Quick Docker Environment for TensorFlowJS Node

.env.sample

root=/path/to/root
project=/path/to/project

README.md

Quick Docker environment for TensorFlowJS Node

Personal docker environment for TensorFlowJS Node for Development purpose

Why?

Because:

1. I didn't want to install Xcode (which also takes up a lot of space)
2. Easy to setup on different machines

How to use this?

Assuming you have Docker setup completed:

If you plan to make use of compose file shared here:

1. Create a .env file from .env.sample and specify root and project volumes.

And then:

1. Run $ docker-compose build command to build the image
2. Run $ docker-compose run tensorflowjs command to run the container.

Otherwise:

1. Run $ docker build . command to build the image
2. Run $ docker run -it --rm <imageid> command to run the container.

Why is this not on DockerHub?

I haven't given it much thought

docker-compose.yml

version: '3'
services:
  tensorflowjs:
    build:
      context: .
      dockerfile: Dockerfile
    image: tfjs-node:cpu
    env_file:
        - .env
    volumes:
        - "${root}:/root"
        - "${project}:/root/project"
    command: /bin/bash

Dockerfile

FROM ubuntu:16.04

LABEL maintainer="Manraj Singh Grover <manrajsinghgrover@gmail.com>"

# Install basic CLI tools etc.
RUN apt-get update && apt-get install -y --fix-missing --no-install-recommends \
        build-essential \
        curl \
        git-core \
        iputils-ping \
        pkg-config \
        rsync \
        software-properties-common \
        unzip \
        wget

# Install NodeJS
RUN curl --silent --location https://deb.nodesource.com/setup_10.x | bash -
RUN apt-get install --yes nodejs

# Install yarn
RUN npm install -g yarn

# Install tfjs-node
RUN yarn add @tensorflow/tfjs
RUN yarn add @tensorflow/tfjs-node

# Clean up commands
RUN apt-get autoremove -y && apt-get clean && \
    rm -rf /usr/local/src/*

RUN apt-get clean && \
    rm -rf /var/lib/apt/lists/*

# Set working directory
WORKDIR "/root/project"
CMD ["/bin/bash"]