As recommended by Heinze et al. 2018 (https://doi.org/10.1002/bimj.201700067)

Dear problem solver,

I can't make the java command work. It started out as a problem with executing the class file
java program
, where I was told the program had been compiled by a more recent version (jdk 13.0.2, displayed as 57 if I remember correctly) than the java runtime environment (being jre.1.8.0-241). After having desperately tried to solve that problem - changing paths both in windows environment as well as Java's settings - I resorted to simply "downgrading" my jdk to the same version as the jre. Now javac still works fine, but as mentioned, now the "java" command can't even be found. I assume I've made some very simple mistake, but I just can't seem to solve it. Please help!

Dear @julienvollering,

We got this error message and have not been able to figure out what's going on.

code:
DVselect <- selectDVforEV(dvdata, alpha = 0.001, quiet = TRUE)

Error message:

Error in if (nrow(ctable) == 1 || ctable$P[1] > alpha) { : 
  missing value where TRUE/FALSE needed

We are willing to share the dvdata object if you provide your email.
Thanks,

From a user:

I'm not sure why I get the following error when I run the line:

grasslandDVs <- deriveVars(grasslandPO, transformtype = c("L", "M", "D", "HF", "HR", "T", "B"))

Alpha: returning -Infinity for pi[f]=0.000000+-0.000000, q[f]=0.000091
Error: goodAlpha: returning -Infinity for pi[f]=0.000000+-0.000000, q[f]=0.000091
Error in read.table(file = file, header = header, sep = sep, quote = quote, :
no lines available in input
In addition: Warning message:
running command 'java -mx512m -jar "C:/Users/mcilwea/Documents/R/R-3.4.3/library/MIAmaxent/java/maxent.jar" removeduplicates=FALSE addsamplestobackground=FALSE autofeature=FALSE betamultiplier=0 quadratic=FALSE product=FALSE hinge=FALSE threshold=FALSE outputformat=raw writebackgroundpredictions=TRUE outputgrids=FALSE pictures=FALSE extrapolate=FALSE writemess=FALSE plots=FALSE doclamp=FALSE writeclampgrid=FALSE autorun=TRUE threads=2 visible=FALSE warnings=FALSE maximumbackground=17480 samplesfile="R:/Koala/R/MAXENT/Test/deriveVars/pca1/HR/pca1_HR1/samples.csv" environmentallayers="R:/Koala/R/MAXENT/Test/deriveVars/pca1/HR/pca1_HR1/environlayers.csv" outputdirectory="R:/Koala/R/MAXENT/Test/deriveVars/pca1/HR/pca1_HR1/"' had status 1

When using one of the MIAmaxent functions that relies on the maxent.jar software (e.g. deriveVars, selectDVforEV, selectEV, plotResp), some users have reported the appearance of the following waring message:
"WARNING: Could not open/create prefs root node Software\JavaSoft\Prefs at root 0x80000002. Windows RegCreateKeyEx(...) returned error code 5."

Issue: Using Robot class i have copied content during run time BCDE and trying to retrieve it from clipboard to print again in console or any other file. But it's fetching old value from the clip board?
Expected Output is: BCDE

import java.awt.AWTException;
import java.awt.HeadlessException;
import java.awt.Robot;
import java.awt.Toolkit;
import java.awt.datatransfer.Clipboard;
import java.awt.datatransfer.DataFlavor;
import java.awt.datatransfer.StringSelection;
import java.awt.datatransfer.Transferable;
import java.awt.datatransfer.UnsupportedFlavorException;
import java.awt.event.InputEvent;
import java.awt.event.KeyEvent;
import java.io.IOException;

import org.sikuli.basics.Settings;
import org.sikuli.script.FindFailed;
import org.sikuli.script.Key;
import org.sikuli.script.Pattern;
import org.sikuli.script.Screen;

public class Sikuli_Demo {

public static void main(String[] args) throws InterruptedException, FindFailed, AWTException, HeadlessException, UnsupportedFlavorException, IOException {

	Screen screen=new Screen();
	Thread.sleep(2000);
	Pattern clickOutlook=new Pattern("images\\1_ClickOutlook.png");

	Pattern clickNewMail=new Pattern("images\\3_NewMail.png");

	screen.click(clickOutlook);
	Thread.sleep(2000);
	screen.click(clickNewMail);
	Thread.sleep(2000);
	screen.type("ABCDE");
	screen.type(Key.LEFT+Key.LEFT+Key.LEFT+Key.LEFT);
	Thread.sleep(2000);

	Robot robot=new Robot();
	Thread.sleep(4000);
	
	robot.keyPress(KeyEvent.VK_SHIFT);
	robot.keyPress(KeyEvent.VK_RIGHT);
	robot.keyPress(KeyEvent.VK_SHIFT);
	robot.keyPress(KeyEvent.VK_RIGHT);
	robot.keyPress(KeyEvent.VK_SHIFT);
	robot.keyPress(KeyEvent.VK_RIGHT);
	robot.keyPress(KeyEvent.VK_SHIFT);
	robot.keyPress(KeyEvent.VK_RIGHT);

	Thread.sleep(2000);
	robot.keyRelease(KeyEvent.VK_RIGHT);
	robot.keyRelease(KeyEvent.VK_SHIFT);
	Thread.sleep(2000);
	System.out.println("Check1");
	robot.keyPress(KeyEvent.VK_CONTROL);
	robot.keyPress(KeyEvent.VK_C);
	Thread.sleep(2000);
	robot.keyRelease(KeyEvent.VK_CONTROL);
	robot.keyRelease(KeyEvent.VK_C);
	Thread.sleep(2000);
	robot.keyPress(KeyEvent.VK_RIGHT);
	robot.keyRelease(KeyEvent.VK_RIGHT);

