p60975123
/
mindspore
forked from mindspore-Ecosystem/mindspore
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

!11483 [Auto parallel] Add an optimization in searching strategies in model_zoo's resnet 

From: @xiaoda_zh
Reviewed-by: @stsuteng,@kisnwang
Signed-off-by: @stsuteng
This commit is contained in:
mindspore-ci-bot 2021-01-28 17:20:13 +08:00 committed by Gitee
parent e3796b3639 2646654a30
commit 58fecfb673
1 changed files with 2 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
model_zoo/official/cv/resnet/train.py
View File

@ -27,6 +27,7 @@ from mindspore.train.loss_scale_manager import FixedLossScaleManager
from mindspore.train.serialization import load_checkpoint, load_param_into_net
from mindspore.communication.management import init, get_rank, get_group_size
from mindspore.common import set_seed
from mindspore.parallel import set_algo_parameters
import mindspore.nn as nn
import mindspore.common.initializer as weight_init
from src.lr_generator import get_lr, warmup_cosine_annealing_lr
@ -82,6 +83,7 @@ if __name__ == '__main__':
context.set_context(device_id=device_id, enable_auto_mixed_precision=True)
context.set_auto_parallel_context(device_num=args_opt.device_num, parallel_mode=ParallelMode.DATA_PARALLEL,
gradients_mean=True)
set_algo_parameters(elementwise_op_strategy_follow=True)
if args_opt.net == "resnet50" or args_opt.net == "se-resnet50":
context.set_auto_parallel_context(all_reduce_fusion_config=[85, 160])
else: