adamlilith / enmsdm Goto Github PK

This includes, but is not limited to, geoFold(), pointDist(), and others.

trainByCrossValid() fails if using weights argument for GLMs and BRTs... need to enable passing of optional arguments by ... while recognizing named arguments

localSpatialCorrForValues() miscalculates distance for rasters... compare same for points in each cell

Hi,
First at all, thanks for this amazing tool!

I'm using the pointDist function to calculate distances between points. My command is like that:
pointDist(i_subset, distFunct = distVincentyEllipsoid, longLat = c('longitude', 'latitude'))
i_subset is a df with columns for points ID, some points attributes and separate columns for long and lat. When I run it I receive the following error: Error in p[, 2] : subscript out of bounds

I believe it might be an issue that emerged with some package update because some months ago I was able to calculate distances with this line of code.

Any thoughts?

Thanks!

If the input dataset contains no points that are closer than minDist, then the progress bar initiated with verbose=TRUE fails, causing the function to fail.

I have copied/pasted the example for spatialCorrForPoints.
There is an error at sacDist <- spatialCorrForPointsSummary(obsAndNullDistrib), the previously created object obsAndNullDistrib has a lot of NaN values.

Error message:

 Obtaining: Observed distances | random points | randomized distances
  |================================================================| 100%
Error in quantile.default(newX[, i], ...) : 
  missing values and NaN's not allowed if 'na.rm' is FALSE

sessionInfo:

R version 3.6.2 (2019-12-12)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: Ubuntu 18.04.3 LTS

Matrix products: default
BLAS:   /usr/lib/x86_64-linux-gnu/openblas/libblas.so.3
LAPACK: /usr/lib/x86_64-linux-gnu/libopenblasp-r0.2.20.so

locale:
 [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C               LC_TIME=en_US.UTF-8        LC_COLLATE=en_US.UTF-8     LC_MONETARY=nl_NL.UTF-8   
 [6] LC_MESSAGES=en_US.UTF-8    LC_PAPER=nl_NL.UTF-8       LC_NAME=C                  LC_ADDRESS=C               LC_TELEPHONE=C            
[11] LC_MEASUREMENT=nl_NL.UTF-8 LC_IDENTIFICATION=C       

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
[1] raster_3.0-12  sp_1.3-2       enmSdm_0.3.9.2

loaded via a namespace (and not attached):
 [1] omnibus_0.3.3.0     compiler_3.6.2      rgdal_1.4-8         tools_3.6.2         geosphere_1.5-10    Rcpp_1.0.3          codetools_0.2-16   
 [8] grid_3.6.2          dismo_1.1-4         statisfactory_0.3.0 lattice_0.20-38 

I was trying to install holoSimCell which depends on enmSdm and I'm getting an error:
Error: package or namespace load failed for ‘enmSdm’ in namespaceExport(ns, exports):
undefined exports: geoFold

After looking at the code, I'm pretty sure that all that needs to happen is to run devtools::document() to update the NAMESPACE file

I am sure I must be doing something wrong here, but my Maxent model produces a convex instead of a concave P/E curve, see image.
I realise it is hard to say something without code, but does this look familiar?

Hi,
I'm using the package to tune maxnet:maxent as part of an analysis of 236 pan-eurpean tree species.
It's working really well, and it saved me considerable time writing a similar (but probably less efficient) code myself. Thanks for that!

Only a small and minor issue to report:
the 'Out' argument in the enmSdm::trainMaxNet function does not accept 'tuning' as a meaningful value. In the code, to return the table, 'Out' should be set to 'table'. It would be nice to change this in the help file.
Best regards,
Yoni Gavish,
School of Geography, University of Leeds.

This is largely a resolved issue, but I'm posting it here because it occasionally sometimes crops up when using functions that can do multicoring, like trainMaxEnt and bioticVelocity. When you try using >1 core, sometimes on some machines you can get the error

object '.doSnowGlobals' not found

I have, for example, received this error on a machine that uses the exact same set of libraries as another, and yet it worked on the second. This StackExchange post explains the issue, or so it seems, but in my case none of the fixes fixed it, and if you're not interested in looking inside the function, it wouldn't help anyway.

What has worked for me:

• An aggressive virus checker can disable access to cores. This was probably my issue, but I don't have admin access to change the antivirus settings.
• However, uninstalling R, then reinstalling (even the same version) seemed to work.

Yes, it was that "simple", and it may not work for you, but it's a solution which I did not find documented on numerous help boards.

Best,
Adam

Hi,

I am trying to run trainByCrossValid for maxnet training. When hinge features are included in the classes, the function fails saying that it could not find the function 'hingeval'. The function works as long as hinge features are not included or if I manually load the 'hingeval' function from the maxnet package into the environment.

mx = trainByCrossValid(data = pts, resp = 'presab', preds = 3:7, folds = pts$block, trainFx = 'trainMaxNet', regMult = seq(0.5,8,0.5), classes = 'lpqh', out = 'tuning')
Modeling k = 1 on Mon Apr 25 09:51:25 2022... Error in hingeval(hmean, 78.3085768563407, 91.973876953125) : could not find function "hingeval"

Thanks,
Lydia

Hello,

Thanks for this great package. The line "=======" raises an error during installation of the package. Removing the line solves the problem.

Hello, R found error that cannot download data from ucdavis website.

worldClim <- getData('worldclim', var='bio', res=5)
试开URL’https://biogeo.ucdavis.edu/data/climate/worldclim/1_4/grid/cur/bio_5m_bil.zip'
Error in utils::download.file(url = aurl, destfile = fn, method = "auto", :
无法打开URL'https://biogeo.ucdavis.edu/data/climate/worldclim/1_4/grid/cur/bio_5m_bil.zip'
此外: Warning message:
In utils::download.file(url = aurl, destfile = fn, method = "auto", :
InternetOpenUrl失败:’不能连接到吊销服务器，或者未能获得最终响应。'