i-robot
|
c04faad4ad
|
!30559 modify gpu opt adapter dynamic shape
Merge pull request !30559 from changzherui/mod_gpu_opt
|
4 years ago |
i-robot
|
059ac45ae8
|
!30920 Call Draw function directly
Merge pull request !30920 from huanghui/fix-draw
|
4 years ago |
greatpanc
|
07255c8939
|
dynamic thread refactor and optimization, add split op and softmax op dynamic thread choose
|
4 years ago |
yuchaojie
|
6a386a6b42
|
add Fracz depthwise case in TransDataSplit
|
4 years ago |
changzherui
|
47ef477fef
|
modify gpu opt adapter dynamic shape
|
4 years ago |
i-robot
|
df55f87899
|
!31029 fix attr length bug of Conv2DBackpropFilter
Merge pull request !31029 from zhangbuxue/fix_bug_of_conv2dbackpropfliter
|
4 years ago |
i-robot
|
3fecc04f26
|
!30886 profiler suport pynative
Merge pull request !30886 from zangqx/zangqx_8
|
4 years ago |
zhupuxu
|
d0d7c26b47
|
set min/max shape
Signed-off-by: zhupuxu <zhupuxu@huawei.com>
|
4 years ago |
buxue
|
c24e9c3e93
|
fix bug of Conv2DBackpropFilter
|
4 years ago |
i-robot
|
6423febaf0
|
!30981 Priority replay buffer backend kernel implementation
Merge pull request !30981 from chenweifeng/priority-replay-buffer-backend-kernel
|
4 years ago |
臧庆香
|
1dd489096e
|
profiler pynative python code
|
4 years ago |
i-robot
|
5fb502fdb4
|
!30978 Priority replay buffer bugfix
Merge pull request !30978 from chenweifeng/priority-replay-buffer-fix
|
4 years ago |
i-robot
|
7c1e6bbbbc
|
!30941 [MSLITE][Func] Add random normal operator.
Merge pull request !30941 from wangshaocong/bugfix_master
|
4 years ago |
i-robot
|
9075e36081
|
!30980 remove duplicate op impl
Merge pull request !30980 from zhujingxuan/master
|
4 years ago |
wilfChen
|
8a358f0817
|
fix priorities update issue
|
4 years ago |
i-robot
|
2ed51f3cc7
|
!30660 Add host_shape to CreateDeviceAddress API Input
Merge pull request !30660 from jiaorui/host_shape
|
4 years ago |
wilfChen
|
d19579b3dd
|
priority kernel backend implementation
|
4 years ago |
i-robot
|
a52e34ae7e
|
!30965 fix bug when init array on mac
Merge pull request !30965 from zhangbuxue/fix_bug_when_init_array_on_mac
|
4 years ago |
i-robot
|
40f282fe92
|
!30840 Fix dynamic shape aicpu launch failed.
Merge pull request !30840 from linqingke/compile-aicpu
|
4 years ago |
i-robot
|
c3b2a08fff
|
!30571 support sparse op on CPU via AKG
Merge pull request !30571 from 杨林枫/sparse_cpu
|
4 years ago |
zhujingxuan
|
45e7c257af
|
remove duplicate op impl
matrix_triangular_solve_gpu_kernel and trsm_solve_gpu_kernel are actually the same.
|
4 years ago |
buxue
|
c0b7aa77c0
|
delete init array because not support on mac
|
4 years ago |
i-robot
|
33495ae1e9
|
!30940 Use dynamic address for recv actor
Merge pull request !30940 from ZPaC/parse-label-for-parameter-update
|
4 years ago |
yanglf1121
|
893587ae55
|
support csr cpu
|
4 years ago |
wangshaocong
|
b2730f590d
|
[MSLITE] Add randomNormal operator.
