Python in Science Conference (SciPy)
Published July 10, 2024 | Version v1
Poster Open

Aeromancy: Towards More Reproducible AI and Machine Learning

Authors/Creators

  • 1. QuantAQ Inc.

Contributors

  • 1. GDI
  • 2. SLB
  • 3. University of North Carolina
  • 4. Curvenote
  • 5. Deloitte
  • 6. Aptos
  • 7. Arm

Description

We present Aeromancy, an opinionated philosophy and open-sourced framework that closely tracks experimental runtime environments for more reproducible machine learning. In existing experiment trackers, it's easy to miss important details about how an experiment was run, e.g., which version of a dataset was used as input or the exact versions of library dependencies. Missing these details can make replicability more difficult. Aeromancy aims to make this process smoother by providing both new infrastructure (a more comprehensive versioning scheme including both system runtimes and external datasets) and a corresponding set of best practices to ensure experiments are maximally trackable.

Files

scipy-2024-aeromancy.pdf

Files (410.1 kB)

Name Size Download all
md5:76674cc4fcdde588a98418d2010cbcda
410.1 kB Preview Download