593 add value print for Parameter
...
Merge pull request !42593 from 吕昱峰(Nate.River)/master
2022-10-04 13:25:55 +00:00
yangzhenzhang
bfc02c9371
add conv3d parallel op
2022-09-29 17:33:40 +08:00
liuluobin
a9b047bc61
Added float64 support for IOU on CPU&GPU, and modify description of pynative shard api
2022-09-26 14:52:20 +08:00
lvyufeng
eff301e3e5
add value print for Parameter
2022-09-26 12:54:13 +08:00
i-robot
5dff519437
!40643 Improve the Transformer APIs
...
Merge pull request !40643 from huangxinjing/improve_transformer_apis
2022-09-24 03:10:37 +00:00
huangxinjing
908975d458
Impove the transforer apis
...
Support attention mask is None
Fix attention check
WIP in custom activation
Support attention
2022-09-23 21:57:58 +08:00
kswang
dc236c4753
add set device interface for prim
2022-09-23 10:58:01 +08:00
i-robot
002054321a
!41862 fix Image, log_matrix_determinant, lp_norm kernel bugs
...
Merge pull request !41862 from zhuzhongrui/pub_master
2022-09-16 02:34:10 +00:00
i-robot
e62cd4e6ce
!41983 Fix bprop bugs of Ops Range
...
Merge pull request !41983 from pkuliuliu/master
2022-09-15 09:08:09 +00:00
i-robot
feeaa435f5
!41811 auto_parallel_ckpt_transform
...
Merge pull request !41811 from yao_yf/auto_parallel_ckpt_transform_add_outter_interface_and_testcases
2022-09-15 01:11:48 +00:00
z00512249
e86ccc636a
fix Image, log_matrix_determinant, lp_norm kernel bugs
2022-09-14 19:42:11 +08:00
pkuliuliu
98f709e6e6
Fix bprop bugs of Ops Range
2022-09-14 18:17:59 +08:00
yangzhenzhang
b3c251208a
modify gather manual split
2022-09-14 10:18:03 +08:00
yao_yf
0937a043b9
add checkpoints transform api
2022-09-14 09:25:39 +08:00
yangzhenzhang
dba41c00da
fix bug for striedslice
2022-09-06 14:55:05 +08:00
lizeting-pcl
59d790e7cc
add matmul op
2022-08-31 10:18:49 +08:00
yangzhenzhang
9489bc9ce6
modify stridedslice parallel op
2022-08-27 15:10:52 +08:00
hezhenhao1
bc12b4ef87
Add function/tensor api for PReLU, Square, SoftplusGrad operators.
2022-08-25 14:28:31 +08:00
i-robot
d222d7393c
!40455 [AutoParallel]optimizer Pipeline parallel
...
Merge pull request !40455 from lichen/opt_pipeline_parallel
2022-08-19 01:15:31 +00:00
lichen
23a3d65995
optimizer_pipeline_parallel
2022-08-17 17:40:26 +08:00
yao_yf
854b7839e5
auto parallel dataset handle in feed mode
2022-08-15 14:41:16 +08:00
yangzhenzhang
b83c99ce71
add flag for broadcast net
2022-08-04 17:07:49 +08:00
yangzhenzhang
5fa63f8e46
delete auto parallel flag
2022-07-30 15:19:56 +08:00
Zhang Qinghua
f90dcc963d
Support isolated side-effect nodes in constant returned func graph.
2022-07-27 10:12:24 +08:00
wangshengnan12@huawei.com
937e65e81a
clean_code_auto_parallel
2022-07-21 15:06:05 +08:00
zhangyinxia
3e1320b246
add lamb and clipbynorm pass for ge
2022-07-20 10:54:06 +08:00
yangzhenzhang
25009f9051
modify gather op
2022-07-07 14:59:50 +08:00
i-robot
58c901923a
!36767 modify the generated strategy for conv2d
...
Merge pull request !36767 from yangzhenzhang/modify-conv2d-gen-strategy-func
2022-06-29 01:39:29 +00:00
wzw
d309dfd41e
add transformer testcase
2022-06-28 20:01:29 +08:00
yangzhenzhang
9b50d2cfee
modify conv2d gen strategy func
2022-06-28 16:32:15 +08:00
wangshengnan12@huawei.com
bdf2004ca4
fix_pipeline_compile_error
2022-06-28 09:55:31 +08:00
huangxinjing
055ab431db
[AutoParallel]Add virtual mul
...
Fix OneHot Error
Add onehot test cast
2022-06-25 09:06:40 +08:00
huangxinjing
d0f2bf37c6
Add cross entory loss
...
[AutoParallel]Add sub graph support for parallel training
Fix forward mark error and virtual dataset error
Add Test cases
Add virtual dataset inserts, forbiden the subgraphs as candidates
2022-06-20 10:53:21 +08:00
i-robot
e902c7c731
!35839 fix bug of moe
...
Merge pull request !35839 from bichaoyang/master
2022-06-17 08:15:25 +00:00
yangzhenzhang
b63e70f6a9
gen data parallel strategy for conv2d auto parallel
2022-06-16 16:44:30 +08:00