e341b7ee97
!10335 bleu score
...
From: @lijiaqi0612
Reviewed-by:
Signed-off-by:
4 years ago
ca84fa1a00
add avx int8 add
4 years ago
ade9a82c2b
fix device memory leak
4 years ago
52953f16fc
!11027 Use logic id instead of physic id in get_distribute_pretrain_cmd.py
...
From: @c_34
Reviewed-by: @ljl0711,@liangchenghui
Signed-off-by: @liangchenghui
4 years ago
8a6d5baeca
find resize nn input tensor 2 num case
4 years ago
8ce17ba568
!11015 fix resize and shape
...
From: @yeyunpeng2020
Reviewed-by:
Signed-off-by:
4 years ago
175a17216a
!11044 [MSLITE] nnacl optimize : delete clip
...
From: @ling_qiao_min
Reviewed-by:
Signed-off-by:
4 years ago
24dc0925db
fix bug in getitem when index is True
4 years ago
5194913bc9
!11077 only split dynamic shape node for ascend
...
From: @kisnwang
Reviewed-by: @zhoufeng54,@chujinjin
Signed-off-by: @chujinjin
4 years ago
565084462b
!11061 fix longrunning cause segmentfault
...
From: @kisnwang
Reviewed-by:
Signed-off-by:
4 years ago
1e7f18a097
Modify dot opt to support pynative mode
4 years ago
92a85d1061
!11075 dynamic op re primitive when infer
...
From: @liubuyu
Reviewed-by: @kisnwang,@zhoufeng54
Signed-off-by: @zhoufeng54
4 years ago
026cdcd6fc
fix build for md pipeline
4 years ago
756f67151b
fix fasttext doc problems
4 years ago
f466776bde
!11052 fix bug of sparse unify mindir
...
From: @hwjiaorui
Reviewed-by:
Signed-off-by:
4 years ago
93b3055a47
fix longrunning segment fault
4 years ago
984f31e901
fix bug of supported data type for LessEqual on CPU
4 years ago
51ac2847e3
fix resize & shape bug
4 years ago
60ac42adca
fix api comments.
4 years ago
aad0f0561f
!11043 encoder no fusion run success
...
From: @cjh9368
Reviewed-by: @zhanghaibo5
Signed-off-by:
4 years ago
bc5ca14132
[MSLITE] nnacl optimize
...
delete clip
unstack to base
unsqueeze to base
batchToSpace to base
4 years ago
69de2c376d
!11091 Correct the description error of the operator Depend.
...
From: @david-he91
Reviewed-by: @liangchenghui,@wuxuejian
Signed-off-by: @liangchenghui
4 years ago
30560be800
!11066 GPU add restrict for bert script
...
From: @VectorSL
Reviewed-by: @gaoxiong1,@dylangeng,@anyrenwei
Signed-off-by: @gaoxiong1
4 years ago
b9b4a5e5f7
Add a restriction for getitem in basic_ops_fusion.
...
this commit reverts the modification for basic_ops_fusion.cc in 8af78cd5c
4 years ago
64b72bd511
!11051 Fixed GNMT reasoning performance
...
From: @gaojing22
Reviewed-by: @wuxuejian,@linqingke
Signed-off-by: @wuxuejian
4 years ago
d6bf058052
!11076 uadate L1Regularizer operation annotation
...
From: @dangjiaqi1
Reviewed-by: @wuxuejian,@liangchenghui
Signed-off-by: @liangchenghui
4 years ago
b3aa620ab9
optimize somas logs:
...
1. add parameter info
2. update atomic clean inputs
3. remove somas_mem_pool.ir
4 years ago
e2b740e812
add physic_logic_ids in get_distribute_pretrain_cmd.py
...
change ways to get physic_logic_ids
4 years ago
19a21858d8
!11006 add single train device_id tiny_darknet
...
From: @wukesong
Reviewed-by: @yingjy
Signed-off-by:
4 years ago
e6e97802ab
resolve encoder model parser problems
4 years ago
d46279dba2
!11078 dataset: minor fixes which resolve codedex CI alarms
...
From: @cathwong
Reviewed-by: @nsyca,@robingrosman
Signed-off-by: @nsyca
4 years ago
39cc9e70cd
dynamic op re primitive when infer
4 years ago
29505ceb4a
!10990 fix: probabilistic failure when get data timeout in retinaface_resnet50 / psenet
...
From: @jonyguo
Reviewed-by:
Signed-off-by:
4 years ago
7ce9056fc8
Correct the description error of the operator Depend
4 years ago
2c90cf6a1f
!11058 [MD][bugfix] Fix GeneratorDataset get_dataset_size issue
...
From: @xiefangqi
Reviewed-by: @liucunwei,@heleiwang
Signed-off-by: @liucunwei
4 years ago
e6a25379ed
dataset: minor fixes which resolve codedex CI alarms
4 years ago
a0c7b7abef
!11035 Change GPU Slice ops shape from int to size_t/int64_t
...
From: @TFbunny
Reviewed-by: @robingrosman,@tom__chen
Signed-off-by: @tom__chen
4 years ago
0e68575e77
port over getter pass
...
port over tensor_op_fusion pass
add fusion support for prebuildOpeartion (TensorOpeartion)
4 years ago
7aac7f694c
change size_t to S type in argmaxwithvalue
4 years ago
6d6ed86a9d
!10427 RangeOp new gpu kernel
...
From: @peilin-wang
Reviewed-by:
Signed-off-by:
4 years ago
908e9a526b
tile's input multiples
...
new mul node
new mul node
convert const to attr
4 years ago
71417617a5
update L1Regularizer annotation
4 years ago
b6ee7f3fb2
only split dynamic shape node for ascend
4 years ago
f6de97fc27
bleu_score
4 years ago
62888f23b8
!10400 roc and auc
...
From: @lijiaqi0612
Reviewed-by:
Signed-off-by:
4 years ago
5169ff24ab
!11041 [GraphKernel] Change `None` input desc to null list in model builder
...
From: @tronzhang
Reviewed-by: @gaoxiong1,@ckey_dou
Signed-off-by: @ckey_dou
4 years ago
7cac3d3a47
Fix continuous calls for recompute api
4 years ago
e2f344f74a
!10086 fix shape of CTCGreedyDecoder
...
From: @yanzhenxiang2020
Reviewed-by:
Signed-off-by:
4 years ago
ba76356702
fix: retinaface_resnet50 & psenet random get data timeout
4 years ago
f544aa5c3b
!11068 Fix some api comments.
...
From: @zhang_yi2020
Reviewed-by: @gemini524,@liangchenghui,@wuxuejian
Signed-off-by: @liangchenghui
4 years ago
mindspore-ci-bot
fuzhiye
chujinjin
zhaozhenlong
buxue
xutianming
xulei2020
zhaojichen
kswang
shibeiji
yeyunpeng
zhangyi
ling
dayschan
laiyongqiang
chenhaozhe
cjh9368
liubuyu
hedongdong
Cathy Wong
Zirui Wu
TFbunny
hwjiaorui
“dangjiaqi1”
Jiaqi
yujianfeng
jonyguo