Transform categorical into numeric variables

Description

Internal function transform categorical variables in a data.frame into numeric variables. Mostly used with maxent which cannot read character

Usage

.categorical2numeric(df, categorical_var)

Arguments

Value

a data.frame without categorical variables

Check duplicated cells

Description

Identify data that are contained in the same raster cells ; print warnings if need be and filter those data if asked for.

Usage

.check_duplicated_cells(env, xy, sp, filter.raster, PA.user.table = NULL)

Arguments

Value

a list

Get categorical variable names

Description

Internal function to get categorical variables name from a data.frame.

Usage

.get_categorical_names(df)

Arguments

Value

a vector with the name of categorical variables

Get class of environmental data provided

Description

Get class of environmental data provided

Usage

.get_env_class(new.env)

Arguments

Value

a character

Load library for GAM models

Description

This function loads library for either GAM and BAM from mgcv package or for GAM from gam package.

Usage

.load_gam_namespace(
  model_subclass = c("GAM_mgcv_gam", "GAM_mgcv_bam", "GAM_gam_gam")
)

Arguments

.transform.outputs.list

Description

Deprecated function name for .transform_outputs_list

Usage

.transform.outputs.list(...)

Arguments

See Also

biomod2-deprecated

Transform predictions data.frame from long to wide with models as columns

Description

This function is used internally in get_predictions to ensure back-compatibility with former output of get_predictions (i.e. for biomod2 version < 4.2-2). It transform a long data.frame into a wide data.frame with each column corresponding to a single model.

Note that the function is intended for internal use but have been made available for compatibility with ecospat

Usage

.transform_model.as.col(df)

Arguments

Value

a wide data.frame

Author(s)

Rémi Lemaire-Patin

BIOMOD_EnsembleModeling() output object class

Description

Class returned by BIOMOD_EnsembleModeling, and used by BIOMOD_LoadModels, BIOMOD_PresenceOnly and BIOMOD_EnsembleForecasting

Usage

## S4 method for signature 'BIOMOD.ensemble.models.out'
show(object)

Arguments

Slots

modeling.id

a character corresponding to the name (ID) of the simulation set

dir.name

a character corresponding to the modeling folder

sp.name

a character corresponding to the species name

expl.var.names

a vector containing names of explanatory variables

data.type

a character corresponding to the data type

models.out

a BIOMOD.stored.models.out-class object containing informations from BIOMOD_Modeling object

em.by

a character corresponding to the way kept models have been combined to build the ensemble models, must be among PA+run, PA+algo, PA, algo, all

em.computed

a vector containing names of ensemble models

em.failed

a vector containing names of failed ensemble models

em.models_kept

a list containing single models for each ensemble model

models.evaluation

a BIOMOD.stored.data.frame-class object containing models evaluation

variables.importance

a BIOMOD.stored.data.frame-class object containing variables importance

models.prediction

a BIOMOD.stored.data.frame-class object containing models predictions

models.prediction.eval

a BIOMOD.stored.data.frame-class object containing models predictions for evaluation data

link

a character containing the file name of the saved object

call

a language object corresponding to the call used to obtain the object

Author(s)

Damien Georges

See Also

BIOMOD_EnsembleModeling, BIOMOD_LoadModels, BIOMOD_PresenceOnly, bm_VariablesImportance, bm_PlotEvalMean, bm_PlotEvalBoxplot, bm_PlotVarImpBoxplot, bm_PlotResponseCurves

Other Toolbox objects: BIOMOD.formated.data, BIOMOD.formated.data.PA, BIOMOD.models.options, BIOMOD.models.out, BIOMOD.options.dataset, BIOMOD.options.default, BIOMOD.projection.out, BIOMOD.rangesize.out, BIOMOD.stored.data, biomod2_ensemble_model, biomod2_model

Examples

showClass("BIOMOD.ensemble.models.out")

