Python Tutorial

Juyper Notebook based Python tutorial

Target Audience

Python beginners with experience in other programming languages (like Java, C++), who want to learn the specialities and features of Python.
Python programmers with medicore experience may find some useful information in this tutorial, too.

This tutorial is not intended for:

Beginners with no (object-oriented) programming experience - basic programming concepts are not introduced in detail, but expected from the reader.
Python experts (obviously...)

Structure

This tutorial is not structured linearly (from chapter 1-x), but the notebooks covering specific topics are rather independent from each other. I tried to illustrate the various points without requiring too much knowledge of other topics, but implicit dependencies could not be avoided.

Docker-based Installation

Install Docker (including docker-compose)

a. Windows 10 Professional: https://docs.docker.com/docker-for-windows/install/

b. other Windows versions: https://docs.docker.com/toolbox/toolbox_install_windows/

c. Linux (link is for Ubuntu): https://docs.docker.com/install/linux/docker-ce/ubuntu/
Install Git for Windows (in Linux usually already available): https://gitforwindows.org/

Clone this repository with Git into an empty directory

 git clone https://github.com/lungben/python_tutorial .

In docker-compose.yml, adjust the local paths (1st part before colon) in the volume statements to local directories (to be created beforehand).
Start the Jupyter and Database server:
```
 docker-compose up -d
```
Start Jupyter Lab: http://localhost:8888/lab

Update Containers

docker-compose down # stops and removes all containers
docker-compose pull # downloads newest image versions
docker-compose build # (re-) builds custom images
docker-compose up -d # starts container

Update Conda Packages

The official Jupyter Docker image usually does not contain the latest versions of the Conda packages. To update all packages, use the following commands:

docker exec -it tutorial_jupyter_1 bash # starts bash shell in Docker container
conda update -n base --yes conda
conda update --all --yes
exit # exists container shell

Warning: While updating all packages usually works fine for a local Conda installation, it seems to break the Jupyter Docker container rather often. Fortunately it is easy to restore Docker containers to their original state, just follow the instructions for Update Containers above.

Local Installation

Anaconda

The Anaconda package contains a large number of libraries and can be used out-of-the-box for most parts of this tutorial (except external databases).

The drawback is that it includes also a number of libraries which are not required, therefore it may be a bit too heavy for many purposes.

https://www.anaconda.com/distribution/#download-section

Download Python 3.x version.

Miniconda

Miniconda contains only the Conda package manager and its dependencies. It allows to create custom virtual environments with exactly the required packages.

https://docs.conda.io/en/latest/miniconda.html

Create virtual environment for tutorial:

conda update conda
conda create -n tutorial python=3.7 ipython notebook juypterlab numpy pandas matplotlib seaborn scipy numba pytest sympy sqlalchemy
activate tutorial

More packages may be installed later using

conda install -n tutorial package_name

Technology

This tutorial is written for Python 3.7+. No effort has been made or will be made for compatibility to legacy Python versions (e.g. Python 2.7).

The notebooks can be opened using Jupyter Notebook or Jupyter Lab.

It is assumed that all packages included in the Anaconda distribution are installed. If Miniconda is used, it may be required to install additional packages using

conda install [-c conda-forge] package --yes

Useful JupyterLab Settings:

Keyboard shortcuts:

{
    // Move cell up
    "notebook:move-cell-up": {
      "selector": ".jp-Notebook:focus",
      "command": "notebook:move-cell-up",
      "keys": [
        "Ctrl ArrowUp"
      ]
    },

    // Move cell down
    "notebook:move-cell-down": {
      "selector": ".jp-Notebook:focus",
      "command": "notebook:move-cell-down",
      "keys": [
        "Ctrl ArrowDown"
      ]
    }
}