Skip to content

Latest commit

 

History

History
83 lines (70 loc) · 3.02 KB

README.md

File metadata and controls

83 lines (70 loc) · 3.02 KB
Raw

dwchallenge-docker

A docker image for Machine Learning exercises. Contains Anaconda plus some extra packages and customisations.

Usage

Setup

Warning: At the first execution this can download and take more then 5GB of your diskspace.

Having docker installed you can execute jupyter notebook to work with your notebooks like that:

docker run -p $your_port:8888 -e JUPYTER_TOKEN=$your_secret_token -v $your_notebooks_folder_path:/data/notebooks horizoncrafts/dwchallenge:latest

where:

  • $your_port can be any port available on your local host (your notebook, for exmple). 8888 works usually.
  • $your_secret_token is the token required when you login to jupyter notebook. Leave it empty or skip -e JUPYTER_TOKEN=$your_secret_token part so your coleagues on the same network can access your jupyter notbook without restricton (if not firewalled). Otherwise they'll need to provide the token (more secure).
  • $your_notebooks_folder_path is a folder where your notebooks are located on your computer.

In the simplest case you can have it started like that:

docker run -p 8888:8888 -v "c:\Users\me\notebooks":/data/notebooks horizoncrafts/dwchallenge:latest

then go to http://localhost:8888

If protecting with a token you'll be asked to type it in. You can also connect to jupyter notbook providing token in the url, for exmple:

http://localhost:8888/tree?token=my_secret_token

Check this note: https://jupyter-notebook.readthedocs.io/en/latest/security.html#alternatives-to-token-authentication to learn more about authentication.

Docker command line operations

Check running containers:

docker ps

all containers, including stoped:

docker ps -a

To kill a container (opposite to docker run described above):

docker kill [container id]

To stop or start a containter (like suspending and waking up):

docker stop [container id]
docker start [container id]

Connect to command line of a container:

docker exec -it [container id] /bin/bash

or use exclamation mark in python cell in jupyter notebook.

docker-compose.yml example

version: '3.7'
services:
  jupyter:
    image: horizoncrafts/dwchallenge:latest
    ports:
      - $your_port:8888
    environment:
      JUPYTER_TOKEN: $your_secret_token
    volumes:
      - type: bind
        source: $your_notebooks_folder_path
        target: /data/notebooks

References

This image is somehow inspired on the Continuum Analytics original one (https://hub.docker.com/_/anaconda), plus it has the minimum stuff required for Data Workshop challenges.

Notes

Data Workshop challengers!

Each tag represents the state of the container at the completion of a given challange. Thus when solving a given challenge always use the image tagged with the previous challenge. For example, when you're at challenge 4 use image tagged "challenge03" instead of latest. It looks like this:

docker run -p 8888:8888 -v "c:\Users\me\notebooks":/data/notebooks horizoncrafts/dwchallenge:challenge03

This way you can excercise all tasks within a challenge.