README.md 1.26 KB
Newer Older
qianyj's avatar
qianyj committed
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
# 简介

该测试用例可用于ResNet50/Vgg16等网络的性能测试及精度验证。 


# 单卡测试 (单精度)

## 运行
    
    export HIP_VISIBLE_DEVICES=0  
    python3 ./benchmarks-master/scripts/tf_cnn_benchmarks/tf_cnn_benchmarks.py --data_format=NCHW --batch_size=128 --model=resnet50 --optimizer=momentum --variable_update=parameter_server  --print_training_accuracy=true  --nodistortions --num_gpus=1 --num_epochs=90 --weight_decay=1e-4 --data_dir=$data_dir_path   --use_fp16=False --data_name=imagenet --train_dir=$save_checkpoint_path

# 单卡测试 (混合精度)

## 运行
qianyj's avatar
qianyj committed
16
    export HIP_VISIBLE_DEVICES=0    
qianyj's avatar
qianyj committed
17
18
19
20
21
22
23
24
25
26
27
28
    python3 ./benchmarks-master/scripts/tf_cnn_benchmarks/tf_cnn_benchmarks.py --data_format=NCHW --batch_size=128 --model=resnet50 --optimizer=momentum --variable_update=parameter_server  --print_training_accuracy=true  --nodistortions --num_gpus=4 --num_epochs=90 --weight_decay=1e-4 --data_dir=$data_dir_path   --use_fp16=True --data_name=imagenet --train_dir=$save_checkpoint_path

# 多卡测试 (单精度)

## 运行

    mpirun -np 4 --hostfile hostfile --bind-to none scripts-run/single_process.shhi  
    
# 参考资料
[https://github.com/tensorflow/benchmarks/tree/master/scripts/tf_cnn_benchmarks]
[https://github.com/horovod/horovod]