Repository for Characterizing Marine Magnetic Anomalies

 

This repository contains the data and code used in our study. It is organized into three main folders: data, matlab, and notebook. These resources collectively enable the training and application of generation and characterization networks for magnetic field data analysis, as well as the processing of field magnetic data from the EMAG2v3 database.

 

 

 

This folder contains:

• CNNTrainingData & CVAETrainingData: Includes all the training datasets used for the generation and characterization networks.

• OriginalMagData: Contains the magnetic field data extracted from the EMAG2v3 database in .xyz format.

 

 

This folder contains all the MATLAB scripts utilized in this study.

• FieldDataSplit: A key script that splits the original .xyz magnetic data into smaller windows, preparing the data for further predictions.

 

 

This folder contains:

• Jupyter Notebooks: These notebooks include the training and prediction workflows for the generation and characterization networks.

• Trained Networks: The well-trained networks are saved in the NetSave subfolder, ready for use in prediction or further evaluation.