Dear BIOMODER, <br><br>I had a look on the Biomod.Models() function and I was wondering why interaction depth was set to 7 in the GBM function. <br><br>gbm(eval(parse(text = paste(SpNames[i], <br> paste(scopeExpSyst(DataBIOMOD[1:10, 1:NbVar], <br>
"GBM"), collapse = "")))), data = DataBIOMOD[calib.lines, <br> ], distribution = "bernoulli", var.monotone = rep(0, <br> length = NbVar), w = Yweights[calib.lines, <br>
i], <b>interaction.depth = 7</b>, shrinkage = 0.001, <br> bag.fraction = 0.5, train.fraction = 1, verbose = F, <br> cv.folds = 5)<br><br><br><br>Interaction depth is the maximum depth of variable interactions. 1 implies an additive model, 2 implies a model with up to 2-way interactions, etc... I was wondering if it was because 7 variables were used in the tutorial but in this case, wouldn't be better to set the interaction depth to the number of variables (NbVar) ?<br>
<br>However I was wondering the effect of interaction depth on overfitting. <br><br>Cheers, <br>Blaise Petitpierre<br>