This is the source code related to the publication
Maike Sonnewald and Redouane Lguensat, 2021. Revealing the impact of global heating on North Atlantic circulation using transparent machine learning. Paper: https://doi.org/10.1029/2021MS002496
Companion paper using a probabilistic framework:
Mariana Clare, Maike Sonnewald and Redouane Lguensat, Julie Deshayes and V. Balaji, 2022 Explainable Artificial Intelligence for Bayesian Neural Networks: Towards trustworthy predictions of ocean dynamics. Paper: https://doi.org/10.1029/2022MS003162
This repository contains the codes used for accessing the data and the weights of the ensemble MLP used for classifying ocean regimes
Go to THOR folder, where you will find notebooks on the individual steps of THOR.
If you are only interested in applying the already trained EnsembleMLP of THOR, go to the folder THOR/ApplicationOnCMIPModels to find an example of the application of THOR on the IPSL-CM6-LR model
For this paper we used the CMIP6 data hosted on AWS, but you can use your preffered source of CMIP6 models. Researchers wishing to use the CMIP6 data on AWS can also apply for cloud grants through the AWS Promotional Credit Program. Please check these links:
Other data for the training of THOR is present in each step-specific folder.
Do not hesitate to send us an email if you have any question !
Maike Sonnewald: maikes "at" princeton.edu
Redouane Lguensat: redouane.lguensat "at" locean.ipsl.fr