Python development environment in the container

Software to install

Install Docker

Install Git

Use dockerhub to find proper images

Examples

1
docker pull jupyter/datascience-notebook

Now we can build and run the container and target it to local folder for output files

1
docker run -it -p 8888:8888 -v D:/Code/Python/Grokking_ML/jupiter_notebooks:/home/jovyan/projects jupyter/datascience-notebook

where arguments are port to listen, folder to target and container to run

-v paramter here has next format -v {host_path}:{container_path}

with container running this way you can either work with jupiter notebook through your browser or using VSCode you can open remote window by
attaching to running container and with jupiter extension write code directly from VSCode

Alternative method

For alternative method we would rely on anaconda docker containers

Anaconda docker

Next step is to create docker container, we will use it to attach through VSCode and work from created environment

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
# For more information, please refer to https://aka.ms/vscode-docker-python
FROM continuumio/anaconda3

# Keeps Python from generating .pyc files in the container
ENV PYTHONDONTWRITEBYTECODE=1

# Turns off buffering for easier container logging
ENV PYTHONUNBUFFERED=1

# Install pip requirements
COPY requirements.txt .
RUN python -m pip install -r requirements.txt

# Install jupiter
RUN bash -c "/opt/conda/bin/conda install jupyter -y --quiet # && mkdir \ /opt/notebooks && /opt/conda/bin/jupyter notebook \ --notebook-dir=/opt/notebooks --ip='*' --port=8888 \ --no-browser --allow-root"

Additionally there are usefull tasks for vscode to automate

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
{
  "version": "2.0.0",
  "tasks": [
    {
      "type": "shell",
      "label": "docker-clean",
      "command": "docker rmi $(docker images --filter \"dangling=true\" -q --no-trunc)",
      "problemMatcher": []
    },
    {
      "label": "run-docker-container",
      "type": "shell",
      "command": "docker run -it -p 8888:8888 -v D:/Code/Python/Grokking_ML/jupiter_notebooks:/app anaconda_ml"
    },
    {
      "type": "docker-build",
      "label": "docker-build",
      "platform": "python",
      "dockerBuild": {
        "tag": "anaconda_ml:latest",
        "dockerfile": "${workspaceFolder}/Dockerfile",
        "context": "${workspaceFolder}",
        "pull": true
      }
    }
  ]
}