Error in BIOMOD_Modeling - [MaxEnt error in file]

[jiayuanxie]

Hello, the maintenance team of Biomod2：

I'm a ecological doctor. recently, I was working with biomod2 to finish my project, but meet some troubles which beyond my ability to solve. So can you help me?

Error and context
When I run Maxent in biomod2, I meet the error, although I have tried many solutions which other people have solved their similar problems, but they didn't work with my code. I have tried to change the file path of maxent.jar to without any spaces or change the system environmental variables(followed: http://modata.ceoe.udel.edu/dev/dhaulsee/class_rcode/r_pkgmanuals/MAXENT4R_directions.pdf), but they don't help me for my trouble.

Code used to get the error
#######################Format sdm data for biomod2
Sp.biomod.data = BIOMOD_FormatingData(resp.name = Resp.name,
resp.var = Resp.vector,
expl.var = Sdm.env,
resp.xy = Resp.coor,
na.rm = T,
dir.name = paste("D:/Rtry/SDM",Resp.name,sep = "/"))

#######################Defining Models Options using default options
sdm.opt.default = BIOMOD_ModelingOptions( MAXENT = list( path_to_maxent.jar = "C://Users//Public//maxent.jar"))

####################Constrcut sdm models by setting parameters
Bio.mod.all = BIOMOD_Modeling( bm.format = Sp.biomod.data,
models = "MAXENT",
models.pa = Sp.biomod.data@data.species,
bm.options = sdm.opt.default,
CV.strategy = 'random',
#Number of Evaluation run
CV.nb.rep = 1,
#80% of data train the model
CV.perc = 0.8,
#models be calibrated and validated over the whole dataset
CV.do.full.models = F,
#Below arg is whether control the weight?
weights=NULL,
#absences will be weighted equally to the presences
prevalence = 0.5,
#Set model evaluations methods
metric.eval = mode.eva,
#Number of permutation to estimate variable importance
var.import = 1,
nb.cpu =1,
do.progress =T)

#show errors
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= Build Single Models -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=

Checking Models arguments...
Warning in .BIOMOD_Modeling.check.args(bm.format = bm.format, modeling.id = modeling.id, :
models.pa has been disabled because no PA datasets have been given

 MAXENT default prevalence option was updated to fit with modeling prevalence (i.e 0.5 )
> Automatic weights creation to rise a 0.5 prevalence

Creating suitable Workdir...

Checking Cross-Validation arguments...

Random cross-validation selection

-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= Bandicota.bengalensis.reg.1 Modeling Summary -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=

8 environmental variables ( PCA_1 PCA_2 PCA_3 PCA_4 PCA_5 PCA_6 PCA_7 PCA_8 )
Number of evaluation repetitions : 1
Models selected : MAXENT

Total number of model runs: 1

-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=

-=-=-=--=-=-=- Bandicota.bengalensis.reg.1_allData_RUN1_MAXENT

Model=MAXENT
> MAXENT modeling...
Creating Maxent Temp Proj Data...
Getting predictions...Error in file(file, "rt") : cannot open the connection
In addition: Warning message:
In file(file, "rt") :
cannot open file 'D:/Rtry/SDM/Bandicota.bengalensis.reg.1/Bandicota.bengalensis.reg.1/models/1696750966/Bandicota.bengalensis.reg.1_allData_RUN1_MAXENT_outputs/Bandicota.bengalensis.reg.1_allData_RUN1_Pred_swd.csv': No such file or directory

Getting predictor contributions...Error in file(fname, "r") : cannot open the connection
In addition: Warning messages:
1: In system(command = maxent.command, wait = TRUE, intern = TRUE) :
running command 'java -mx512m -cp "C:\Users\Public/maxent.jar" density.Project "" "D:/Rtry/SDM/Bandicota.bengalensis.reg.1/Bandicota.bengalensis.reg.1/models/1696750966/Bandicota.bengalensis.reg.1_allData_RUN1_MAXENT_outputs/Predictions/Pred_swdBis_5125.csv" "D:/Rtry/SDM/Bandicota.bengalensis.reg.1/Bandicota.bengalensis.reg.1/models/1696750966/Bandicota.bengalensis.reg.1_allData_RUN1_MAXENT_outputs/projMaxent.asc" doclamp=false visible=false autorun nowarnings notooltips' had status 1
2: In file(fname, "r") :
cannot open file 'D:/Rtry/SDM/Bandicota.bengalensis.reg.1/Bandicota.bengalensis.reg.1/models/1696750966/Bandicota.bengalensis.reg.1_allData_RUN1_MAXENT_outputs/projMaxent.asc': No such file or directory
Error in { : task 1 failed - "Unable to make model prediction"

#show(Bio.mod.all)

show(Bio.mod.all)
Error in h(simpleError(msg, call)) : There was an error evaluating the 'object' parameter when selecting a method for the 'show' function: object 'Bio.mod.all' not found

Inaddition, my computer system is windows 10; R version is R-4.3.1; Java version is "21" 2023-09-19 LTS; Biomod2 package version is 4.2-4

To now, I really have tried many methods to solve my problem, but them didn't work, If you can help me , I will be great grateful!!!