## ----------------------------------------------------------------------- #
library(terra)

# Load species occurrences (6 species available)
data(DataSpecies)
head(DataSpecies)

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

# Get corresponding presence/absence data
myResp <- as.numeric(DataSpecies[, myRespName])

# Get corresponding XY coordinates
myRespXY <- DataSpecies[, c('X_WGS84', 'Y_WGS84')]

# Load environmental variables extracted from BIOCLIM (bio_3, bio_4, bio_7, bio_11 & bio_12)
data(bioclim_current)
myExpl <- terra::rast(bioclim_current)



## ----------------------------------------------------------------------- #
file.out <- paste0(myRespName, "/", myRespName, ".AllModels.models.out")
if (file.exists(file.out)) {
  myBiomodModelOut <- get(load(file.out))
} else {

  # Format Data with true absences
  myBiomodData <- BIOMOD_FormatingData(resp.name = myRespName,
                                       resp.var = myResp,
                                       resp.xy = myRespXY,
                                       expl.var = myExpl)

  # Model single models
  myBiomodModelOut <- BIOMOD_Modeling(bm.format = myBiomodData,
                                      modeling.id = 'AllModels',
                                      models = c('RF', 'GLM'),
                                      CV.strategy = 'random',
                                      CV.nb.rep = 2,
                                      CV.perc = 0.8,
                                      OPT.strategy = 'bigboss',
                                      metric.eval = c('TSS', 'AUCroc'),
                                      var.import = 3,
                                      seed.val = 42)
}


## ----------------------------------------------------------------------- #
# Model ensemble models
myBiomodEM <- BIOMOD_EnsembleModeling(bm.mod = myBiomodModelOut,
                                      models.chosen = 'all',
                                      em.by = 'all',
                                      em.algo = c('EMmean', 'EMca'),
                                      metric.select = c('TSS'),
                                      metric.select.thresh = c(0.7),
                                      metric.eval = c('TSS', 'AUCroc'),
                                      var.import = 3,
                                      seed.val = 42)
myBiomodEM

BIOMOD_FormatingData() output object class

Description

Class returned by BIOMOD_FormatingData, and used by bm_Tuning, bm_CrossValidation and BIOMOD_Modeling

Usage

## S4 method for signature 'numeric,data.frame'
BIOMOD.formated.data(
  sp,
  env,
  xy = NULL,
  dir.name = ".",
  data.type = NULL,
  sp.name = NULL,
  eval.sp = NULL,
  eval.env = NULL,
  eval.xy = NULL,
  na.rm = TRUE,
  data.mask = NULL,
  shared.eval.env = FALSE,
  filter.raster = FALSE
)

## S4 method for signature 'data.frame,ANY'
BIOMOD.formated.data(
  sp,
  env,
  xy = NULL,
  dir.name = ".",
  data.type = NULL,
  sp.name = NULL,
  eval.sp = NULL,
  eval.env = NULL,
  eval.xy = NULL,
  na.rm = TRUE,
  filter.raster = FALSE
)

## S4 method for signature 'numeric,matrix'
BIOMOD.formated.data(
  sp,
  env,
  xy = NULL,
  dir.name = ".",
  data.type = NULL,
  sp.name = NULL,
  eval.sp = NULL,
  eval.env = NULL,
  eval.xy = NULL,
  na.rm = TRUE,
  filter.raster = FALSE
)

## S4 method for signature 'numeric,SpatRaster'
BIOMOD.formated.data(
  sp,
  env,
  xy = NULL,
  dir.name = ".",
  data.type = NULL,
  sp.name = NULL,
  eval.sp = NULL,
  eval.env = NULL,
  eval.xy = NULL,
  na.rm = TRUE,
  shared.eval.env = FALSE,
  filter.raster = FALSE
)

