Closed stevehadd closed 4 years ago
I'm happy that the notebook is now in place to reproduce the plots in the paper. There seems to be a discrepancy in the number of labelled (model and manufacturer) profiles in the dataset I am working with, compared to the dataset in the paper. This needs to investigated further, so I have created a new issue (#20), and am closing this one.
The Pallmer et. al paper describing the metadata infill algorithm iMeta: https://journals.ametsoc.org/doi/full/10.1175/JTECH-D-17-0129.1
includes several plots showing distributions of data and metadata for various probe types. We should aim to recreate these plots to check comparability.