	System.out.println("Check2");

	String str=(String)Toolkit.getDefaultToolkit().getSystemClipboard().getData(DataFlavor.stringFlavor);
	Thread.sleep(2000);

	System.out.println("Copied content is :"+str);
	screen.type(str);
}

}

Console:
Jan 11, 2018 5:41:31 PM java.util.prefs.WindowsPreferences
WARNING: Could not open/create prefs root node Software\JavaSoft\Prefs at root 0x80000002. Windows RegCreateKeyEx(...) returned error code 5.
[log] CLICK on L(540,1401)@s(0)[0,0 2560x1440]
[log] CLICK on L(58,214)@s(0)[0,0 2560x1440]
[log] TYPE "ABCDE"
[log] TYPE "#LEFT.#LEFT.#LEFT.#LEFT."
Check1
Check2
Copied content is :[email protected]
[log] TYPE "[email protected] "

Problem: with certain real world data (in a typical model selection workflow) model predictions end up Inf.
This seems to be the result of high collinearity, unstable parameter estimates.

Demonstration:

library(MIAmaxent)
packageVersion("MIAmaxent")
#> [1] '1.2.0.9000'

reprexdata <- structure(list(RV = c(1, 1, 1, NA, NA, NA, NA, NA, NA, NA), 
                             EV_L = c(0.98, 0.83, 0.94, 0.72, 0.89, 0.7, 0.67, 0.64, 0.39, 0.83), 
                             EV_D2 = c(0, 0.03, 0, 0.07, 0.01, 0.09, 0.1, 0.13, 0.37, 0.03), 
                             EV_M = c(0.96, 0.74, 0.9, 0.6, 0.83, 0.58, 0.54, 0.5, 0.26, 0.74)), 
                        class = "data.frame", row.names = c("1", "2", "3", "4", "5", "6", "7", "8", "9", "10"))

plot(reprexdata[,-1])

cor(reprexdata[,-1], reprexdata[,-1])
#>             EV_L      EV_D2       EV_M
#> EV_L   1.0000000 -0.9451377  0.9946556
#> EV_D2 -0.9451377  1.0000000 -0.9069347
#> EV_M   0.9946556 -0.9069347  1.0000000

iwlr <- MIAmaxent:::.runIWLR(formula("RV ~ EV_L + EV_D2 + EV_M"), reprexdata)
coef(iwlr)
#> (Intercept)        EV_L       EV_D2        EV_M 
#>  -2531.4654   7431.9418    802.6664  -4954.6279
iwlr$entropy
#> [1] 0
iwlr$alpha
#> [1] -Inf

The instability is contingent on the data in unexpected ways. Leave only the highly correlated variables and the problem disappears:

iwlr <- MIAmaxent:::.runIWLR(formula("RV ~ EV_L + EV_M"), reprexdata)
coef(iwlr)
#> (Intercept)        EV_L        EV_M 
#>   -262.0798    782.4296   -530.8548
iwlr$entropy
#> [1] 1.41998
iwlr$alpha
#> [1] -258.5736

This problem can appear in a real world MIAmaxent workflow, as highly correlated derived variables are selected together given a sufficient number of background points:

set.seed(42)
longerdata <- dplyr::slice_sample(reprexdata, n = 1e4, replace = TRUE)
selection <- selectDVforEV(list(RV = longerdata$RV, EV = longerdata[,-1]))
#> Forward selection of DVs for 1 EVs
#>   |                                                                              |                                                                      |   0%  |                                                                              |======================================================================| 100%
selection$selection
#> $EV
#>   round           variables m   Dsq    Chisq df        P
#> 1     1               EV_D2 1 0.099 4033.427  1 0.00e+00
#> 2     1                EV_L 1 0.097 3945.131  1 0.00e+00
#> 3     1                EV_M 1 0.094 3834.647  1 0.00e+00
#> 4     2        EV_D2 + EV_L 2 0.100   26.073  1 3.29e-07
#> 5     2        EV_D2 + EV_M 2 0.099    9.107  1 2.55e-03
#> 6     3 EV_D2 + EV_L + EV_M 3 0.140 1660.015  1 0.00e+00

I have only encountered the issue when both 'L' and 'M'-type transformations are used. The derived variables that result from these may often be highly correlated. So I expect that picking only one of these transformation types in deriveVars() will resolve the problem in most real world cases.

^{Created on 2022-11-22 by the reprex package (v2.0.1)}

From Maxent readme:

The problem is either that you don't have Java installed (you need to
install from java.sun.com as above), or Java is not mentioned in your
"path" variable. To fix the latter problem:

1. From the desktop, right click My Computer and click properties.

2. In the System Properties window, click on the Advanced tab.

3. In the Advanced section, click the Environment Variables button.

4. Finally, in the Environment Variables window, highlight the path
   variable in the Systems Variable section and click edit. Add the
   directory where the java executable is.

For example, if you downloaded Java 1.6.0_13, it was probably
installed in C:\Program Files\Java\jdk1.6.0_13, in
which case you will add ";C:\Program Files\Java\jdk1.6.0_13\bin" to
the end of the path variable, like in the following example:

  C:\Program Files;C:\Winnt;C:\Winnt\System32;C:\Program Files\Java\jdk1.6.0_13\bin

The function always produces values that sum to 1, and therefore corresponds to RVA (relative variation accounted) not FTVA in Halvorsen 2015.

Feature request: a convenient way to visualize the effect of pairwise interaction terms, similar to the plotResp() functions. Either with 3D plot or with facetting.