CvFit () cross validation error for nls model: "Error: object of type" character "is not a subset of"

Question

I have this model nls()that works pretty smoothly and I want to be able to cross-check with the package cvTools, but I get this error when the function starts cvFit():

Error: an object of type 'symbol' is not a subset

RStudio automatically opens the debugger, and I get this in the source code viewer:

function (x, env = parent.frame(), ...) 
{
    notAtomic <- !is.atomic(x)
    notnull <- function(z) notAtomic && !is.null(z)
    if (notnull(x$formula)) #this part is hightlited in the source viewer for some reason
        eval(x$formula)
    else if (notnull(x$terms)) {
        z <- x$terms
        oldClass(z) <- "formula"
        z
    }
    else if (notnull(x$call$formula)) 
        eval(x$call$formula)
    else if (!is.null(attr(x, "formula"))) 
        attr(x, "formula")
    else {
        form <- switch(mode(x), `NULL` = structure(NULL, class = "formula"), 
            character = formula(eval(parse(text = x)[[1L]])), 
            call = eval(x), stop("invalid formula"))
        environment(form) <- env
        form
    }
}

I have included an excerpt of my frame in the following link in .csv format: https://www.sendspace.com/file/ec1aoa

And here is the code I used:

fctmod<-function(x,y,z){
  resultat<-(1+x*((z-mean(z))/mean(z))+y*((z-mean(z))/mean(z))^2)
  return(resultat)
}

fctmodlin<-function(x,z){
  resultat<-(1+x*((z-mean(z))/mean(z)))
  return(resultat)
}

nlscut<- nls(BAI ~ (a*(B_A)^(b))*
                         fctmodlin(c,PPT.an)*
                         fctmod(e,f,maxTmax.an)*
                         fctmod(g,h,GPautre),
                       start = list(a = 1.14338, b = 0.504549,c=0, e=0, f=0, g=0, h=0),
                       data = datacut,
                       weights= wfct(DHPc))

library(cvTools)
testcut<-cvFit(nlscut, data=datacut,y=datacut$BAI,cost=mspe,K=10,R=1,seed=1234)