Describe the bug
Not updating linear predictors in Theme

In the loop for(t in seq(1, length(out))){ ...} line 373, did not work because "out" is a list of length 1 containing a list of themes. Then length(out) is always equal to one.

solutions :
a/ change length(out) by sum(H>0)
or
b/ modify out object itself

Include in SCGLR work done by Jocelyn Chauvet on mixedSCGLR.

Describe the bug
demo files have to be updated

Expected behavior
Missing A = for additional geology covariates in formula (cf scglr, scglrCrossVal demo functions)
current formula as form <- multivariateFormula(ny,c(nx,"I(latlon)"),c("geology")) must be replaced by
form <- multivariateFormula(ny,c(nx,"I(latlon)"),A=c("geology"))

• setting title to FALSE should not display it

Describe the bug
Linear predictors not well evaluated

replace line from 366 to 378 by

 # compute global linear predictors
  comp <- cbind(1,as.matrix(do.call("cbind",lapply(out$themes, function(x) x$comp))))
  if(additional) {
    comp <- cbind(comp, model.matrix(theme_A, data)[,-1])
  }
  # lin.pred.global <- multivariatePredictGlm(comp, family=family, beta=gamma, offset=offset)
  lin.pred.global <- comp%*%gamma
  # Linear predictors by theme
  for(t in seq_along(out$themes)) {
    # par <- out$themes[[t]]$gamma[1:(H[t]+1),]
    par <- out$themes[[t]]$gamma[2:(H[t]+1),]#without intercept
    # out$themes[[t]]$lin.pred <- multivariatePredictGlm(cbind(1,as.matrix(out$themes[[t]]$comp)), family=family, beta=par, offset=offset)
    out$themes[[t]]$lin.pred <- as.matrix(out$themes[[t]]$comp)%*%par
    out$themes[[t]]$gamma <- gamma
    out$themes[[t]]$beta <- beta
  }

library(SCGLR) version 3.0

data(genus)
# depuis l'exemple scglrTheme
n <- names(genus)
n <-n[!n%in%c("geology","surface","lon","lat","forest","altitude")]
ny <- n[grep("^gen",n)]    # Y <- names that begins with "gen"
nx1 <- n[grep("^evi",n)]   # X <- remaining names
nx2 <- n[-c(grep("^evi",n),grep("^gen",n))]

keep first Y

form <- multivariateFormula(ny[1],nx1,nx2,A=c("geology"))
fam <- rep("poisson",1)#longueur 1
testthm <-scglrTheme(form,data=genus,H=c(2,2),family=fam,offset = genus$surface)

Error

Error in gamma[2:(sum(H) + 1), , drop = FALSE] : 
  incorrect number of dimensions

Describe the bug
Remove unneeded problematic dependency

Dependency expm has issues that are not corrected yet and may lead to archival of SCGLR on CRAN. Fortunately it appears that it is not used anymore in the package and can be safely removed.

Is your feature request related to a problem? Please describe.
Currently scglrCrossVal and themeBackward did not allow spatial cross validation.

Describe the solution you'd like
Provide clusters as a vector of sample size in folds parameter.

Hello
I am trying to use SCGLR package for a multivariate analysis that includes 3 dependent variables, 1 categorical dependent variable (burnt and unburnt), and two additional covariates (temperature and humidity), but I get this error when trying to complete the script:
Error in apply(x, 2, mean) : dim(X) must have a positive length

Can anyone tell me what is causing this?

I present my script (and its results) below:

build multivariate formula

ny <- c("taxa_pulsos","duracao","energia")
nx1 <- c("fogo")
nadd <- c("temperatura","umidade")
form <- multivariateFormula(ny,nx1,A=nadd)
form
Multivariate formula
taxa_pulsos + duracao + energia ~ fogo | temperatura + umidade
Response:
Y = taxa_pulsos + duracao + energia
Covariates:
T1 = fogo
A = temperatura + umidade

summary(form)
Length1 Length2 Class1 Class2
1 2 MultivariateFormula Formula
Class3 Mode
formula call

define family

fam <- rep("poisson",length(ny))
genus.scglr <- scglr(formula=form,,data = dados,family=fam)
Error in apply(x, 2, mean) : dim(X) must have a positive length

Contrasts error generated when providing factor variable as Additional

Error in `contrasts<-`(`*tmp*`, value = contr.funs[1 + isOF[nn]]) : 
  contrasts can be applied only to factors with 2 or more levels

The plsdepot package that is used by mixed SCGLR is no longer available on CRAN.

We should provide an alternate method for component initialization.

Rewrite code parallelization to use future backend

future, future.apply or maybe furrr

Error message should be completed with iteration number and/or fold id in case of cross validation.

When in cross validation, errors should only invalidate current fold but allow others to complete (eg when asked to do 10 folds, returning 9 is maybe enough)

Describe the bug

scglrPrediction is phased out

DELETED

• One function returns beta as matrix the other as a data.frame
• One has Y names as column the other V1