Reproducibility in materials informatics: lessons from ‘A general-purpose machine learning framework for predicting properties of inorganic materials’
Daniel Persaud, Logan Ward, Jason Hattrick‐Simpers
Abstract
Reproducing results from a foundational materials informatics tool (magpie) is difficult and in this study, a failure. This failure yields tangible suggestions to promote easy adoption and trust of materials informatics in the future.
Topics & Concepts
InformaticsMaterials informaticsComputer scienceHealth informaticsData scienceEngineeringEngineering informaticsMedicineElectrical engineeringPublic healthNursingMachine Learning in Materials ScienceComputational Drug Discovery MethodsAdvanced X-ray and CT Imaging