Thanks, glad it was helpful!

]]>Hi Beto,

The facets in lekprofile are created with the facet_grid() function from ggplot2. This function forces a fixed grid based on the y ~ x inputs. You can cheat by using the xsel argument from lekprofile, then patch the individual plots together with patchwork.

library(NeuralNetTools) library(nnet) library(patchwork) mod <- nnet(Y1 ~ X1 + X2 + X3, data = neuraldat, size = 5) p1 <- lekprofile(mod, xsel = c('X1', 'X2')) p2 <- lekprofile(mod, xsel = 'X2') p1 + p2 + plot_layout(ncol = 1, guides = 'collect')

Hope that helps.

-Marcus

]]>Thanks for your great work on sensitivity analysis and I find it really helpful.

I am actually new to R and programming , i have applied the lek profile function and i have a very huge plot, i have tried to arrange the plots , with par(mfrow) but haven’t succeeded .i tried using the code below. but honestly i wasnt too sure of what i was doing. please i will be happy if you can help out, Thanks

p1<-lek.fun(mod1)

class(p1)

p1 +

par(mfrow=c(3, 4))

Hi Anita, glad you were able to use the function. You can use this reference for the VIF method:

Zuur, A.F., Ieno, E.N. Smith, G.M. Analysing Ecological Data. Springer, New York.

Montgomery, D.C., Peck, E.A. 1992. Introduction to Linear Regression Analysis. Wiley, New York.

Thank you ]]>

I got this error message and stumbled on a solution. My model name giving the problem was net.alloy4. I re-ran the model and named it net.alloya; the compute command worked perfectly.

Don’t know if it will work for you, but it did so for me

]]>There might be a distinction between majors, you’d have to check out the UMN web page to find out. If you want to parse the raw code you can access it here.

]]>