Tuesday, September 3, 2013

Fixed error in resampling in ENMTools 1.4.3

I've fixed a bug in 1.4.3 that kept the "resample from raster" command from printing results to the output file.  It was a very silly error; basically I had disabled printing for debugging and forgot to turn it back on!

Anyway, it's fixed now and should be working fine.  While I was at it, I fixed it so that the resample command now uses the output directory set in the ENMTools Options tab, instead of printing to the directory where the layers you're resampling from are located.  The new version is here:


Saturday, July 6, 2013

ENMTools 1.4.3

While trying to iron out the weirdness of Perl with Mac line endings in .csv files (unsuccessfully), I added some bits of code that seem to have caused the model selection functions in ENMTools to stop working on some input files.  Here's a fixed version.

Sunday, June 16, 2013

Version 1.4.2, adding sampling without replacement to "Resample From Raster" function

By request, I have added a radio button for sampling with or without replacement to the "resample from raster" function.  This function was initially intended for simulating data for methodological studies, but can also be used to sample random points for conducting significance tests for AUC values a la Raes and ter Steege 2007 (using the "constant" setting).  The initial setup was to always resample with replacement.  This isn't ideal for the Raes and ter Steege test, but was unlikely to have any real impact except on models built over very small geographic regions and/or those with very coarse resolution (i.e., study areas with a very small number of grid cells).

I'll post a detailed tutorial eventually, once I get a spare moment to breathe.  Long story short: if you have N data points and want to do X replicates, you load up a raster file that has data in grid cells for your study area and nodata values outside the study area.  This can even be the .asc file for your model itself.  Use the resample from raster tool, constant sampling function, to sample N data points for X replicates.  Then build a single model for each of those replicates using the same study area, model construction settings, and environmental predictors as in your model for your empirical data.  Collect all of the AUC train and test scores from those replicate models, and use those as the null distribution against which to compare your empirical values for AUC train and test.  Guidance on how to do that is here:

Species In Space

The new version is here:

ENMTools 1.4.2

Perl version only, see my previous kvetching about Active State if you want to know why.

Thanks to Marie-France Ostrowski for the suggestion and Renee Catullo for testing it.

Sunday, April 7, 2013

Version 1.4.1 with minor bug fixes

Hey everybody,
Here's a new version that fixes a couple of minor problems with version 1.4.  There was a problem with newlines in the sorted.csv file for the identity test in the previous version, so that all values would appear on a single line.  That should be fixed now, as should a bug that was causing extraneous text to appear after visiting the ENMTools Options page.

ENMTools 1.4.1