This repo contains the EIS dataset and trained DNN models that are described in paper titled "Serialization of electrochemical impedance spectra to forecast low frequency impedances via deep learning neural networks".

There is a simple jupyter notebook example on how to load, normlize and split the dataset, as well as how to load the DNN models.

Further files will be added depend on the process of the paper.

Note: the models were created and trained on Ubuntu 20.04 system with tensorflow = 2.6, I have tried to run the models on Apple's macbook with apple silicon, but the prediction results are completely different for reasons unknown. Users must beware that it is preferablly to run the models on Linux or Windows system.