## S4 method for signature 'BIOMOD.formated.data'
show(object)

Arguments

Slots

data.type

a character corresponding to the data type

dir.name

a character corresponding to the modeling folder

sp.name

a character corresponding to the species name

coord

a 2-columns data.frame containing the corresponding X and Y coordinates

data.species

a vector containing the species observations (0, 1 or NA)

data.env.var

a data.frame containing explanatory variables

data.mask

a SpatRaster object containing the mask of the studied area

has.data.eval

a logical value defining whether evaluation data is given

has.filter.raster

a logical value defining whether filtering have been done or not

eval.coord

(optional, default NULL)
A 2-columns data.frame containing the corresponding X and Y coordinates for evaluation data

eval.data.species

(optional, default NULL)
A vector containing the species observations (0, 1 or NA) for evaluation data

eval.data.env.var

(optional, default NULL)
A data.frame containing explanatory variables for evaluation data

biomod2.version

a character corresponding to the biomod2 version

call

a language object corresponding to the call used to obtain the object

Author(s)

Damien Georges

See Also

BIOMOD_FormatingData, bm_Tuning, bm_CrossValidation, BIOMOD_Modeling, bm_RunModelsLoop

Other Toolbox objects: BIOMOD.ensemble.models.out, BIOMOD.formated.data.PA, BIOMOD.models.options, BIOMOD.models.out, BIOMOD.options.dataset, BIOMOD.options.default, BIOMOD.projection.out, BIOMOD.rangesize.out, BIOMOD.stored.data, biomod2_ensemble_model, biomod2_model

Examples

showClass("BIOMOD.formated.data")

## ----------------------------------------------------------------------- #
library(terra)

# Load species occurrences (6 species available)
data(DataSpecies)
head(DataSpecies)

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

# Get corresponding presence/absence data
myResp <- as.numeric(DataSpecies[, myRespName])

# Get corresponding XY coordinates
myRespXY <- DataSpecies[, c('X_WGS84', 'Y_WGS84')]

# Load environmental variables extracted from BIOCLIM (bio_3, bio_4, bio_7, bio_11 & bio_12)
data(bioclim_current)
myExpl <- terra::rast(bioclim_current)



## ----------------------------------------------------------------------- #
# Format Data with true absences
myBiomodData <- BIOMOD_FormatingData(resp.name = myRespName,
                                     resp.var = myResp,
                                     resp.xy = myRespXY,
                                     expl.var = myExpl)
myBiomodData
plot(myBiomodData)
summary(myBiomodData)

BIOMOD_FormatingData() output object class (with pseudo-absences)

Description

Class returned by BIOMOD_FormatingData, and used by bm_Tuning, bm_CrossValidation and BIOMOD_Modeling

Usage

## S4 method for signature 'numeric,data.frame'
BIOMOD.formated.data.PA(
  sp,
  env,
  xy = NULL,
  dir.name = ".",
  sp.name = NULL,
  eval.sp = NULL,
  eval.env = NULL,
  eval.xy = NULL,
  PA.nb.rep = 1,
  PA.strategy = "random",
  PA.nb.absences = NULL,
  PA.dist.min = 0,
  PA.dist.max = NULL,
  PA.sre.quant = 0.025,
  PA.fact.aggr = NULL,
  PA.user.table = NULL,
  na.rm = TRUE,
  filter.raster = FALSE,
  seed.val = NULL
)

## S4 method for signature 'numeric,SpatRaster'
BIOMOD.formated.data.PA(
  sp,
  env,
  xy = NULL,
  dir.name = ".",
  sp.name = NULL,
  eval.sp = NULL,
  eval.env = NULL,
  eval.xy = NULL,
  PA.nb.rep = 1,
  PA.strategy = "random",
  PA.nb.absences = NULL,
  PA.dist.min = 0,
  PA.dist.max = NULL,
  PA.sre.quant = 0.025,
  PA.fact.aggr = NULL,
  PA.user.table = NULL,
  na.rm = TRUE,
  filter.raster = FALSE,
  seed.val = NULL
)

