examples

Python example files

About this directory

The idea of this course is to be programming-driven. We offer a set of example programs using the basic functions of Python. Interested people should go through them and try to understand by themself. The links to the corresponding documentation is given in the header of each individual example file. The main Python documentation is online available under docs.python.org.

For the example which require data, the corresponding example data files are given in the files directory.

The order you should go through the examples is given below. After that tutorial, you should be able to reuse these examples to create your own applications.

How to start?

Go through the example files given in the next section and try to understand them. If you have problems with one file: The link to the documentation is usually given in the header. The complexity is increasing with each step. After going through all of these files, you should be able to write a Python script for plotting data with the following capabilities:

• A command line interface where the input filename and some other options like the title, axis description etc. can be defined
• The Data is read and parsed from the given file. Maybe some further analysis is done (min, mean, max values)
• The results should be exported to a pdf or text file so they can be used for further documentation (i.e., in a report)

Things to implement

Some ideas what you could implement for further understanding of the language:

• Hello world example
• Print a square / triangle
• Enter name and gender to formulate a greeting
• Print prime numbers
• Try out the random number generators
• Calculate some statistics from user-given numbers
• Create a simple phone book
• Plot a sine / cosine using matplotlib
• Read data from the files directory and plot statistics
• ...

Commented example files

The examples are sorted in an increasing order, i.e., they are getting more complex. The files are written in Python 3 not in Python 2. Please check weather you have installed the correction version. If you are new to Python, you should start at the beginning and go through all examples step by step. If you are a more experienced programmer, you can use these examples to lookup how to use certain functions.

1. 01_helloworldExample.py
2. 02_numberExample.py
3. 03_conditionsExample.py
4. 04_stringExample.py
5. 05_textInputExample.py
6. 06_datastructuresExample.py
7. 07_loopsExample.py
8. 08_castingExample.py
9. 09_enumerateExample.py
10. 10_functionExample.py
11. 11_datetimeExample.py
12. 12_randomExample.py
13. 13_numpyExample.py
14. 14_classExample.py, myClassExample.py
15. 15_csvExample.py
16. 16_argumentParserExample.py
17. 17_matplotlibExample.py
18. 18_pickleExample.py
19. 19_jsonExample.py
20. 20_regularExpressionExample.py
21. 21_exceptionsExample.py

Libraries and the corresponding references

This section briefly introduces commonly used libraries for Python. Some libraries offer similar functionality. The choice depends on the project requirements and often on the personal preferences.

SciPy

SciPy is a library for scientific calculations and offers for example algorithms for Fourier Transforms, Interpolation, Statistics etc.

• Webpage
• Reference documentation

numpy

Numpy is a library for efficient numeric calculations. It is a base library of SciPy.

• Webpage
• Reference documentation

Pandas

Pandas is a library for high-performance data analysis.

• Webpage
• Reference documentation

matplotlib

matplotlib is a graphing library for Python. It can create nice graphs comparable to the ones creates using Matlab. It is commonly used together mit Scipy and numpy.

• Webpage
• Examples
• Tutorials
• Reference documentation

Seaborn

Seaborn is based on matplotlib and specialized on statistical data visualization.

• Webpage
• Reference documentation

pyserial

Library for serial communication in Python.

• Reference and webpage

tkinter

tkinter is a library for platform-independent GUI-design using Python.

• Webpage
• Referecne documentation

kivy

kivy is Python library for the development of touch GUI applications on Windows, Linux, OS X, Android, iOS and RaspberryPi.

• Webpage
• Examples
• Reference documentation

SQLAlchemy

SQLAlchemy is an ORM (object-relational mapper), i.e. it represents database interactions like for example with SQLite as objects in Python. Using SQLAlchemy, one can write data to a database without knowing SQL.

• Webpage
• Tutorial
• API

Virtual environment

Sometimes a project requires the installation of libraries which are either not available as packages for the current used system or the user can not install them (limited system access, beta version etc.).

For this case, Python3 comes with a module called virtualenv (virtual environment). It creates a separate directory with an isolated Python path. virtualenv is used on the command line and is handled as follows:

• Create a new virtual environment in the current directory: python3 -m virtualenv . This directory only has access to the core libraries of the system.
• Create a new virtual environment in the current directory which has access to the system wide installed libraries: python3 -m virtualenv . --system-site-packages
• Activate the virtual environment: . ./bin/activate. The name of the environment will be indicated with its name on the command line
• Deactivate the currently used virtual environment: deactivate
• Install packages in the environment: pip3 install <packagename>

Additional information can be found in the official Python3 documentation of the package

FAQ

This section contains the frequently asked questions

Package cannot be found

Depending on your Python installation, you might get an error message complaining that a package is missing like for example the numpy package. You can install most packages using the pip tool.

• On Unix / Linux / MacOS, you can usually use pip3 install <package name> to install packages
• On Windows, you will find the corresponding command in the C:\Python2x\Scripts directory. Open a shell, change into that directory and execute the command as mentioned above

For this tutorial, the following packages are required:

• pip3 install numpy
• pip3 install python-dateutil
• pip3 install matplotlib

For Linux based systems, you can also install the packages using the system package manager. On a Debian / Ubuntu system, the following commands will install the corresponding packages. You should prefer this way over the pip command:

• sudo apt install python-numpy
• sudo apt install python-dateutil
• sudo apt install python-matplotlib

I got an IndentationError

Maybe you are mixing space an tabs. According to the documentation (https://www.python.org/dev/peps/pep-0008/#tabs-or-spaces), you should stick to spaces. It is also a good idea to setup you text editor to highlight tabs and replace tabs by spaces (preferable 4 spaces).

Timestamp vs. Datetime

Dates and times can be represented in different formats. The human-readable format (2017-04-08 01:00PM or 8.4.17 13:00) and Unix timestamp are the most common ones. The timestamp gives the number of seconds since 1970-01-01 and has several advantages in performing calculations and storing the values. In most cases, it is easier to convert dates and times to timestamps and only convert them back if they are shown to the user.

https://en.wikipedia.org/wiki/Unix_time

Naming your programs

When naming your program, you should ensure that you are not using a reserved name (i.e. a module name) as Python will try to import your app instead of the system module. So do not use filenames like matplotlib.py, numpy.py, int.py etc.

Tuples vs. Lists

Tuples and lists look quite similar but are showing crucial differences. Tuples are fixed (immutable) and cannot be changed after the assignment but are faster compared to lists. Lists are dynamically assigned (mutable) and thus are more flexible but are slower in processing.

So use tuples if

• You need a write-protected structure
• A faster structure

For everything else you should prefer lists