i-robot
ded1c77bbf
!25765 neighborExchangeV2 & grad
Merge pull request !25765 from TuDouNi/neighborExchangeV2
4 years ago
ttudu
e953c15cd2
NeighborExchangeV2 & Grad
4 years ago
Margaret_wangrui
ae7ff1d17d
Add Primitive type into ValueNameToConverterVector
4 years ago
He Wei
fd8bbb4d96
Optimize isa() and type id related code.
1. Let 'isa<T>()' calls 'IsSameTypeId()' when T is final,
and we add 'final' mark to some final classes derived from Base;
2. Generate type id at compile time using constexpr hash function,
remove TypeIdManager and related code;
3. Add a tool script 'scripts/check_tid.sh' to check tid uniqueness;
4. Improve type search in 'proto_exporter.cc' and 'convert_utils_py.cc';
5. Add some missed MS_DECLARE_PARENT;
6. Ensure calls of 'parent_t::IsFromTypeId()' are all inlined.
4 years ago
Yang Jiao
9bcf28d15d
adapt custom op to pyfunc kernel
4 years ago
i-robot
324a767fe4
!25415 insert atomic ops process split
Merge pull request !25415 from liubuyu/master
4 years ago
LaiYongqiang
7f251e3f08
add attr kAttrSkipNopOpAddr for nop node hidden in execution order
4 years ago
lby
fe18bd8695
insert atomic clean ops
4 years ago
i-robot
ff924d6e4b
!25395 unified runtime support skip nop node
Merge pull request !25395 from limingqi107/new_actor_runtime
4 years ago
limingqi107
2fe803f81d
unified runtime support skip nop node
4 years ago
looop5
0b34bf33bc
add custom op pynative testcases
4 years ago
i-robot
aa63062595
!25082 Set akg kernel attrs in backend pass
Merge pull request !25082 from DeshiChen/1015_set_kernel_attr
4 years ago
dayschan
6600c0c474
Set akg kernel attrs in backend pass
it's unreasonable to change the node when generating kernel json.
instead, it should be set in a pass.
most of the operators in original akg_kernel_attrs_process are not longer used,
so we deleted them, leaving only the "Cast" and "MatMul/BatchMatMul".
4 years ago
i-robot
8cf8e4ddd3
!21778 support tensor slice when slice result should be caculated in runtime mode
Merge pull request !21778 from lianliguang/support-dynamic-slice
4 years ago
i-robot
ee70517a5b
!25070 fix single mode bug
Merge pull request !25070 from baihuawei/single_mode_bug
4 years ago
baihuawei
a16864069b
fix single mode bug
4 years ago
i-robot
4268cf5de0
!25026 support dynamic shape for cpu kernel of mae net
Merge pull request !25026 from fangzehua/dynamic_mae_1014
4 years ago
fangzehua
53009276d1
add cpu dynamic memory
4 years ago
lianliguang
72c7706d22
support dynamic slice in runtime mode
4 years ago
Yang Jiao
2d83f0e9ef
fix static-check
4 years ago
dayschan
7502345c8f
Add GraphKernelFlags into namespace mindspore::graphkernel
4 years ago
i-robot
06b0beced7
!22731 GraphKernel supports CPU
Merge pull request !22731 from DeshiChen/0901_graphkernel_cpu
4 years ago
dayschan
32ecd8ee79
GraphKernel supports CPU
only Linux system is supported now.
change the default value of `ENABLE_AKG` to off, and controlled by option `-K`.
the `ENABLE_AKG` is auto enabled when `ENABLE_GPU` or `ENABLE_D` is on.
since now, we can use `ENABLE_AKG` to control the compilation of graphkernel
and akg codes.
fix usage description for option "-K", it should be "[-K on|off]".
LLVM is required by akg for cpu kernels, so AKG for cpu is default disabled now.
4 years ago
limingqi107
0a9cbd2220
imporve performance of CPU LSTMGrad
4 years ago
i-robot
f98cc8dce2
!23599 Handle real to complex in bprob
Merge pull request !23599 from zhouyaqiang0/complex_ops
4 years ago
zhouyaqiang
3d01cb0c3d
handle real to complex bprob
4 years ago
i-robot
d9e6edfc9f
!23669 Fix inline pass problem in switch.
Merge pull request !23669 from LiangZhibo/inline
4 years ago
l00591931
092fcc40c9
Fix inline pass problem on switch
4 years ago
i-robot
9309a11999
!23352 [Fallback] Handle interpreted node as input, set function Parameters and update local params if need.
Merge pull request !23352 from 张清华/opt_fallback
4 years ago
i-robot
b825fd281d
!23697 fix warnings of pclint plus, codedex and codecheck at 0917
Merge pull request !23697 from huanghui/clean-warnings-0917
4 years ago
Zhang Qinghua
9fbd118319
Handle interpreted node as input, set function Parameters and update local params if need.
4 years ago
i-robot
e87f0be14f
!23707 fix summary isolation
Merge pull request !23707 from jiangshuqiang/temp
4 years ago
i-robot
917b83e9b1
!23715 codedex
Merge pull request !23715 from TuDouNi/master
4 years ago
i-robot
825897f15f
!23522 fix log to file bug of Could not create logging file: No such file or directory/File exists
Merge pull request !23522 from huanghui/fix-log-to-file
4 years ago
huanghui
80416a71e0
fix warnings of pclint plus, codedex and codecheck
4 years ago
jiangshuqiang
c2bd061889
fix summary isolation
4 years ago
ttudu
c8f24a18f7
codedex
4 years ago
i-robot
dc87bf7f4d
!23051 add TransposedUpdateFusion
Merge pull request !23051 from yuchaojie/ir_fusion3
4 years ago
i-robot
fb58cfa273
!23414 update pynative profiling codes and delete some useless code
Merge pull request !23414 from lvchangquan/master
4 years ago
huanghui
4a873ff4ed
Fix the bug of "Could not create logging file: No such file or directory/File exists"
4 years ago
yuchaojie
6cd16d61c4
add TransposedUpdateFusion
4 years ago
i-robot
409e0af313
!23455 support value_depend input
Merge pull request !23455 from laiyongqiang/value_depend
4 years ago
i-robot
6f09891501
!22092 [executor] Add mem scheduler
Merge pull request !22092 from kisnwang/add-mem-extend-cache
4 years ago
lvchangquan
b908f40d94
delete some useless codes and update pynative profiling codes
4 years ago
LaiYongqiang
9082b70708
support value_depend input
4 years ago
Gaoxiong
1b3707b919
add dayschan to graphkernel committers
4 years ago
i-robot
7e3dca5d5b
!23426 fix empty group check
Merge pull request !23426 from zhoufeng/xiu-ba-ge-2
4 years ago
i-robot
61b3e33898
!23376 Support relative path in create file
Merge pull request !23376 from huanghui/support-relative-path
4 years ago
zhoufeng
4719338bd2
fix empty group check
5 years ago
kswang
a0dc465041
add mem scheduler
4 years ago