docker搭建python/tensorflow学习环境

编写dockerfile

• 从Docker Hub拉取tensorflow/tensorflow:2.10.0-jupyter的Docker镜像，里面包含了tensorflow的全部依赖
• 指定工作区目录
• 从requirements.txt获取其他依赖包
• 安装jupyter依赖
• 指定端口
• 配置默认启动命令

FROM tensorflow/tensorflow:2.10.0-jupyter

WORKDIR ~/tf-learning

COPY requirments.txt requirements.txt

RUN pip install jupyter_core --upgrade -i https://pypi.tuna.tsinghua.edu.cn/simple
RUN pip install -r requirements.txt -i https://pypi.tuna.tsinghua.edu.cn/simple some-package
RUN pip install mkl mkl_service

EXPOSE 8888

ENTRYPOINT ["jupyter","lab","--ip=0.0.0.0","--allow-root","--no-browser"] 

编写requirements.txt

可以将所需安装的pip包放在此处

jupyterlab
pandas
matplotlib
numpy
gensim
joblib
networkx
numba
scipy
scikit-learn

编写docker-compose.yaml文件

描述应用程序的服务、网络、卷等资源，并以声明式的方式定义这些资源如何协同工作

version: '1.0'
services:
  jupyter-lab:
    build: .
    ports:
      - "8888:8888"
    volumes:
      - ~/tf-learning:~/tf-learning
    deploy:
      resources:
        limits:
          cpus: "2.00"
          memory: 2G