i-robot
3f9fed78c4
!21860 PyNative kerenl parallel build in FIRST step
Merge pull request !21860 from caifubi/master-kernel-parallel-build-simple
4 years ago
caifubi
537fce0ee1
PyNative Kernel Parallel Build
1. Create Tensor and DeviceAddress for output before Launch.
2. Push Launch/Build Task to Queue and execute togather.
4 years ago
Parastoo Ashtari
bf034bddb5
Apply comments on tensor stat online and offline debugger
4 years ago
i-robot
01ade5857d
!22123 update pynative profiling codes in fp and bp
Merge pull request !22123 from lvchangquan/profiling_formal
4 years ago
i-robot
32281f84e7
!19000 update LayerNormGrad split pass to V2
Merge pull request !19000 from yuchaojie/ir_fusion2
4 years ago
lvchangquan
bab311f0c7
update pynative profiling codes in fp and bp
4 years ago
i-robot
7d461ed6a4
!22349 fix Conv3D precision under fp16
Merge pull request !22349 from zuochuanyong/fix_conv3d_precision_fp16
4 years ago
i-robot
64f931f1cf
!22307 code check clean
Merge pull request !22307 from yuchaojie/code-clean
4 years ago
zuochuanyong
8fa68ebd98
fix Conv3D precision under fp16
4 years ago
i-robot
abc9d8e6fe
!22068 DynamicRNN supports scenarios that hidden_size is not multiples of 16
Merge pull request !22068 from yuchaojie/op_select2
4 years ago
i-robot
2e0ae45a67
!22322 clean code
Merge pull request !22322 from hwjiaorui/clean-code-master
4 years ago
i-robot
f88445dcf0
!22260 Avoid printing meaningless error message
Merge pull request !22260 from tanghuikang/tbe_em
4 years ago
yuchaojie
da4f6c51dc
code check clean
4 years ago
hwjiaorui
2008a2a78c
clean code
4 years ago
yuchaojie
1d1490df0b
xupdate LayerNormGrad split pass to V2
4 years ago
yuchaojie
7b3943bc91
DynamicRNN supports scenarios that hidden_size is not multiples of 16
4 years ago
tanghuikang
9d08583307
Avoid printing meaningless error message
4 years ago
i-robot
02db74ab2c
!22110 opt ascend single op mode runtime code
Merge pull request !22110 from baihuawei/graph_mode_nonsink_part3-1
4 years ago
i-robot
562088ad1b
!21894 Fix VM random failed issue
Merge pull request !21894 from chenweifeng/vm-random-failed
4 years ago
baihuawei
6eec288c39
opt ascend single op mode runtime code
4 years ago
tanghuikang
81e7f7173a
Report tbe error code to ErrorManager
4 years ago
ms_yan
36a8886ca2
Revert "[feat] [assistant] [I3T96T] add new Dataset operator CMUARCTICDataset"
This reverts commit b077aa1cab .
Revert "[feat] [assistant] [I3T96X] add new Dataset operator LibriSpeechDataset"
This reverts commit 4e6f7dc97d .
delete pass_registry_test.cc
comment hiai_nlu_model_multi.pb related line
4 years ago
djc
b077aa1cab
[feat] [assistant] [I3T96T] add new Dataset operator CMUARCTICDataset
4 years ago
djc
4e6f7dc97d
[feat] [assistant] [I3T96X] add new Dataset operator LibriSpeechDataset
4 years ago
zjun
35aab6144d
Fix pynative memory leak
Signed-off-by: zjun <zhangjun0@huawei.com>
4 years ago
i-robot
09a1a7f1f2
!21924 Add subdirectory for glog and ir_dump
Merge pull request !21924 from huanghui/add-submodule-for-dfx-files
4 years ago
wilfChen
8b8680c308
fix dead lock issue
4 years ago
i-robot
d5b61f23ab
!22051 fix serving kernel build failed
Merge pull request !22051 from laiyongqiang/serving
4 years ago
i-robot
db1995562e
!21917 Opt kernel runtime performance in vm non_task sink mode.
Merge pull request !21917 from liangzelang/opt_runtime
4 years ago
LaiYongqiang
2ab988fdcb
fix serving kernel build failed
4 years ago
huanghui
1630dcb0c8
add subdirectory for log and ir_dump
4 years ago
chenfei
4428db97e3
check abstract ref of update parameters
4 years ago
i-robot
1c7c7dc6da
!21913 add cpu mask_select type and fix visitkernel bug
Merge pull request !21913 from baihuawei/mask_select_and_visit_kernel_bug
4 years ago
i-robot
8553d342b2
!21957 get kernel build server dir path from python
Merge pull request !21957 from laiyongqiang/compile_server
4 years ago
i-robot
792d7afcd1
!21954 Fix 'kPrimTransData' from upper case 'K' to lower case 'k'
Merge pull request !21954 from hewei/core_user_api
4 years ago
LaiYongqiang
5e79bd6349
get kernel build server dir path from python
4 years ago
He Wei
18adbe3cd2
Fix 'kPrimTransData' from upper case 'K' to lower case 'k'
4 years ago
liangzelang
3e4ffd2025
opt runtime
4 years ago
yelihua
72e6058265
get rank id when set hccl env for single card train
4 years ago
baihuawei
178e918e91
add mask_select type
4 years ago
yelihua
a6dc9a0a07
get rank id when set hccl env for single card train
4 years ago
i-robot
fca1cb34c8
!21490 Ascend control use vm
Merge pull request !21490 from chenfei_mindspore/ascend-control-use-vm
4 years ago
chenfei
d6bc8c2fe3
ascend control in vm
inner test case fix
rm MS_CTX_MULTIGRAPH in ME func grpah cloner
add logs
4 years ago
i-robot
24526f2c0d
!21784 code opt and add manager to child graph
Merge pull request !21784 from hwjiaorui/opt-tensor
4 years ago
hwjiaorui
23b1fb7f3f
opt code and add childgraph manager
4 years ago
i-robot
36027a7485
!21770 Remove return value from datadump
Merge pull request !21770 from sabrinasun_59ee/dev
4 years ago
i-robot
445681830d
!21737 fix bug of common opt repeatly match a same node
Merge pull request !21737 from hwjiaorui/fix-node-pass-backend
4 years ago
sabrinasun
279717fb8e
remove return value of data dump
4 years ago
i-robot
d701d7a512
!21686 Fix dump logging when dump not enabled
Merge pull request !21686 from sabrinasun_59ee/dev
4 years ago
i-robot
6ca6d7e42d
!20973 keep tuple_getitem as much as possible to reduce the number of func graphs.
Merge pull request !20973 from xychow/dont-eliminate-tuple-getitem
4 years ago