search

lucpaoli / SAFT_ML

2 stars 0 forks source link

Add temperature dependence to neural network, incorporating Gibbs-Duhem into the loss #5

Open lucpaoli opened 11 months ago

lucpaoli commented 11 months ago

relevant paper: https://arxiv.org/abs/2306.07937 though as we aim to generate date from GERG, their decision to create synthetic data using just Gibbs-Duhem loss is likely not relevant

MichaelGadaloff commented 10 months ago

Is Gibbs-Duhem not incorporated into SAFT? (i.e. not sure about benefit of incorporating into loss)

lucpaoli commented 10 months ago

Yeah, I think this is a silly idea