Model axes contribution evaluation: Error in if (stat != "") { : argument is of length zero

[henriqueantoniolli]

Dear Sylvain and contributors,

I am getting the error at the "Model axes contribution evaluation" step for all algorithms, except MAXENT. Here I bring an example for the first algorithm (GLM):

library(SSDM)
library(raster)
Environ <- load_var(path="layers/",files=NULL,format=".tif")
Occur <- load_occ(path="occur/",Environ,Xcol='lon',Ycol='lat',sep=",",dec=".",Spcol='sp')
SSDM <- ensemble_modelling(algorithms = "all",Occurrences = Occur,Env = Environ,
                                  Xcol='lon', Ycol='lat', Pcol=NULL,
                                  rep=1, tmp=T, cv="holdout",
                                  PA=NULL, cv.param = c(0.75, 10), axes.metric = "AUC",
                                  ensemble.metric = c("AUC"), ensemble.thresh = c(0.75))

#### Algorithms models creation for species ##### 2022-01-19 16:55:40

Modelling : GLM.1

Data check ...
No presence column, presence-only data set is supposed.
Pseudo-absence selection will be computed.
done.

Pseudo absence selection...
random selection
done.

Model evaluation...
done.

Model projection...
done.

Model axes contribution evaluation...
Error in if (stat != "") { : argument is of length zero
Error in if (stat != "") { : argument is of length zero
GLM.1 done for species 2022-01-19 16:55:44

I was not able to find help on this matter elsewhere.
I am running the Github version of SSDM (0.2.8.9002), on R version 4.0.0, Ubuntu 20.04.3 LTS.

Thanks
Henrique Antoniolli

[sylvainschmitt]

Could you make a reproducible example with the package data?