28 August 2023 to 1 September 2023
FZ Jülich
Europe/Berlin timezone

The Palestinian-German Science Bridge and Helmholtz AI offer a basic course and a workshop on machine learning for PIs and supervisors/senior scientists in both natural and computer science. Potential applications of machine learning and deep learning in natural science are numerous; they include detector development, data analysis techniques, and even physical modelling. The event consists of an introductory and an advanced module, which you can book independently if you are not a core participant. The content beyond the basic module with an introduction to deep learning will be decided based on a survey conducted among the participants before the event. 

 

The highly interactive, hands-on introductory module will equip the participants with basic tools to implement and use machine learning methods. All participants are expected to contribute to the course actively. A Python course that introduces participants that are not familiar using basic Python, pandas, and Matplotlib to the necessary basics and packages will be offered prior to the machine learning modules as a virtual course. 
 

Modules:

  • deeplearning in 540 minutes: In the first module on 28-30 August, the machine learning course will enable participants to embark on clustering, classification, and regression projects using “scikit-learn” or “keras”. (If you are already familiar with implementing machine learning in Python and do not belong to the core participants of the workshop, you can skip this module.)
  • Advanced machine learning: In the second module on 31 August and 1 September, ample opportunities will be provided for working on advanced topics and current research projects, which will be selected considering the group's needs. Advanced topics may include--for example--reinforcement learning, genetic algorithms, computer vision, how to determine the robustness and reliability of a model, strategies if the basic algorithm fails, and physics-informed machine learning. In addition to advanced machine-learning topics, the advanced content may include an extended mathematical introduction to theoretical concepts, enriching the workshop's benefits and outcome. (If you want to learn the basics of machine learning only and do not belong to the core participants of the workshop, you can skip this module.)
     

Prerequisites: Knowledge of basic Python, Pandas, and Matplotlib is required for successful participation. It will be provided to those that are not yet familiar in our preparatory PGSB Python course that will be delivered online (Registration: https://events.hifis.net/e/pgsb_python_23).

 

Participants: Please note that this course is aimed at the PI/supervisor/IT/staff scientist/postdoc level and is not intended for PhD students. A first selection of participants will take place soon after 6 August 2023.

 

Timetable: The timing may be adjusted to meet the needs of the participants.

Starts
Ends
Europe/Berlin
FZ Jülich
building 04.8, room 142/143
Room change on Friday!

Should I participate in the preparatory Python course?

 

To determine whether you should participate in the preparatory online Python course before participating in Machine Learning, please read through the following questions ask yourself whether you are able to solve these tasks. Please note that this list is not complete! In addition, please try to judge how well you can deal with large datasets (pandas) and plotting (Matplotlib). The modules of the Python course "Basic Python", "pandas", and "Matplotlib" can also be booked separately in case you do have previous knowledge in some but not all areas!

- You are provided with a Python list of integer values. The list has length 1024 and you would like to obtain all entries from index 50 to 101. How would you do this?
- You need to open 102 data files and extract an object from them. For this, you compose a small function to open a single file which requires 3 input parameters. The parameters are a file location, the name of the object to retrieve and a parameter that controls the verbosity of the function. The latter parameter has the default value False.
- You call a function in your code that sometimes throws a ValueError exception. You’d like to intercept that exception and process the traceback before re-raising the exception. How would you do that?
- You are provided a list of 512 random float values. These values range between 0 and 100. You would like to remove any entry in this list that is larger than 90 or smaller than 10.
- You are provided with a CSV file. The file has 35,000 rows. The file has 45 columns. Each column is separated by the “,” symbol from the other. You would like to open the file in Python, calculate the arithmetic mean, the minimum and maximum of column number 5, 12 and 39.

Application
Application for this event is currently open.
Surveys
There is an open survey.