evaluate: biomod2 modelling outputs evaluation

In biomod2: Ensemble Platform for Species Distribution Modeling

Description

This function will evaluate biomod2 modelling output for given metrics (e.g 'TSS', 'ROC'...) for a given dataset.

Usage

evaluate(model, data, stat, as.array=FALSE)

Arguments

model	the model you want evaluate (either "BIOMOD.models.out", "BIOMOD.EnsembleModeling.out" or "biomod2_model" )
data	the data.set on which you want to perform analyses. Must be a dataset with first column containing the observed data for your species. The flowing columns must be the explanatory variables at observed points. Be sure that columns names of your dataset are the name of your species then the names of variables used for building models at previous steps.
stat	vector of statistic metrics names (e.g 'TSS','ROC') you want to perform. (see BIOMOD_Modeling) to get the list of all available metrics)
as.array	logical, (FALSE by default) if FALSE a list of evaluation tables is returned (one item by models). If TRUE, the output will be return under 'classical' biomod2 array objects (see BIOMOD_Projection)

Details

Given model predictive score is evaluated on the new data set. It is done comparing binary transformed model predictions (on data set) to species occurrences (first column of data arg). A list of available evaluation metrics is given in BIOMOD_Modeling help file. For metrics that compared binary/binary data, a set of threshold will be test to transform continuous model prediction within binary ones. The return scores are the ones obtained for the threshold optimising tested metric (best score).

Value

a list or an array containing for each evaluation metric the score, the threshold considered to transform continuous data into binary ones (for all metrics excepted 'ROC') and associated sensibility and specificity.

Author(s)

Damien Georges

See Also

BIOMOD_Modeling, BIOMOD_EnsembleModeling, variables_importance

Examples

# species occurrences
DataSpecies <- read.csv(system.file("external/species/mammals_table.csv",
                                    package="biomod2"), row.names = 1)
head(DataSpecies)

# the name of studied species
myRespName <- 'GuloGulo'

# the presence/absences data for our species 
myResp <- as.numeric(DataSpecies[,myRespName])

# the XY coordinates of species data
myRespXY <- DataSpecies[,c("X_WGS84","Y_WGS84")]


# Environmental variables extracted from BIOCLIM (bio_3, bio_4, bio_7, bio_11 & bio_12)
myExpl = raster::stack( system.file( "external/bioclim/current/bio3.grd", 
                     package="biomod2"),
                system.file( "external/bioclim/current/bio4.grd", 
                             package="biomod2"), 
                system.file( "external/bioclim/current/bio7.grd", 
                             package="biomod2"),  
                system.file( "external/bioclim/current/bio11.grd", 
                             package="biomod2"), 
                system.file( "external/bioclim/current/bio12.grd", 
                             package="biomod2"))

# 1. Formatting Data
myBiomodData <- BIOMOD_FormatingData(resp.var = myResp,
                                     expl.var = myExpl,
                                     resp.xy = myRespXY,
                                     resp.name = myRespName)

# 2. Defining Models Options using default options.
myBiomodOption <- BIOMOD_ModelingOptions()

# 3. Doing Modelisation

myBiomodModelOut <- BIOMOD_Modeling( myBiomodData, 
                                     models = c('SRE','CTA','RF'), 
                                     models.options = myBiomodOption, 
                                     NbRunEval=1, 
                                     DataSplit=80, 
                                     Yweights=NULL, 
                                     VarImport=3, 
                                     models.eval.meth = c('TSS'),
                                     SaveObj = TRUE,
                                     rescal.all.models = FALSE,
                                     do.full.models = FALSE,
                                     modeling.id='test')
                                     
# 4. Evaluate model over another dataset (here the full one)

## creation of suitable dataset
data <- cbind(GuloGulo=get_formal_data(myBiomodModelOut,'resp.var'), 
              get_formal_data(myBiomodModelOut,'expl.var'))

## evaluation
evaluate(myBiomodModelOut, data=data, stat=c('ROC','TSS'))

biomod2 documentation built on May 2, 2019, 5:08 p.m.