Thanks , from JY Xie

[MayaGueguen]

Hello JY Xie,

Thank you for reporting 🙏

Could you try changing your BIOMOD_FormatingData command by this please ?

Sp.biomod.data = BIOMOD_FormatingData(resp.name = Resp.name,
                                      resp.var = Resp.vector,
                                      expl.var = Sdm.env,
                                      resp.xy = Resp.coor,
                                      na.rm = TRUE,
                                      dir.name = "D:/Rtry/SDM/")

It seems biomod2 get a bit lost through file names.

Maya

[jiayuanxie]

Hello JY Xie,

Thank you for reporting 🙏

Could you try changing your BIOMOD_FormatingData command by this please ?

Sp.biomod.data = BIOMOD_FormatingData(resp.name = Resp.name,
                                      resp.var = Resp.vector,
                                      expl.var = Sdm.env,
                                      resp.xy = Resp.coor,
                                      na.rm = TRUE,
                                      dir.name = "D:/Rtry/SDM/")

It seems biomod2 get a bit lost through file names.

Maya

Maya,

Thank you for your prompt reply！

According to your advising, I just have tried the method, but it seem to not work with the same error above, do you have any other suggestion? Indeed, I have tried on two computer, so it should not be responsible for computer or file name.

JY XIE

#######################Format sdm data for biomod2
Sp.biomod.data = BIOMOD_FormatingData(resp.name = Resp.name,
resp.var = Resp.vector,
expl.var = Sdm.env,
resp.xy = Resp.coor,
na.rm = T,
dir.name = "D:/Rtry/SDM/")

#######################Defining Models Options using default options
sdm.opt.default = BIOMOD_ModelingOptions( MAXENT = list( path_to_maxent.jar = "D:/Rtry/SDM/maxent.jar"))

####################Constrcut sdm models by setting parameters
Bio.mod.all = BIOMOD_Modeling( bm.format = Sp.biomod.data,
models = "MAXENT",
models.pa = Sp.biomod.data@data.species,
bm.options = sdm.opt.default,
CV.strategy = 'random',
#Number of Evaluation run
CV.nb.rep = 1,
#80% of data train the model
CV.perc = 0.8,
#models be calibrated and validated over the whole dataset
CV.do.full.models = F,
#Below arg is whether control the weight?
weights=NULL,
#absences will be weighted equally to the presences
prevalence = 0.5,
#Set model evaluations methods
metric.eval = mode.eva,
#Number of permutation to estimate variable importance
var.import = 1,
nb.cpu =1,
do.progress =T)

[MayaGueguen]

Thank you for trying 🙏

Could you try and check into your files browser if the files mentioned really exist or not ?

[jiayuanxie]

the files "D:/Rtry/SDM/maxent.jar" & "D:/Rtry/SDM/" both exist,

In addition, It's interesting, my error was:
Getting predictions...Error in file(file, "rt") : cannot open the connection
In addition: Warning messages:
1: executing %dopar% sequentially: no parallel backend registered
2: In file(file, "rt") :
cannot open file 'D:/Rtry/SDM//Bandicota.bengalensis.mach.1/models/1696861384/Bandicota.bengalensis.mach.1_allData_RUN1_MAXENT_outputs/Bandicota.bengalensis.mach.1_allData_RUN1_Pred_swd.csv': No such file or directory

but there exist a file named "Pred_swd.csv" in the path（my species name is "Bandicota.bengalensis.mach.1"）: D:\Rtry\SDM\Bandicota.bengalensis.mach.1\models\1696860853\Bandicota.bengalensis.mach.1_allData_RUN1_MAXENT_outputs\Predictions\Pred_swd.csv. I guess the file bring the bug to me. it make me fell confused, the file is generated by my codes, why they can't found it?

[MayaGueguen]

Huuum it is not the same path in the error message :

D:/Rtry/SDM//Bandicota.bengalensis.mach.1/models/1696861384/Bandicota.bengalensis.mach.1_allData_RUN1_MAXENT_outputs/Bandicota.bengalensis.mach.1_allData_RUN1_Pred_swd.csv

and in your folder :

D:\Rtry\SDM\Bandicota.bengalensis.mach.1\models\1696860853\Bandicota.bengalensis.mach.1_allData_RUN1_MAXENT_outputs\Predictions\Pred_swd.csv

The numbers do not match. You can give a modeling.id parameter value to the BIOMOD_Modeling function to have more explicit path.
Try starting again from a blank repository.

[jiayuanxie]

Thanks for your reminding!
Yeah, it's not the same path in the error message due to I have tried many times, So created many similar folders, but their hierarchical structures and error are same!

Try starting again from a blank repository seem have no impact on my error, so , it's really a headache problem！