zhouyuanshen
|
048fc49aed
|
add support to op RandomCategorical
|
4 years ago |
mindspore-ci-bot
|
0de9d3e5b7
|
!6989 split sort for parallel run
From: @kisnwang
Reviewed-by:
Signed-off-by:
|
4 years ago |
|
mindspore-ci-bot
|
98e2a65aab
|
!8444 fix format insert
From: @yankai10
Reviewed-by: @hangangqiang,@zhanghaibo5
Signed-off-by: @hangangqiang
|
4 years ago |
|
mindspore-ci-bot
|
fed8702a29
|
!8457 update test_resnet50_quant loss threshold to 2.6
From: @yuchaojie
Reviewed-by: @c_34,@xsmq
Signed-off-by: @c_34
|
4 years ago |
zhanyuan
|
ae0e3dab65
|
[MSLITE] Fix the fp32 FC's resize bug
|
4 years ago |
huanghui
|
27b9b2870e
|
fix ReduceEltwiseFusionPass not set output_used_num
|
4 years ago |
|
mindspore-ci-bot
|
156980778b
|
!8416 Fix RealDiv Type Error when insert an VirtualDiv Operation
From: @huangxinjing
Reviewed-by: @stsuteng
Signed-off-by: @stsuteng
|
4 years ago |
|
mindspore-ci-bot
|
8e3bf54fd2
|
!7032 add CTCGreedyDecoder ops for aicpu
From: @yanzhenxiang2020
Reviewed-by:
Signed-off-by:
|
4 years ago |
|
mindspore-ci-bot
|
0dedcceefa
|
!8374 fix tile op
From: @wuxuejian
Reviewed-by: @liangchenghui,@c_34
Signed-off-by: @liangchenghui
|
4 years ago |
|
mindspore-ci-bot
|
93a48a8425
|
!8445 [MSLITE][Develop] optimize arm cpu conv depthwise fp32: add 3x3
From: @yangruoqi713
Reviewed-by: @zhang_xue_tong,@hangangqiang
Signed-off-by: @zhang_xue_tong
|
4 years ago |
|
mindspore-ci-bot
|
26dd9d1905
|
!8454 [MSLITE]fix train mode in package_lite.cmake
From: @guohonhzilonghw
Reviewed-by: @HilbertDavid,@ddwsky
Signed-off-by: @HilbertDavid
|
4 years ago |
ling
|
3fd7bc78c2
|
[MSLITE] mnet full quant bug
|
4 years ago |
hwjiaorui
|
c1a57c83b5
|
move dynamic shape mark before backend common optimization
format
|
4 years ago |
gengdongjie
|
b7cbd94eeb
|
add hccl timeout annotation for maskrcnn
|
4 years ago |
yangruoqi713
|
6d3e3be524
|
[MSLITE][Develop] reopen arm cpu int8 deconv depthwise assembly
|
4 years ago |
zhaozhenlong
|
2afed72140
|
strided_slice_with_axes
|
4 years ago |
|
mindspore-ci-bot
|
59abca824c
|
!8446 [MSLITE] tanh int8 op
From: @ling_qiao_min
Reviewed-by: @zhang_xue_tong,@HilbertDavid
Signed-off-by: @zhang_xue_tong
|
4 years ago |
|
mindspore-ci-bot
|
696e6d959e
|
!8452 [MS][LITE][CPU]optimize int8 spaceToBatch
From: @fuzhiye
Reviewed-by: @zhang_xue_tong,@hangangqiang
Signed-off-by: @zhang_xue_tong
|
4 years ago |
yuchaojie
|
733a37be13
|
update test_resnet50_quant loss threshold to 2.6
|
4 years ago |
|
mindspore-ci-bot
|
466fdcbd42
|
!8439 update run 20201110
From: @jonyguo
Reviewed-by: @ljl0711,@pandoublefeng
Signed-off-by: @pandoublefeng
|
4 years ago |
guohongzilong
|
585f29a1f7
|
fix train mode in package_lite.cmake
|
4 years ago |
|
mindspore-ci-bot
|
222a0bccf4
|
!8431 Add more check in unsorted_segment_sum_fission pass to avoid doing wrong optimize which cause runtime error
