Dockerfile to Create an Image with scikit learn and jupyter notebook Dependencies

This repo has a Dockerfile for python 3, scikit-learn, and Jupyter configured to access the app folder contents on host.

Usage:

Docker is essentially like a VM (Virtual Machine), but much lighter.

Running this command, in the folder containing the Docker file
('.' stands for current folder):

docker build -t mlnb .

• creates an image (similar to a VM disk image) based on this Dockerfile,
• with a name/tag "mlnb" (chose it for "machine learning notebook"),

To see available images on your system:

docker image ls

This Docker run command:

docker run -p 9999:8888 \
--mount 'type=bind,src="$(pwd)"/app,target=/app' mlnb

• creates a new instance for "mlnb" image and runs it
• maps internal port 8888 to the host's 9999 and
• bind mounts the app folder in this folder to app folder in the docker container

To list running instances

docker ps

To make a terminal connection to a running instance:

docker exec -it <containerName> /bin/bash

More

You can study my Docker Workshop: https://github.com/ATLD/docker_workshop

Serving jupyter notebook as a Slideshow

In folder containing the notebook:

 git submodule add https://github.com/hakimel/reveal.js.git reveal.js
 jupyter nbconvert "<notebook name>.ipynb" --to slides --post serve

Also add 8000 to mapped internal ports when starting this image

docker run -p 9999:8888 -p 9000:8000 \
--mount 'type=bind,src="$(pwd)"/app,target=/app' mlnb