After successful completion of the course, students are able to program in Python in a data-oriented way, using SciPy, NumPy and Pandas; explain the fundamentals of machine learning and network analysis, and implement a Data Science project.

The following topics are covered in the lectures:

• Introduction to doing Data Science
• SciPy, NumPy, vectorisation, execution performance measurement
• Data preparation, structuring, fusion with Pandas
• Data Science solution approaches and case studies
• Introduction to machine learning
• Introduction to network analysis

Lectures about the fundamentals

2 practical exercises (Exercise 1 is done individually, Exercise 2 is done in a group)

The lectures are online. The link to the online lectures is on TUWEL.

All other sessions are in presence (if they must be moved online due to the pandemic, then an announcement will be made).

Dates

All Lectures on Tuesday 12:00 c.t.-13:45.

1. Kickoff-Session, data science process, community, solution examples, introduction to DOPP (4.10.2022)

2. SciPy, NumPy, vectorisation, visualisation, benchmarking (11.10.2022)

3. Preprocessing, Pandas (18.10.2022)

4. Intro to Machine Learning (25.10.2022)

5. Network Analysis (8.11.2022)

6. Introduction to Text Processing (22.11.2022)
7. Data suitability, Data biases (29.11.2022)

Exercise-related sessions

Review meetings for exercise 2 (15 minutes for each group):

• 13.12.2022, 9:00-16:00
• 14.12.2022, 9:00-16:00

Exercise 2 consultation sessions in EI11 at the usual lecture times (voluntary):

• 20.12.2022
• 10.1.2023
• 17.1.2023

Project presentation: 24.1.2023, 9:00-18:00

The effort breakdown is:

Python test: 3h
Lectures: 7 sessions @ 2h: 14h
Exercises:
    EX1 (data science process): 22h
    EX2 (project): 36h [includes review meeting (topic + questions + work plan)]
SUM: 75h

It is necessary to pass the Python test at the beginning of the course to be able to complete the course. Support is available for this - see the details in the section "Previous knowledge". The self-assessment is a good indicator of what you need to know for the test. Note that only one of the two offered Python tests must be taken.

Two practical exercises. The second exercise requires a report, Jupyter Notebook, and presentation of the results.

Basic proficiency in programming with Python is expected for this lecture. A self-assessment is provided: https://github.com/tuw-python/tuw-python-2022WS/blob/main/self_assessment.ipynb

To assist in achieving the required proficiency in Python, the one week intensive course "194.123 Programming in Python" can be taken. The materials for this course are available to all and can also be worked through without attending the course.

A Python proficiency test is held at the beginning of the course. It is necessary to pass this test to be able to pass the course. Failing the Python test means that you cannot continue the course, but will not result in a negative certificate for the whole course.