简介

TensorFlow Serving is a flexible, high-performance serving system for machine learning models, designed for production environments. TensorFlow Serving makes it easy to deploy new algorithms and experiments, while keeping the same server architecture and APIs. TensorFlow Serving provides out-of-the-box integration with TensorFlow models, but can be easily extended to serve other types of models and data.

搭建

具体搭建可以参考官方文档：https://tensorflow.github.io/serving/。

https://medium.com/zendesk-engineering/how-zendesk-serves-tensorflow-models-in-production-751ee22f0f4b

第一步

mkdir test_dir
touch WORKSPACE
mkdir test_dir/data
# cp
cp $train_files $test_files vocab test_dir/data

example

git clone https://github.com/tensorflow/serving.git

启动服务

bazel-bin/tensorflow_serving/model_servers/tensorflow_model_server --port=9000 --model_name=output --model_base_path=$DIR/output

参考