Arguments

Slots

dir.name

a character corresponding to the modeling folder

sp.name

a character corresponding to the species name

coord

a 2-columns data.frame containing the corresponding X and Y coordinates

data.species

a vector containing the species observations (0, 1 or NA)

data.env.var

a data.frame containing explanatory variables

data.mask

a SpatRaster object containing the mask of the studied area

has.data.eval

a logical value defining whether evaluation data is given

eval.coord

(optional, default NULL)
A 2-columns data.frame containing the corresponding X and Y coordinates for evaluation data

eval.data.species

(optional, default NULL)
A vector containing the species observations (0, 1 or NA) for evaluation data

eval.data.env.var

(optional, default NULL)
A data.frame containing explanatory variables for evaluation data

PA.strategy

a character corresponding to the pseudo-absence selection strategy

PA.table

a data.frame containing the corresponding table of selected pseudo-absences (indicated by TRUE or FALSE) from the pa.tab list element returned by the bm_PseudoAbsences function

Author(s)

Damien Georges

See Also

BIOMOD_FormatingData, bm_PseudoAbsences, bm_Tuning, bm_CrossValidation, BIOMOD_Modeling, bm_RunModelsLoop

Other Toolbox objects: BIOMOD.ensemble.models.out, BIOMOD.formated.data, BIOMOD.models.options, BIOMOD.models.out, BIOMOD.options.dataset, BIOMOD.options.default, BIOMOD.projection.out, BIOMOD.rangesize.out, BIOMOD.stored.data, biomod2_ensemble_model, biomod2_model

Examples

showClass("BIOMOD.formated.data.PA")

## ----------------------------------------------------------------------- #
library(terra)

# Load species occurrences (6 species available)
data(DataSpecies)
head(DataSpecies)

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

# Keep only presence informations
DataSpecies <- DataSpecies[which(DataSpecies[, myRespName] == 1), ]

# Get corresponding presence/absence data
myResp <- as.numeric(DataSpecies[, myRespName])

# Get corresponding XY coordinates
myRespXY <- DataSpecies[, c('X_WGS84', 'Y_WGS84')]

# Load environmental variables extracted from BIOCLIM (bio_3, bio_4, bio_7, bio_11 & bio_12)
data(bioclim_current)
myExpl <- terra::rast(bioclim_current)



## ----------------------------------------------------------------------- #
# Format Data with pseudo-absences : random method
myBiomodData <- BIOMOD_FormatingData(resp.name = myRespName,
                                     resp.var = myResp,
                                     resp.xy = myRespXY,
                                     expl.var = myExpl,
                                     PA.nb.rep = 4,
                                     PA.strategy = 'random',
                                     PA.nb.absences = 1000)
myBiomodData
plot(myBiomodData)

bm_ModelingOptions output object class

Description

Class returned by bm_ModelingOptions and used by BIOMOD_Modeling

Usage

## S4 method for signature 'BIOMOD.models.options'
show(object)

## S4 method for signature 'BIOMOD.models.options'
print(x, dataset = "_allData_allRun")

Arguments

Slots

models

a vector containing model names for which options have been retrieved and defined, must be algo.datatype.package.function

options

a list containing BIOMOD.options.dataset object for each model

Author(s)

Maya Guéguen

See Also

BIOMOD.options.default, BIOMOD.options.dataset, bm_ModelingOptions, bm_Tuning, BIOMOD_Modeling

Other Toolbox objects: BIOMOD.ensemble.models.out, BIOMOD.formated.data, BIOMOD.formated.data.PA, BIOMOD.models.out, BIOMOD.options.dataset, BIOMOD.options.default, BIOMOD.projection.out, BIOMOD.rangesize.out, BIOMOD.stored.data, biomod2_ensemble_model, biomod2_model

Examples

showClass("BIOMOD.models.options")

BIOMOD_Modeling() output object class

Description

Class returned by BIOMOD_Modeling, and used by BIOMOD_LoadModels, BIOMOD_PresenceOnly, BIOMOD_Projection and BIOMOD_EnsembleModeling

