[TensorFlow Serving] Linux下TenforFlow Serving的安装

2020-07-07 743 阅读1分钟

介绍

借助Docker，将训练好的TensorFlow模型直接上线并提供服务

安装准备

docker的安装：Linux下Docker的安装

安装

# 通过docker拉取tensorflow serving镜像
docker pull tensorflow/serving

# 下载官方示例代码
mkdir -p /tmp/tfserving
cd /tmp/tfserving
git clone https://github.com/tensorflow/serving

踩坑：docker pull超时

替换加速器镜像

rm /etc/docker/daemon.json
vim /etc/docker/daemon.json
# content:
# {
#     "registry-mirrors": ["https://docker.mirrors.ustc.edu.cn/","https://hub-mirror.c.163.com","https://registry.docker-cn.com"],
#     "insecure-registries": ["10.0.0.12:5000"]
# }

# 重启加速器
systemctl restart docker

使用

# 运行TF Serving
docker run -dt -p 8501:8501 -v "/tmp/tfserving/serving/tensorflow_serving/servables/tensorflow/testdata/saved_model_half_plus_two_cpu:/models/half_plus_two" -e MODEL_NAME=half_plus_two -t tensorflow/serving

# 客户端验证
curl -d '{"instances": [1.0, 2.0, 5.0]}' \
  -X POST http://localhost:8501/v1/models/half_plus_two:predict 
  
# 返回结果
# { "predictions": [2.5, 3.0, 4.5] }

Reference