Reproducible Machine Learning Workflows for Scientists

Scientific researchers need reproducible software environments for complex applications that can run across heterogeneous computing platforms. Modern open source tools, like Pixi, provide automatic reproducibility solutions for all dependencies while providing a high level interface well suited for researchers.

This in-person workshop will provide a practical introduction to using Pixi to easily create computing environments for scientific and AI/ML workflows that benefit from hardware acceleration, across multiple machines and platforms. The focus will be on applications using Python machine learning libraries with CUDA enabled, as well as deploying these environments to production settings in Linux container images. This workshop will not teach machine learning concepts, but will focus on the methodologies and tools to make existing machine learning workflows reproducible.

This workshop was supported by the US Research Software Sustainability Institute (URSSI) via grant G-2022-19347 from the Sloan Foundation, NVIDIA, and the University of Wisconsin–Madison Data Science Institute.