Welcome to the "Between Models and Reality" Ph.D. school on Machine Learning in physics. The focus of this school is the application of Machine Learning to physics research and the many challenges and solutions that this entails. The focus is mainly on application, and will not only consist of lectures.

So if you are a Ph.D. student in a field of physics (or a related "large research data" science), who would like to learn more about both basic but also more advanced Machine Learning approaches and ways of thinking, don't hesitate to apply.

Lectures will be run by a diverse group of physicists and computer scientists, led by the world-class researchers Tilman Plehn (Univerity of Heidelberg) and Thea Aarrestad (ETH Zurich), on topics such as:

- Bayesian neural networks
- Uncertainty quantification & learning
- Generative machine learning
- Representation learning
- Fast machine learning
- Physics-informed neural networks

Collecting, processing and analyzing data are central activities for virtually every researcher. Topics like data sharing and data publication are becoming increasingly important. Nevertheless, many research projects lack a structured and well-organized data management. This course is meant to give a general, discipline-independent introduction into various topics central to an efficient management of research data with a special focus on questions related to data archiving and data sharing. Both are central aspects of good scientific practice. Archiving and long-term preservation of research data are prerequisites for the scrutiny of scientific results based on the analysis of this data. Data sharing on the other hand increases transparency of research results and enables possible re-usage of data for new research questions, in combination with additional data sets and in interdisciplinary contexts.

In particular, the course will cover the following topics:

Requirements on research data handling from universities, research funders and scientific journals
Short-term and long-term preservation: formats, metadata, documentation, standards
Open Research data, data publication and data citations: Where? How? Why and why not?
Creation of data management plans for research projects

Support for researchers at Heidelberg University: the Competence Centre for Research Data (http://www.data.uni-heidelberg.de/index.en.html)

Prerequisites:

Experience or interest in publishing your Python code and a laptop is required.

Summary:

In this course we will learn how to package a Python library, how to publish it on PyPI and on conda-forge, as well as look at more advanced topics like building pre-compiled wheels including c++ extensions using pybind11, and automatically publishing new releases using continuous integration and cibuildwheel.
Learning Objectives

After the course participants should be able to:

- Create a modern pyproject.toml Python package
- Publish this package to PyPI
- Set up continuous integration to automatically publish to PyPI
- Understand the basics of conda-forge publishing
- Create binary wheels including c++ pybind11 extensions