Albert Steppi
Albert Steppi (@steppi) is a Senior Software Engineer at Quansight Labs and a maintainer of the SciPy library. Previously he worked as a Machine Learning Scientist at Lendbuzz and before that as a Scientific Software Developer in the Labroratory of Systems Pharmacology at Harvard Medical School. He earned a PhD in Statistics from Florida State University in 2018 with research focusing on network aware bioinformatics analysis and biomedical text mining. He is broadly interested in numerical mathematics and scientific and statistical computing.
Sessions
The SciPy library provides objects representing well over 100 univariate probability distributions. These have served the scientific Python ecosystem for decades, but they are built upon an infrastructure that has not kept up with the demands of today’s users. To address its shortcomings, SciPy 1.15 includes a new infrastructure for working with probability distributions. This talk will introduce users to the new infrastructure and demonstrate its many advantages in terms of usability, flexibility, accuracy, and performance.
SciPy provides a number of useful mathematical special functions within its scipy.special module. This talk will discuss developments over the past year to translate and move scipy.special's underlying compiled code into a separate library called xsf which, when practical, works in CUDA. We will discuss the impact of this project has had and will have for improving special function support in GPU-aware array libraries such as CuPy and PyTorch, and suggest ways that interested contributors can get involved.