mindspore-ci-bot
|
64abbeaa89
|
!705 add pynative cache
Merge pull request !705 from chujinjin/add_pynative_cache
|
6 years ago |
chujinjin
|
e2b0a28142
|
add pynative cache
|
6 years ago |
chenjianping
|
6d47036f95
|
erase datatype raise kernel
|
6 years ago |
mindspore-ci-bot
|
2a31fce599
|
!647 change profiling iter_end point after last tbe kernel
Merge pull request !647 from caifubi/profiling_iter_end_point
|
6 years ago |
caifubi
|
3f69b5b518
|
insert profiling iteration end point after last tbe kernel
|
6 years ago |
mindspore-ci-bot
|
ffdad1acd4
|
!617 Match format when kernel selecting using raise or reduce precision
Merge pull request !617 from lianliguang/fix-bug-of-raise-precision
|
6 years ago |
mindspore-ci-bot
|
bfab0bc856
|
!618 clean ascend runtime codex
Merge pull request !618 from caifubi/clean-runtime-codex
|
6 years ago |
caifubi
|
246fc290d0
|
clean runtime codex
|
6 years ago |
lianliguang
|
c4aeb5a0b8
|
add format chooice when kernel selecting reduce or raise precision
|
6 years ago |
liubuyu
|
0b6b5e5123
|
fix codedex warning
|
6 years ago |
caifubi
|
0cef4d8abd
|
set fp_start profiling point after getnext node
|
6 years ago |
mindspore-ci-bot
|
715c0735a8
|
!487 add dtype trans template
Merge pull request !487 from liubuyu/master
|
6 years ago |
liubuyu
|
ac2d5df2a1
|
add dtype trans template
|
6 years ago |
caifubi
|
94a455dace
|
insert profiling kernel for hccl automaticly
|
6 years ago |
zhoufeng
|
c2b3360d69
|
update clang format rule
|
6 years ago |
gukecai
|
f8208c7c52
|
Support GetNext Parallel
|
6 years ago |
mindspore-ci-bot
|
093c2caed4
|
!337 optimize execute order sort
Merge pull request !337 from kisnwang/optimize-execute-order-sort
|
6 years ago |
mindspore-ci-bot
|
7c06d292c8
|
!387 auto mix precision
Merge pull request !387 from liubuyu/master
|
6 years ago |
liubuyu
|
852e61d46c
|
bug fix
|
6 years ago |
liubuyu
|
b1585f862d
|
auto mix precision
|
6 years ago |
liubuyu
|
fc07cd908e
|
add 6d format transfer
|
6 years ago |
kswang
|
83eeac9310
|
optimize execute order sort
|
6 years ago |
jjfeing
|
f9ef78609f
|
add nc1hwc0_c04 format
|
6 years ago |
caifubi
|
bce5f57752
|
use GraphId as key of DavinciModel in ascend_kernel_runtime.cc
|
6 years ago |
mindspore-ci-bot
|
f9dd47620c
|
!300 refactor kernel select priority scheme
Merge pull request !300 from lianliguang/refactor-kernel_select
|
6 years ago |
mindspore-ci-bot
|
285e258b3b
|
!347 fix e2e dump shape not match
Merge pull request !347 from dengwentao/dump
|
6 years ago |
jojobugfree
|
762bf9ac25
|
fix tensoradd grad op run fail
|
6 years ago |
dengwentao
|
593c4fc700
|
fix shape used for dump
|
6 years ago |
lianliguang
|
5365678eee
|
refactor kernel select
|
6 years ago |
lianliguang
|
5d225f934f
|
change the padding strategy & refactor insert transdata
|
6 years ago |
mindspore-ci-bot
|
18b9a0957e
|
!203 fix reshape as output and release mem exception
Merge pull request !203 from kisnwang/optimize-allreduce-mem-malloc
|
6 years ago |
kswang
|
b8a7e73f7d
|
fix reshape output and clearres error
|
6 years ago |
lianliguang
|
3348e5a7c2
|
deal something special of adam's kernel select
|
6 years ago |
kswang
|
bef62db128
|
add ascend mem pool
|
6 years ago |
kswang
|
fb343bd607
|
add mem manager
|
6 years ago |
jojobugfree
|
effdb483d6
|
profiling feature enhancement
|
6 years ago |
wanghua
|
5b176f258b
|
modify bert test file
|
6 years ago |
wanghua
|
da123c5b3e
|
fix bert precison bug
|
6 years ago |
mindspore-ci-bot
|
55916351ee
|
!52 remove ge depend
Merge pull request !52 from flywind/remove-ge-depend
|
6 years ago |
Wei Luning
|
73ba399364
|
remove ge depend in cpu
|
6 years ago |
laiyongqiang
|
bb421b70db
|
fix bug when dump op with trans_flag true
|
6 years ago |
jojobugfree
|
e6c15b82c8
|
remove cce dependency
|
6 years ago |
lvliang
|
b3a306489d
|
auto enbale dynamic mem pool
|
6 years ago |
zhunaipan
|
930a1fb0a8
|
initial version
Signed-off-by: leonwanghui <leon.wanghui@huawei.com>
|
6 years ago |