|
4 years ago |
i-robot
|
55916a8394
|
!30711 move AkgCallBack to the first input of akg cpu kernel; add an extend_data value
Merge pull request !30711 from DeshiChen/0301_akgcallback
|
4 years ago |
i-robot
|
e8e35966a5
|
!30930 decouple ascend_stream_assign from frontend-parallel
Merge pull request !30930 from zhoufeng/backend-frontend-decoupling
|
4 years ago |
i-robot
|
438a4081fb
|
!28887 [assistant][ops] Add the MulNoNan operator
Merge pull request !28887 from 胡代旺/OpMulNoNan
|
4 years ago |
i-robot
|
ef5249b067
|
!30914 [MSLITE][CPU] op Dynamic thread choose Optimization, refactor and Optimization
Merge pull request !30914 from Greatpan/dynanic_thread_new
|
4 years ago |
ZPaC
|
4d3ded56a9
|
Use dynamic address for recv actor
|
4 years ago |
i-robot
|
64544b142a
|
!30340 develop maxpool3d, avgpool3d and conv3d on CPU(including farward and backward)
Merge pull request !30340 from zhangbuxue/develop_conv3d_maxpool3d_and_avgpool3d_on_cpu
|
4 years ago |
i-robot
|
54c39430fd
|
!30910 PriorityReplayBuffer completion
Merge pull request !30910 from chenweifeng/fifo-replay-buffer
|
4 years ago |
hu-daiwang
|
2cc4a44cbf
|
[assistant][ops] Add the MulNoNan operator
|
4 years ago |
zhoufeng
|
1029d576e6
|
decouple ascend_stream_assign from frontend-parallel
Signed-off-by: zhoufeng <zhoufeng54@huawei.com>
|
4 years ago |
buxue
|
50f3ad00ab
|
develop maxpool3d, avgpool3d and conv3d on CPU(including forward and backward)
|
4 years ago |
i-robot
|
513663ea0d
|
!30896 add passing fracz_group through optimizer's output
Merge pull request !30896 from yuchaojie/op_select
|
4 years ago |
i-robot
|
77cd10ce04
|
!30850 add transpose_x2 check in MatmulConfusionTranposeFusionPass
Merge pull request !30850 from yuchaojie/ub_fusion2
|
4 years ago |
zong-shuai
|
cdcbd9aeb0
|
debug
|
4 years ago |
wilfChen
|
0b279dce20
|
fifo replay buffer
|
4 years ago |
huanghui
|
ce0f500a3f
|
convert draw_ to call Draw function derectly
|
4 years ago |
i-robot
|
5e51dda2f7
|
!30837 priority replay buffer
Merge pull request !30837 from chenweifeng/priority-replay-buffer
|
4 years ago |
i-robot
|
69ada65e0a
|
!27273 [feat] [assistant] [I48O78][I48O4O]Add Rsqrt and RsqrtGrad Operators
Merge pull request !27273 from 桂宁馨/RsqrtGrad
|
4 years ago |
linqingke
|
42005bd5df
|
fix dynamic shape aicpu launch failed.
|
4 years ago |
greatpanc
|
05fbb7cf9c
|
dynanic thread refactor and optimization
|
4 years ago |
i-robot
|
960d551b8f
|
!30889 parse core type from pre-build for tbe kernel compile
Merge pull request !30889 from liubuyu/master
|
4 years ago |
i-robot
|
9b9760a481
|
!30831 [MS][LITE] fix lite train resize issue
Merge pull request !30831 from zhengjun10/fix
|
4 years ago |
yuchaojie
|
4490ef82de
|
add passing fracz_group through optimizer's output
|
4 years ago |
i-robot
|
048205e033
|
!30861 Fix send recv correction.
Merge pull request !30861 from ZPaC/parse-label-for-parameter-update
|
4 years ago |
lby
|
2b1b539b36
|
parse core type from pre-build for tbe kernel compile
|
4 years ago |
i-robot
|
ad808b3b30
|
!30872 [MS][LITE][Develop] fix memory leaks
Merge pull request !30872 from sunsuodong/fix_memory_leaks
|
4 years ago |