!11799 vgg16+imagenet performance optimizing

From: @caojian05
Reviewed-by: @oacjiewen
Signed-off-by:

model_zoo/official/cv/vgg16/scripts/run_distribute_train.sh

@@ -52,21 +52,25 @@ cpus=`cat /proc/cpuinfo| grep "processor"| wc -l`
 avg=`expr $cpus \/ $RANK_SIZE`
 gap=`expr $avg \- 1`
 
+script_dir="$( cd "$( dirname "${BASH_SOURCE[0]}" )" && pwd )"
+src_dir=$script_dir/..
+
+start_idx=0
 for((i=0;i<RANK_SIZE;i++))
 do
     start=`expr $i \* $avg`
     end=`expr $start \+ $gap`
     cmdopt=$start"-"$end
 
-    export DEVICE_ID=$i
+    export DEVICE_ID=`expr $i \+ $start_idx`
     export RANK_ID=$i
-    rm -rf ./train_parallel$i
-    mkdir ./train_parallel$i
-    cp *.py ./train_parallel$i
-    cp -r src ./train_parallel$i
-    cd ./train_parallel$i || exit
+    rm -rf ./train_parallel$DEVICE_ID
+    mkdir ./train_parallel$DEVICE_ID
+    cp $src_dir/*.py ./train_parallel$DEVICE_ID
+    cp -r $src_dir/src ./train_parallel$DEVICE_ID
+    cd ./train_parallel$DEVICE_ID || exit
     echo "start training for rank $RANK_ID, device $DEVICE_ID, $dataset_type"
     env > env.log
-    taskset -c $cmdopt python train.py --data_path=$2 --device_target="Ascend" --device_id=$i --is_distributed=1 --dataset=$dataset_type &> log &
+    taskset -c $cmdopt python train.py --data_path=$2 --device_target="Ascend" --device_id=$DEVICE_ID --is_distributed=1 --dataset=$dataset_type &> log &
     cd ..
 done

model_zoo/official/cv/vgg16/src/config.py

@@ -47,14 +47,14 @@ cifar_cfg = edict({
 # config for vgg16, imagenet2012
 imagenet_cfg = edict({
     "num_classes": 1000,
-    "lr": 0.01,
+    "lr": 0.04,
     "lr_init": 0.01,
     "lr_max": 0.1,
     "lr_epochs": '30,60,90,120',
     "lr_scheduler": 'cosine_annealing',
     "warmup_epochs": 0,
-    "batch_size": 32,
-    "max_epoch": 150,
+    "batch_size": 64,
+    "max_epoch": 90,
     "momentum": 0.9,
     "weight_decay": 1e-4,
     "loss_scale": 1024,

model_zoo/official/cv/vgg16/train.py

@@ -61,7 +61,7 @@ def parse_args(cloud_args=None):
     parser.add_argument('--lr_gamma', type=float, default=0.1,
                         help='decrease lr by a factor of exponential lr_scheduler')
     parser.add_argument('--eta_min', type=float, default=0., help='eta_min in cosine_annealing scheduler')
-    parser.add_argument('--T_max', type=int, default=150, help='T-max in cosine_annealing scheduler')
+    parser.add_argument('--T_max', type=int, default=90, help='T-max in cosine_annealing scheduler')
 
     # logging and checkpoint related
     parser.add_argument('--log_interval', type=int, default=100, help='logging interval')
@@ -140,7 +140,7 @@ if __name__ == '__main__':
         device_num = args.group_size
         context.reset_auto_parallel_context()
         context.set_auto_parallel_context(device_num=device_num, parallel_mode=ParallelMode.DATA_PARALLEL,
-                                          gradients_mean=True, all_reduce_fusion_config=[3, 10, 12, 15])
+                                          gradients_mean=True, all_reduce_fusion_config=[2, 18])
     else:
         if args.device_target == "Ascend":
             context.set_context(device_id=args.device_id)