PHYS-E0549 - Introduction to Machine Learning for Materials Science D, Lecture, 5.9.2022-11.10.2022
This course space end date is set to 11.10.2022 Search Courses: PHYS-E0549
Topic outline
-
In machine learning, we write programs that the machine executes to process data and to learn. To understand machine learning therefore means to understand also how these instructions to the machine are composed. Python has evolved into the standard programming language in machine learning and we will use it in this course for the machine learning tutorials. The course will provide a gentle introduction into Python in the first tutorial, but it would be advantageous, if you have some programming experience (not necessarily in Python) prior to taking the course. We have devised a short pre-assessment notebook for you to test, if you have sufficient programming knowledge. Please go through this pre-assessment, before you decide to sign up for the lecture.
Pre-assessment
Here is the link for a short Google Colab notebook that we have designed for you to test your Python knowledge. If you can easily complete the tasks, you have sufficient knowledge for the course. If you know how to complete the tasks in a different language (e.g. C, Fortran, Scala, Matlab), but are unsure about how to do them in Python, you can brush up your Python knowledge before the course (see below) and sign up for the course. If you have no programming experience at all, it would be advisable to first acquire rudimentary Python skills and take this course next year instead.
Python and machine learning resources
The University of Helsinki has developed the Elements of AI free online course. This is an excellent resource to start familiarising yourself with machine learning and its practical aspects. The course can be taken in your own time. It is not a prerequisite for this course, but the 2nd part "Building AI" might be useful for you, if you are not sure about your Python knowledge.
CSC - IT Center for Science provides a Beginner Python course (~10h to complete), which is also available as Jupyter Notebooks. You can also find many Python learning resources online and we encourage you to explore options that work best for you.