TensorFlow serving远程访问引擎的容器部署

时间 2019-12-05

标签 tensorflow serving 远程访问引擎容器部署繁體版

原文原文链接

TensorFlow Serving是经过rpc接口远程访问tensorflow引擎的服务器。git

TensorFlow Serving文档 https://tensorflow.github.io/serving/architecture_overview

TensorFlow Serving能够经过容器启动，方法以下：github

docker run --name tfserver --restart always tensorflow/tf_grpc_server:0.8.0 \
    --cluster_spec="worker|localhost:2220;localhost:2221,ps|localhost:3220;localhost:3221" \
    --job_name=worker --task_id=0

详细的TensorFlow Serving容器构建方法参考： https://github.com/openthings/serving/blob/master/tensorflow_serving/g3doc/setup.mddocker
在分布式环境中运行TensorFlow的详细指南： https://www.tensorflow.org/how_tos/distributed/#distributed-tensorflow服务器