This is a practical course designed for chemical engineers that want to learn the basics of AI. The course is very practical and will cover various topics, such as basic Python syntax, solving ODEs, data visualization, machine learning and deep learning models, natural language processing, image processing and explainability.
This course covers the following topics: (i) basics of the Python syntax, data pre-processing and analysis, (ii) fundamentals in machine learning, including supervised and unsupervised learning, classification and regression techniques, (iii) deep-dives into various sub-fields of artificial intelligence such as natural language processing (NLP), image processing, graph neural networks (GNN) and how they can be applied in chemical engineering, and (iv) explainable artificial intelligence. In this course, several algorithms are introduced and discussed through practical coding tutorials, in order to provide students with a solid knowledge on how to implement various models in different frameworks, such as scikit-learn, PyTorch and TensorFlow. The course is composed by 7 weeks and is designed to be comprehensive and self-paced.
Python is one of the most popular languages in use today, thanks to its easy syntax, readability, libraries available and broad documentation. It is open-source, unlike other very popular programs such as MATLAB, which require a license. It has a big community, meaning that other people have probably already asked any possible question you might be having right now. Some forums where to ask and look for questions: StackOverflow (for practical questions about code), StackExchange (for more theoretical questions). It is the to-go language for data analysis and data science, both in academia and industry.
We have created other Python courses tailored for Chemical Engineers, such as Chemical Reaction Engineering in Python.
If you have any questions regarding the exercises or any feedback on the course, feel free to contact Fiammetta Caccavale (fiacac@kt.dtu.dk).
Please fill in our online survey if you have any feedback. We really appreciate your suggestions and we will try to use them to improve the course.
If you would like to cite this work, please reference our paper: SPyCE: A structured and tailored series of Python courses for (bio)chemical engineers as:
@article{CACCAVALE202390,
title = {SPyCE: A structured and tailored series of Python courses for (bio)chemical engineers},
journal = {Education for Chemical Engineers},
volume = {45},
pages = {90-103},
year = {2023},
issn = {1749-7728},
doi = {https://doi.org/10.1016/j.ece.2023.08.003},
url = {https://www.sciencedirect.com/science/article/pii/S1749772823000404},
author = {Fiammetta Caccavale and Carina L. Gargalo and Krist V. Gernaey and Ulrich Krühne},
keywords = {Python in engineering education, Digital education, Programming in engineering curriculum, Artificial Intelligence},
abstract = {In times of educational disruption, significant advances in adopting digitalization strategies have been accelerated. In this transformation climate, engineers should be adequately educated to face the challenges and acquire the new skills imposed by Industry 4.0. Among these, one of the most highly requested tools is Python. To tackle these aspects, this work establishes a pedagogical framework to teach Python to chemical engineers. This is achieved through a hands-on series of Python courses (sPyCE), covering topics as chemical reaction engineering and machine learning. Part of the series has been embedded in the curriculum of a Bachelor’s-level course at the Technical University of Denmark (DTU). Overall, students found the course to be useful; using Python, they solved systems of differential equations, mass and energy balances, set stoichiometric tables, regressions, simulations and more. Motivated by the large applicability and relevance of the covered topics, sPyCE is made publicly available on GitHub.}
}