Usage

## S4 method for signature 'BIOMOD.models.out'
show(object)

Arguments

Slots

modeling.id

a character corresponding to the name (ID) of the simulation set

dir.name

a character corresponding to the modeling folder

sp.name

a character corresponding to the species name

data.type

a character corresponding to the data type

expl.var.names

a vector containing names of explanatory variables

models.computed

a vector containing names of computed models

models.failed

a vector containing names of failed models

has.evaluation.data

a logical value defining whether evaluation data is given

scale.models

a logical value defining whether models have been rescaled or not

formated.input.data

a BIOMOD.stored.formated.data-class object containing informations from BIOMOD_FormatingData object

calib.lines

a BIOMOD.stored.data.frame-class object containing calibration lines

models.options

a BIOMOD.stored.options-class object containing informations from bm_ModelingOptions object

models.evaluation

a BIOMOD.stored.data.frame-class object containing models evaluation

variables.importance

a BIOMOD.stored.data.frame-class object containing variables importance

models.prediction

a BIOMOD.stored.data.frame-class object containing models predictions

models.prediction.eval

a BIOMOD.stored.data.frame-class object containing models predictions for evaluation data

link

a character containing the file name of the saved object

call

a language object corresponding to the call used to obtain the object

Author(s)

Damien Georges

See Also

BIOMOD_Modeling, BIOMOD_LoadModels, BIOMOD_PresenceOnly, BIOMOD_Projection, BIOMOD_EnsembleModeling, bm_VariablesImportance, bm_PlotEvalMean, bm_PlotEvalBoxplot, bm_PlotVarImpBoxplot, bm_PlotResponseCurves

Other Toolbox objects: BIOMOD.ensemble.models.out, BIOMOD.formated.data, BIOMOD.formated.data.PA, BIOMOD.models.options, BIOMOD.options.dataset, BIOMOD.options.default, BIOMOD.projection.out, BIOMOD.rangesize.out, BIOMOD.stored.data, biomod2_ensemble_model, biomod2_model

Examples

showClass("BIOMOD.models.out")

## ----------------------------------------------------------------------- #
library(terra)

# Load species occurrences (6 species available)
data(DataSpecies)
head(DataSpecies)

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

# Get corresponding presence/absence data
myResp <- as.numeric(DataSpecies[, myRespName])

# Get corresponding XY coordinates
myRespXY <- DataSpecies[, c('X_WGS84', 'Y_WGS84')]

# Load environmental variables extracted from BIOCLIM (bio_3, bio_4, bio_7, bio_11 & bio_12)
data(bioclim_current)
myExpl <- terra::rast(bioclim_current)



## ----------------------------------------------------------------------- #
# Format Data with true absences
myBiomodData <- BIOMOD_FormatingData(resp.name = myRespName,
                                     resp.var = myResp,
                                     resp.xy = myRespXY,
                                     expl.var = myExpl)

## ----------------------------------------------------------------------- #
# Model single models
myBiomodModelOut <- BIOMOD_Modeling(bm.format = myBiomodData,
                                    modeling.id = 'AllModels',
                                    models = c('RF', 'GLM'),
                                    CV.strategy = 'random',
                                    CV.nb.rep = 2,
                                    CV.perc = 0.8,
                                    OPT.strategy = 'bigboss',
                                    metric.eval = c('TSS', 'AUCroc'),
                                    var.import = 3,
                                    seed.val = 42)
myBiomodModelOut

bm_ModelingOptions output object class

Description

Class returned by bm_ModelingOptions (a list of BIOMOD.options.dataset more exactly), and used by BIOMOD_Modeling

Usage

## S4 method for signature 'character'
BIOMOD.options.dataset(
  mod,
  typ,
  pkg,
  fun,
  strategy,
  user.val = NULL,
  user.base = NULL,
  tuning.fun = NULL,
  bm.format = NULL,
  calib.lines = NULL
)

## S4 method for signature 'BIOMOD.options.dataset'
show(object)

## S4 method for signature 'BIOMOD.options.dataset'
print(x, dataset = "_allData_allRun")

Arguments

Slots

model

a character corresponding to the model

type

a character corresponding to the data type (binary, binary.PA, abundance, compositional)

package

a character corresponding to the package containing the model function to be called

func

a character corresponding to the model function name to be called

args.names

a vector containing character corresponding to the model function arguments

args.default

a list containing for each dataset the default values for all arguments listed in args.names

args.values

a list containing for each dataset the to-be-used values for all arguments listed in args.names

