Python Transition Course (EN)¶
Context of the Data Science Python Transition Course (dsci-pytrans):
Some basic programming prerequisites are already existent. We expect about 1 year (200h) of programming study time.
dsci-pytrans refreshes these basic prerequisites based on Python, and also explains some typical pythonic elements which probably will be new for many students.
dsci-pytrans is a nonobligatory strongly recommended precondition to enter the course dsci-main
Programming Prerequisites¶
We assume that you already have acquired basic programming skills and passed the courses with at least a mediocre (DE: Note ausreichend) result
Informatik 1: 100 hours total studies of imperative programming with a language like Python, C, Java etc.
variables, datatypes, control structures, functions, modules, I/O with text files, formtatted printing, string handling, basic type conversion
be able to implement simple algorithms like bubble sort
Informatik 2: 100 hours total studies of OOP
We assume your programming experience dates back up to 3 years, thus much of the content may be forgotten. However, there was a point in time where the main ideas of programming and some basic algorithms where understood. This basic understanding is an inevitalble prerequisite of our course.
We offer a Python transition Course on the basis of Python to
catch up forgotten (but already understood) basic programming skills
migrate from other procedural languages (typically C or Java) to Python
The course ends with a lightweight Python Transition exam, based on a set of short programming quests. The are available publicly, as it is also true for you curriculum. Experienced learners can teach and test themselfes completely self-paced. Other learners may want to interact with our course lecturer.
The course is voluntary, nonobligatory. Passing the course is not a requirement to participate in the main course DSCI. However, the lecturer of the main course dsci will not provide support w.r.t. basic programming skills. Thus we strongly advice to take the transition course (and to take it seriously).
Curriculum¶
The curriculum is contained in this little book:
available on github as a set of ipynb: https://github.com/jakevdp/WhirlwindTourOfPython
Subject 1: Linux Command Line¶
Objective: Most students do have user-lever experience on Windows. However, in order to fully benefit from the jupyter ecosystem we need to work with a Linux (here: XUbuntu) Platform.
Installing tools like Jupyter, Conda, Jupyterbook, LaTeX etc. on Ubuntu is simple - but configuring them to work together can be a challenge. Thus we provide a fully configurated virtual data science lab (an XUbuntu Oracle VirtualBox virtual machine) to the students.
Operating System Subjects include:
command line, bash essentials etc.
jupyter notebook: start, stop, access with firefox etc.
where are your files? locate them in bash, in notebook
We already have installed miniconda. Many libraries are missing: install them with conda and/or pip
Subject 2: Fundamental Python for Java / C Programmers¶
Objective:
Refresh fundamental programming concepts which are already known from lanuages like C, Java etc.
Fundamental Python Subjects include (but are not restricted to)
Data types
List, Dictionary (not only) Comprehensions, lambda, map, reduce, filter
Collection Data types
Control Structures and Functions
Modules and packages
File management, input, output
Object Oriented Programming
Exercises¶
Yes, a student has to know the subjects of the course. This means she can explain every single code snippet in the whirlwind book.
But knowing isn’t enough. In the end a student has to solve problems with a computing language. Consequently a student has to be able to demonstrate practical programming skills to pass the course.
We provide a bunch of micro challences in order to test practical programming skills and to allow for goal-oriented self-paced learning. Currently we explore several cloud platforms to reuse and/or to amend sets of Python micro challenges:
platforms wich provide test facilities:
Gamification:
TBD (to be done): Are there enough and specific programming self test opportunities already aroud, or do we want to / have to provide additional own micro challenges?