Error message while using "humboldt" package.

General
1

Hey! I am currently using a package called humboldt (GitHub - jasonleebrown/humboldt: Welcome to the offical webpage for the R-package ‘humboldt’) in order to reduce the the climate factors, that affect some animal species. I wrote the following:

env1<-read.delim("Env1.csv",h=T,sep=",")
env2 <-read.delim("Env1.csv",h=T,sep=",")
sp1<- na.exclude(read.delim("Occ_Data_Total.csv",h=T,sep=","))
sp2<- na.exclude(read.delim("Occ_Data_Total.csv",h=T,sep=","))
reduc.vars<- humboldt.top.env(env1=env1,env2=env2,sp1=sp1,sp2=sp2,rarefy.dist=50, rarefy.units="km", env.reso=0.416669,learning.rt1=0.01,learning.rt2=0.01,e.var=(4:17),pa.ratio=4,steps1=50,steps2=50,method="contrib",contrib.greater=5)

When I try to run the last command I end up with the following error message:
Error in gbm.fit(x = x, y = y, offset = offset, distribution = distribution, :
The data set is too small or the subsampling rate is too large: nTrain * bag.fraction <= n.minobsinnode

If any of you guys can help i'd bre grateful!!!

2

Hi @Nickmofoe,
Welcome to the RStudio Community Forum.

Although I have no knowledge of the humboldt package, I assume Env1 and Env2 should contain different data, yet you are reading the same file into both objects. Same applies to sp1 and sp2.

3

Hey! So with the sp1 and sp2 I made a typo it is different data sets. Regarding env1 and env2 they're supposed to be the same (the package allows it)

4

It seems you will either need to get more data or reduce the subsampling rate.

5

Any clue how I can reduce the subsampling rate?

6

This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.

If you have a query related to it or one of the replies, start a new topic and refer back with a link.