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
i-robot
be10bb7af6
!23517 fix parallel opt recompute pass warning
Merge pull request !23517 from yao_yf/parallel_opt_recompute_pass_warning_fix
4 years ago
yao_yf
b647bbf444
fix warning
4 years ago
huangxinjing
c3a98bab2b
Add code check
fix pangu error
fix leakyrelu error
4 years ago
baihuawei
af0d315279
fix pclint and error log
4 years ago
baihuawei
a9694a9230
ascend add nontask sink mode
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
4e6f7dc97d
[feat] [assistant] [I3T96X] add new Dataset operator LibriSpeechDataset
4 years ago
yao_yf
a83bf73298
union auto_parallel_context interface dataset_strategy
4 years ago
yao_yf
dc7dc7d3fa
dataset strategy set
4 years ago
i-robot
c9d3c1d346
!20411 enable optimizer parallel for inference
Merge pull request !20411 from gziyan/enable_opt_shard_predict
4 years ago
Ziyan
1c9166e0a6
remove restriction for opt shard in inference
4 years ago
yao_yf
4aae231a8a
fix_pipeline_opt_shard
4 years ago
Ziyan
be1f5a43d7
opt shard fit micro batch
4 years ago
huangxinjing
e79db658e8
Fix codex for python file
4 years ago
Ziyan
2a752f24bf
enable not fully use opt shard
5 years ago
Ziyan
b3eebea4de
fix shard strategy for batchnorm
4 years ago
Ziyan
27c7c8618e
fix get seed validation
5 years ago
Ziyan
6e475bea3f
add warning for parameter share among multi devices
5 years ago
lizhenyu
f17534af08
ps cache support sparse
5 years ago
yangzhenzhang
cbca482e59
delete useless parameter in pipeline parallel
5 years ago
Ziyan
2c3b99ce91
fix infer rank list typo
5 years ago
Ziyan
660f578988
fix standalone prediction
5 years ago
Ziyan
e7a24611f4
fix distribtued predict
5 years ago
Ziyan
e7e9dae54d
support distributed predict
5 years ago
Yi Huaijie
b28a6ff88e
refactor seed interfaces
5 years ago
Yi Huaijie
c5f7700992
refactor get_seed() interface
5 years ago
Ziyan
cc131193ec
remove parameter broadcast
5 years ago
Yi Huaijie
e4cd67596f
raise RuntimeError when using full_batch neither under semi_auto_parallel nor auto_parallel
5 years ago
yao_yf
d4cfe55c04
rename mirror_mean to gradients_mean
5 years ago
Yi Huaijie
394be43492
raise RuntimeError when set different mode after Initializer created
5 years ago
Yi Huaijie
89a4ebf8a1
parallel mode must be set before create an initializer
5 years ago
yao_yf
cbb4363fa7
remove to_full_tensor and load_inputs in exexute stage
5 years ago
He Wei
f337c6bc14
Decouple ParamValue from python
5 years ago
Yi Huaijie
e5c351690b
support load full dataset on each device
5 years ago
yangzhenzhang
8c9730b3c5
add parallel mode for cell
5 years ago
lirongzhen
4ff418084c
enable/disable allreduce_fusion
5 years ago
zhunaipan
930a1fb0a8
initial version
Signed-off-by: leonwanghui <leon.wanghui@huawei.com>
6 years ago