Model parameterisation, testing and variations from Coupling machine learning and epidemiological modelling to characterize optimal fungicide doses when fungicide resistance is partial or quantitative

We present parameter values for the epidemiological model as well as the optimal hyperparameters for the gradient-boosted trees model. We evaluate the gradient-boosted trees model performance on both training and test data. We also present model variations which allow us to explore: the effect of model parameters within a single year only; the results in terms of cumulative yield over time; the effect of changing partial resistance parameterisation.