Research Compendium for Himes et al. (2024): "Using neural networks for near-real-time aerosol retrievals from OMPS Limb Profiler measurements"

This archive is the Reproducible Research Compendium for

Using neural networks for near-real-time aerosol retrievals from OMPS Limb Profiler measurements

by Himes et al. (2024), submitted to Atmospheric Measurement Techniques.

This compendium includes all files related to MARGE associated with the manuscript.

NN model files are split into smaller files for convenience, given their sizes.  To recombine the files, do, e.g.,     cat cnn_weights_NH-LW.h5* > cnn_weights_NH-LW.h5

User interested in running MARGE will need to clone the GitHub repo (https://github.com/exosports/MARGE), apply the patch file to checksum fc95b3c, organize the relevant files into directories as listed in the configuration files (Zenodo does not support organizing files into directory structures) and calculate the number of training, validation, and test cases to be stored in the relevant input file specified in the configuration file.  MARGE is under the Reproducible Research Software License (https://planets.ucf.edu/resources/reproducible-research/software-license/).  For more details on MARGE, see the User Manual on GitHub.