docker hub: https://hub.docker.com/r/tensorflow/serving/tags/

git clone https://github.com/tensorflow/serving
cd serving/tensorflow_serving/tools/docker

以 ubuntu 18.04 為映像檔基底

FROM ubuntu:18.04

略過一些預設的套件

--no-install-recommends

安裝 Tensorflow Serving Serving 套件，複製 /usr/local/bin/tensorflow_model_server 到 /usr/bin/tensorflow_model_server。

COPY --from=build_image /usr/local/bin/tensorflow_model_server /usr/bin/tensorflow_model_server

指定 MODEL_BASE_PATH 位置

ENV MODEL_BASE_PATH=/models

建立資料夾

RUN mkdir -p ${MODEL_BASE_PATH}

設定模組名稱

ENV MODEL_NAME=model

產生 tf_serving_entrypoint.sh 檔，script 裡面的內容是

#!/bin/bash
tensorflow_model_server --port=8500 --rest_api_port=8501
--model_name=${MODEL_NAME} --model_base_path=${MODEL_BASE_PATH}/${MODEL_NAME}

RUN echo '#!/bin/bash \n\n\
tensorflow_model_server --port=8500 --rest_api_port=8501 \
--model_name=${MODEL_NAME} --model_base_path=${MODEL_BASE_PATH}/${MODEL_NAME} \
"$@"' > /usr/bin/tf_serving_entrypoint.sh \
&& chmod +x /usr/bin/tf_serving_entrypoint.sh

啟動 /usr/bin/tf_serving_entrypoint.sh

ENTRYPOINT ["/usr/bin/tf_serving_entrypoint.sh"]

ENTRYPOINT：和 CMD 一樣，用來設定映像檔啟動 Container 時要執行的指令，但不同的是，ENTRYPOINT 一定會被執行，而不會有像 CMD 覆蓋的情況發生。

完整的 Dockerfile 檔案

ARG TF_SERVING_VERSION=latest
ARG TF_SERVING_BUILD_IMAGE=tensorflow/serving:${TF_SERVING_VERSION}-devel

FROM ${TF_SERVING_BUILD_IMAGE} as build_image
FROM ubuntu:18.04

ARG TF_SERVING_VERSION_GIT_BRANCH=master
ARG TF_SERVING_VERSION_GIT_COMMIT=head

LABEL maintainer="gvasudevan@google.com"
LABEL tensorflow_serving_github_branchtag=${TF_SERVING_VERSION_GIT_BRANCH}
LABEL tensorflow_serving_github_commit=${TF_SERVING_VERSION_GIT_COMMIT}

RUN apt-get update && apt-get install -y --no-install-recommends \
        ca-certificates \
        && \
    apt-get clean && \
    rm -rf /var/lib/apt/lists/*

# Install TF Serving pkg
COPY --from=build_image /usr/local/bin/tensorflow_model_server /usr/bin/tensorflow_model_server

# Expose ports
# gRPC
EXPOSE 8500

# REST
EXPOSE 8501

# Set where models should be stored in the container
ENV MODEL_BASE_PATH=/models
RUN mkdir -p ${MODEL_BASE_PATH}

# The only required piece is the model name in order to differentiate endpoints
ENV MODEL_NAME=model

# Create a script that runs the model server so we can use environment variables
# while also passing in arguments from the docker command line
RUN echo '#!/bin/bash \n\n\
tensorflow_model_server --port=8500 --rest_api_port=8501 \
--model_name=${MODEL_NAME} --model_base_path=${MODEL_BASE_PATH}/${MODEL_NAME} \
"$@"' > /usr/bin/tf_serving_entrypoint.sh \
&& chmod +x /usr/bin/tf_serving_entrypoint.sh

ENTRYPOINT ["/usr/bin/tf_serving_entrypoint.sh"]

https://github.com/tensorflow/serving/blob/master/tensorflow_serving/tools/docker/Dockerfile

參考來源

• Dockerfile - ENV 與 ARG 參數