yanghaitao1
c94aa6b872
use profiler acl api instead
4 years ago
linqingke
bef2923acf
MindSpore support load custom aicpu ops.
4 years ago
Parastoo Ashtari
7f682ba2f6
partial support for multi root graph in online debugger
4 years ago
i-robot
8bf903ba19
!26320 Add Exception log when AscendMemoryPool::AllocTensorMem fail
Merge pull request !26320 from tanghuikang/oom_nullptr_log
4 years ago
ZPaC
611de83fd8
Fix dynamic load error
4 years ago
i-robot
81aff685b1
!26299 Modify Events Index
Merge pull request !26299 from hwjiaorui/event-index
4 years ago
tanghuikang
f0995e7899
Add Exception log when AscendMemoryPool::AllocTensorMem fail
4 years ago
i-robot
73bb099deb
!26056 easy using for kernel select log print
Merge pull request !26056 from liubuyu/master
4 years ago
hwjiaorui
613aa757c3
modify events index
4 years ago
lby
ab0efba450
easy using for kernel select log print
4 years ago
i-robot
9aafd972cd
!26219 fix the wrong iteration number issue with async dump + data sink mode
Merge pull request !26219 from yelihua/dev
4 years ago
i-robot
abcb458b28
!26167 fusion atomic clean
Merge pull request !26167 from liubuyu/bug_fix
4 years ago
yelihua
cb8aa0f31e
fix the dump structure error in data sink mode
4 years ago
hwjiaorui
1f8773e00d
add sparse unify mindir pass
4 years ago
i-robot
8072e6d7f7
!26062 add Custom, custom_op_info_register, CustomRegOp to __init__
Merge pull request !26062 from looop5/custom_init_commit
4 years ago
i-robot
8a15784782
!26117 [MSLITE] add nccl and mpi distribution in tensorrt delegate
Merge pull request !26117 from Liu_Xuu/trt_1109_nccl
4 years ago
lby
9b2e6e4aca
fusion atomic clean
4 years ago
looop5
58e27d87bc
add Custom, custom_op_info_register, CustomRegOp to __init__
4 years ago
hwjiaorui
6a39f5093b
fix stream bug
4 years ago
Liu_Xuu
255e2c03b4
[MSLITE] add nccl and mpi distribution in tensorrt delegate 1111_05
4 years ago
baihuawei
a112376b36
fix hete_mix_ctrlflow
4 years ago
i-robot
8b6beddca3
!21121 optimize thread pool
Merge pull request !21121 from kisnwang/optimize-thread-pool
4 years ago
i-robot
1bfedfcb7f
!26007 Ascend Device Context for MindRT
Merge pull request !26007 from hwjiaorui/ascend_context
4 years ago
i-robot
d4c8dde6d2
!25923 Move IsRealKernel and IsRealCNodeKernel from AnfAlgo to AnfUtils
Merge pull request !25923 from DeshiChen/1101_anfalgo
4 years ago
kswang
38d35ae96b
optimize common thread pool
4 years ago
hwjiaorui
d6b2a34a69
MindRT Ascend Devcie Context
4 years ago
i-robot
14c3517b9a
!25452 Add Gradient Sharding
Merge pull request !25452 from huangxinjing/pipeline_shard_mini_step
4 years ago
dayschan
cbb84ff580
Move IsRealKernel and IsRealCNodeKernel from AnfAlgo to AnfUtils
the function IsOneOfPrimitive and IsOneOfPrimitiveCNode is useful,
we can move them into anf.cc
4 years ago
i-robot
ba5a9529e2
!26010 fix cpu windows smoke
Merge pull request !26010 from baihuawei/fix_cpu_windows
4 years ago
i-robot
11bec4d85e
!25995 Add nvidia collective lib implementation.
Merge pull request !25995 from ZPaC/dir-of-distributed
4 years ago
huangxinjing
f354ab22a3
add pipeline shard interface
Add support for no pipeline accugradient
Add delay tag for fusion op
Optimizer the visite order
add mirror for mini step control
Move the group to attributes
Add gradient_shard control for the mini step
Fix code stype
Fix ut description
Add interface
4 years ago
ZPaC
9e18bad126
Add nvidia collective lib implementation.
4 years ago
i-robot
a229d056fa
!25939 Upgrade Ascend software package Nov 3
Merge pull request !25939 from TronZhang/upgrade_ascend_1104
4 years ago
baihuawei
372c3edfde
fix cpu windows smoke
4 years ago
i-robot
a58eb682d6
!25778 Custom op supports no reg info
Merge pull request !25778 from looop5/reg_info_commit
4 years ago
tronzhang
a8f9b81311
upgrade Ascend software package Nov 5
4 years ago
i-robot
67cb2abcd6
!25931 Sync parameter output after execute kernel when using MemScheduler
Merge pull request !25931 from tanghuikang/swap_output_input
4 years ago
i-robot
a0091160ed
!25732 set default HBM size for MindSpore to 30GB
Merge pull request !25732 from laiyongqiang/initial_mem
4 years ago
looop5
b89d744e80
Custom op supports no reg info
4 years ago
tanghuikang
87c7f72cf1
Sync parameter output after execute kernel when using MemScheduler
4 years ago
i-robot
02aa2838ed
!25803 unified runtime delete result arrow
Merge pull request !25803 from limingqi107/new_actor_runtime
4 years ago
i-robot
76f4f77cc2
!24520 Add graph run history to dump structure
Merge pull request !24520 from parastooashtari/multi_root_graph_dump
4 years ago
baihuawei
8607905b18
fix cpu control flow bug
4 years ago
limingqi107
1ba78fcbd6
unified runtime delete result arrow
4 years ago
liangzelang
1068183eb6
set default HBM size for MindSpore to 30GB
4 years ago
i-robot
0f1d3e5baf
!25739 PyNative RunOps support string input
Merge pull request !25739 from tanghuikang/runops_string
4 years ago
i-robot
5596e855db
!25807 fix cpu control flow bug
Merge pull request !25807 from baihuawei/fix_cpu_control
4 years ago
baihuawei
4db003303d
fix cpu ctrlflow bug
4 years ago
baihuawei
e59d07899b
fix reset8p pynative performance
4 years ago
parastooashtari
1a59dc37bf
add graph execution order history to dump
4 years ago