Python Programming for Machine Learning (PyML)

Python Programming for Machine Learning (3 ECTS credits) is an optional course within one of the modules:

• "Cognitive Algorithms" (summer and winter semesters)

• "Machine Learning 1" (summer semester).

It's not possible to take the class as a standalone, seminar, or free-of-choice module.

Participation in the exercise sessions is not mandatory but highly recommended. Homework assignments must be submitted every week. You must be enrolled on ISIS to submit homework. If you do not register on time, you cannot pass the course.

You can choose between two (almost identical) courses. More information will follow as we approach the semester.

PyML A:

• Course Period	April 24th - May 19th 2023
  	In-person: Friday, 09:00 - 12:00 p.m. (90 min lecture and 90 min exercise)
  	Virtual: to be announced (90 min lecture, recorded)
  Trainers	Christopher Anders: anders [at] tu-berlin.de
  	Panagiotis Tomer Karagiannis

PyML B:

• Course Period	June 5th - June 30th 2023
  	In-person: Monday, 14:00 - 17:00 p.m. (90 min lecture and 90 min exercise)
  	Virtual: to be announced (90 min lecture, recorded)
  Trainers	Jannik Wolff: wolff.jannik [at] icloud.com
  	Panagiotis Tomer Karagiannis

General information (valid for PyML A and B):

• Language	English
  Exams (choose one)	June 9th, 12:30 - 02:30 p.m.
  	TBD
  	TBD
  ISIS	Link

Homework

• One homework per week

• Due dates: usually Mondays 23:55. Late submissions will not be graded!

• Homework must be submitted via the ISIS submission portal.
• Homework must be completed by yourself.

Passing the course

• The course is passed if the exam is passed (grade 4.0 or better) and all homeworks are completed successfully. (50/100 points or more, taking part in the exam requires successful completion of all homework)

• The final course grade is determined by the exam only.

Frequently Asked Questions (FAQ)

• Where is the link for the class?

  • Links for the online lectures and exercises will be announced on ISIS just before they start.

• Is prior programming/Python knowledge necessary?

  • Knowledge of elementary programming concepts (in Python or another language) will be helpful. Lack of such knowledge will increase the time demand of the class.

• Which exam is compatible with PyML A/B?

  • All exams are compatible with PyML A or B. You can choose the exam that fits your time schedule best.

• How to register for the course?

  • Pre-registration via the exam registration office is not needed. However, register for the ISIS course in time (see above) to be able to submit the exercises.

• May I participate in the class during this semester and take part in the corresponding module in one of the following semesters ?

  • Yes, just after you've passed the class, your results are also valid for the next semesters.

• I've already successfully passed all of the homework in the previous semester but failed/missed the exam. Should I resubmit them again?

  • No, you do not need to resubmit the homework. You may take part directly in the final exam. Enroll for the class via ISIS and wait for the announcements.

• Until when do I have to register at QISPOS? I do not have an ISIS account yet.

  • You have to register for a corresponding module, either ML or CA, until the deadline. Usually it is six weeks before the first exam.

• Do I have to attend both exercise events every week or only one of them?

  • The exercise sessions are not mandatory. You can think of them as Q&A sessions, where you can ask questions regarding the homework.

• I will not be able to attend a lecture. Is this a problem?

  • All lectures and learning materials will be made accessible online. You only need to submit your homework before the deadline. Ask in the ISIS forum if you have questions regarding lecture content.

Students from other universities

If you are not a student at TU and want to earn credit, you have to solicit ''Nebenhörerschaft'':

• First fill the online ''form'' with the following details on the final page.

  Course title and type	Program number	Number of course hours per week (SWS)	Lecturer
  Python Programming for Machine Learning (KU)	0434 L 543	2	(name of the trainer)

• Send the form to one of the trainers (see contact information above) to sign
• Then send the signed form to Manuela Gadow (manuela.gadow at tu-berlin.de), who is authorized to sign on behalf of our dean.

• After getting your signed form back send it together with your current matriculation letter ('Immatrikulationsbescheinigung') to the student registration office (nebenhoerer@studsek.tu-berlin.de).

Please make sure to trigger the above process on time. You may need to go through some bureaucracy to attain ISIS (the university's web portal) access, which is necessary to submit the mandatory homework.