Error in VariableSelection() with Maxent: "character string expected"

54 views
Skip to first unread message

Wondessen Ayalew

unread,
Jan 15, 2025, 5:51:33 PMJan 15
to Maxent

Hi folks,

I am using Maxent in R to model species distribution for Ethiopian sheep. I encountered the following error during the VariableSelection() step:

Error in system(command, as.integer(flag), f, stdout, stderr, timeout,  :

 character string expected as the first argument

Here’s my setup:

Operating System: Windows 10

Maxent Version: 3.4.1

R Version: 4.3.1

Java Version: 1.8.0_291

Relevant Packages: dismo

My R script:

 

maxent <- "D:/ENM_Analysis/maxent/maxent/maxent.jar"

gridfolder <- dir(path="D:/ENM_Analysis/maxent/maxent", pattern='asc', full.names=TRUE)

outdir <- "D:/ENM_Analysis/maxent/maxent/outdir"

occurrencelocations <- "D:/ENM_Analysis/maxent/maxent/VariablesAtOccurrencelocations.csv"

backgroundlocations <- "D:/ENM_Analysis/maxent/maxent/VariablesAtBackgroundlocations.csv"

additionalargs <- "nolinear noquadratic noproduct nothreshold noautofeature"

betamultiplier <- seq(0.5, 15, 0.2)

 VariableSelection(

  maxent,

  outdir,

  gridfolder,

  occurrencelocations,

  backgroundlocations,

  additionalargs,

  contributionthreshold = 5,

  correlationthreshold = 0.9,

  betamultiplier

)

I’ve verified the paths and that the .asc files exist in the maxent/maxent folder. What might be causing this error, and how can I resolve it?

Thanks in advance for your help!

 

Best regards,

Wondossen

Bede-Fazekas Ákos

unread,
Jan 16, 2025, 12:23:05 AMJan 16
to max...@googlegroups.com
Hi Wondossen,
Try
gridfolder <- "D:/ENM_Analysis/maxent/maxent"
Hope this helps,
Ákos
_____________
Ákos Bede-Fazekas
Centre for Ecological Research, Hungary
--
You received this message because you are subscribed to the Google Groups "Maxent" group.
To unsubscribe from this group and stop receiving emails from it, send an email to maxent+un...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/maxent/33ef86cd-0016-4e51-9a41-73475ba1e4d3n%40googlegroups.com.

Wondessen Ayalew

unread,
Jan 16, 2025, 8:52:23 AMJan 16
to Maxent

Dear Akos and all,

Thank you for your valuable suggestions.

I am encountering an error despite following your suggestion. Here is the error message:

 

maxent <- "D:/ENM_Analysis/maxent/maxent/maxent.jar"

> occurrencelocations<- ("D:/ENM_Analysis/maxent/maxent/VariablesAtOccurrencelocationsEthiopianCattle.csv")

> backgroundlocations<- ("D:/ENM_Analysis/maxent/maxent/VariablesAtBackgroundlocationsEthiopianCattle.csv")

> gridfolder <- ("D:/ENM_Analysis/maxent/maxent")

> outdir<- ("D:/ENM_Analysis/maxent/maxent/outdir")

> additionalargs="nolinear noquadratic noproduct nothreshold noautofeature"

> #we set the threshold to 5%, which means that all variables will be excluded when they

> #contribute less than 5% to the model:

> contributionthreshold<- 5

> #Correlated variables should be removed Of the correlated

> #variables, only the variable with the highest contribution score will be kept

> correlationthreshold <- 0.9

> #This argument sets the values of beta multipliers (regularization multipliers) for which variable selection shall

> #be performed. The smaller this value, the more closely will the projected distribution fit to the training data

> #set. Models that are overfitted to the occurrence data are poorly transferable to novel environments and, thus,

> #not appropriate to project distribution changes under environmental change. Performance will be compared

> #between models created with the beta values given in this betamultiplier vector. Thus, providing a range

> #of beta values from 1 (the default in Maxent) to 15 or so, will help you to spot the optimal beta multiplier

> #for your specific model.

> #Here, we are testing betamultipliers from 2 to 6 with steps of 0.5 with the following setting:

> betamultiplier=seq(0.5,15,0.2)

> VariableSelection(maxent,

+                   outdir,

+                   gridfolder,

+                   occurrencelocations,

+                   backgroundlocations,

+                   additionalargs,

+                   contributionthreshold,

+                   correlationthreshold,

+                   betamultiplier)

----------------------- 

Choosing betamultiplier  0.5

 

 Number of remaining variables 35

Testing variable contributions...

Calculating average AUC values from 10 maxent models...

Error in file(file, "rt") : cannot open the connection

In addition: Warning messages:

1: In system(paste("java -jar", maxent, "-o", outdir, "-j", gridfolder,  :

  running command 'java -jar D:/ENM_Analysis/maxent/maxent/maxent.jar -o D:/ENM_Analysis/maxent/maxent/outdir -j D:/ENM_Analysis/maxent/maxent -s D:/ENM_Analysis/maxent/maxent/outdir/VariablesAtOccurrencelocationsEthiopianCattle.csv_VariableSubset.csv -e D:/ENM_Analysis/maxent/maxent/outdir/VariablesAtBackgroundlocationsEthiopianCattle.csv_VariableSubset.csv plots=false writeplotdata=false visible=false autorun=true randomseed=true writebackgroundpredictions=false replicates=10 replicatetype=subsample randomtestpoints=50 redoifexists writemess=false writeclampgrid=false askoverwrite=false pictures=false outputgrids=false -b 0.5 nolinear noquadratic noproduct nothreshold noautofeature' had status 1

2: In file(file, "rt") :

  cannot open file 'D:/ENM_Analysis/maxent/maxent/outdir/maxentResults.csv': No such file or directory

 

Thank you once again for your unwavering support.

Regrds,

Wondossen

Reply all
Reply to author
Forward
0 new messages