From: @irmo
Reviewed-by:
Signed-off-by:
|
4 years ago |
yankai
|
f1115dd077
|
fix format insert format
|
4 years ago |
wilfChen
|
07d8622c7e
|
modify dynamic shape condition to fit ME
|
4 years ago |
huangxinjing
|
3e9fac7f59
|
Fix repeat error
|
4 years ago |
cjh9368
|
dfc7b3e27e
|
rewrite tflite model parsers
|
4 years ago |
wuxuejian
|
f3a0ae0b1b
|
fix tile op
|
4 years ago |
|
yangruoqi713
|
3bb937e718
|
[MSLITE][Develop] optimize arm cpu conv depthwise fp32: add 3x3
|
4 years ago |
anancds
|
96d8c411e7
|
added message handler unit test
|
4 years ago |
fuzhiye
|
638b7c92c0
|
optimize int8 spaceToBatch
|
4 years ago |
|
ling
|
7694dda4f4
|
[MSLITE] tanh int8 op
|
4 years ago |
|
mindspore-ci-bot
|
8aa78c2c8e
|
!8443 [MS][LITE][GPU]lite gpu add spacetodepth
From: @chenzupeng
Reviewed-by: @ddwsky,@zhang_xue_tong
Signed-off-by: @ddwsky
|
4 years ago |
|
mindspore-ci-bot
|
d760cee90f
|
!8442 clean exector before clear runtime resource
From: @kisnwang
Reviewed-by: @jjfeing,@chujinjin
Signed-off-by: @chujinjin
|
4 years ago |
|
mindspore-ci-bot
|
623df51e06
|
!8420 mode_train_file
From: @bai-yangfan
Reviewed-by: @chenfei52,@zh_qh
Signed-off-by: @zh_qh
|
4 years ago |
|
huanghui
|
1c6c280da7
|
fix unsorted_segment_sum_fission pass
|
4 years ago |
dayschan
|
195b1fe8d5
|
Add Transpose into fusible list.
|
4 years ago |
bai-yangfan
|
a9cba56191
|
mode_train
|
4 years ago |
chenzupeng
|
edc4ac2c25
|
add spacetodepth
|
4 years ago |
kswang
|
3440b66a9e
|
clean exector before clear runtime resource
|
4 years ago |
|
mindspore-ci-bot
|
bedc733e42
|
!8440 fix resize issues of reshape and conv2d lite ops
From: @liuwenhao4
Reviewed-by: @zhang_xue_tong,@zhanghaibo5
Signed-off-by: @zhang_xue_tong
|
4 years ago |
|
mindspore-ci-bot
|
bf1c08ba47
|
!8274 himindspore add net that can download new version apk and install
From: @sishuikang
Reviewed-by: @zhang_xue_tong
Signed-off-by:
|
4 years ago |
|
huangxinjing
|
12e9107162
|
Fix VirtualDiv Int32 error
|
4 years ago |
l00591931
|
c5b5a6719c
|
Enable multi-dimensional list value assignment
|
4 years ago |
liuwenhao4
|
b1a868744c
|
fix resize issues of reshape and conv2d lite ops
|
4 years ago |
|
mindspore-ci-bot
|
eb5ae1a0fc
|
!8423 nn.Dense support ND*2D
From: @wanyiming
Reviewed-by:
Signed-off-by:
|
4 years ago |
jonyguo
|
f275c5fea2
|
update run 20201110
|
4 years ago |
Peilin Wang
|
c5ad19deba
|
dynamic shape op initial changes
fix ci
fix ci
|
4 years ago |
TFbunny
|
24ea2ddf92
|
Adapt changes to GPU framework for ScatterAdd/Update and SparseFtrl
|
4 years ago |
|
mindspore-ci-bot
|
70f5775711
|
!8387 Add support for GetClassIndexing in C++ API
From: @alex-yuyue
Reviewed-by:
Signed-off-by:
|
4 years ago |
tom__chen
|
9263145990
|
gpu unique op kernel and st test
|
4 years ago |
