Today's article comes from Elsevier's SoftwareX journal. The authors are Garcia-Moreno et al., from the University of Granada, in Spain. In this paper they attempt to train a neural-network that can help historians restore fine art pieces. But, let's just say that their analysis leaves something to be desired. First we're going to look at what they built. But then, we'll take some time to examine why I believe their model validation process is problematic and flawed. In the end, we'll just see what we can learn from their mistakes.
DOI: 10.1016/j.softx.2024.101917
You must be an active Journal Club member to access this content. If you're already a member, click the blue button to login. If you're not a member yet, click the sign-up button to get started.
Login to My Account
Sign Up