Author(s)

Maya Guéguen

See Also

BIOMOD.options.default, bm_ModelingOptions, bm_Tuning, BIOMOD_Modeling, bm_RunModelsLoop

Other Toolbox objects: BIOMOD.ensemble.models.out, BIOMOD.formated.data, BIOMOD.formated.data.PA, BIOMOD.models.options, BIOMOD.models.out, BIOMOD.options.default, BIOMOD.projection.out, BIOMOD.rangesize.out, BIOMOD.stored.data, biomod2_ensemble_model, biomod2_model

Examples

showClass("BIOMOD.options.dataset")

bm_ModelingOptions output object class

Description

Class returned by bm_ModelingOptions (a list of BIOMOD.options.dataset more exactly), and used by BIOMOD_Modeling

Usage

## S4 method for signature 'character,character'
BIOMOD.options.default(mod, typ, pkg, fun)

Arguments

Slots

model

a character corresponding to the model

type

a character corresponding to the data type (binary, binary.PA, abundance, compositional)

package

a character corresponding to the package containing the model function to be called

func

a character corresponding to the model function name to be called

args.names

a vector containing character corresponding to the model function arguments

args.default

a list containing for each dataset the default values for all arguments listed in args.names

Author(s)

Maya Guéguen

See Also

BIOMOD.options.dataset, bm_ModelingOptions, bm_Tuning, BIOMOD_Modeling, bm_RunModelsLoop

Other Toolbox objects: BIOMOD.ensemble.models.out, BIOMOD.formated.data, BIOMOD.formated.data.PA, BIOMOD.models.options, BIOMOD.models.out, BIOMOD.options.dataset, BIOMOD.projection.out, BIOMOD.rangesize.out, BIOMOD.stored.data, biomod2_ensemble_model, biomod2_model

Examples

showClass("BIOMOD.options.default")

BIOMOD_Projection() output object class

Description

Class returned by BIOMOD_Projection, and used by BIOMOD_EnsembleForecasting

Usage

## S4 method for signature 'BIOMOD.projection.out,missing'
plot(
  x,
  coord = NULL,
  plot.output,
  do.plot = TRUE,
  std = TRUE,
  scales,
  size,
  maxcell = 5e+05,
  ...
)

## S4 method for signature 'BIOMOD.projection.out'
show(object)

Arguments

Slots

modeling.id

a character corresponding to the name (ID) of the simulation set

proj.name

a character corresponding to the projection name

dir.name

a character corresponding to the modeling folder

sp.name

a character corresponding to the species name

expl.var.names

a vector containing names of explanatory variables

coord

a 2-columns matrix or data.frame containing the corresponding X and Y coordinates used to project the species distribution model(s)

scale.models

a logical value defining whether models have been rescaled or not

models.projected

a vector containing names of projected models

models.out

a BIOMOD.stored.data object

type

a character corresponding to the class of the val slot of the proj.out slot

data.type

a character corresponding to the data type

proj.out

a BIOMOD.stored.data object

call

a language object corresponding to the call used to obtain the object

Author(s)

Damien Georges

See Also

BIOMOD_Projection, BIOMOD_EnsembleForecasting

Other Toolbox objects: BIOMOD.ensemble.models.out, BIOMOD.formated.data, BIOMOD.formated.data.PA, BIOMOD.models.options, BIOMOD.models.out, BIOMOD.options.dataset, BIOMOD.options.default, BIOMOD.rangesize.out, BIOMOD.stored.data, biomod2_ensemble_model, biomod2_model

Examples

showClass("BIOMOD.projection.out")

## ----------------------------------------------------------------------- #
library(terra)

# Load species occurrences (6 species available)
data(DataSpecies)
head(DataSpecies)

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

# Get corresponding presence/absence data
myResp <- as.numeric(DataSpecies[, myRespName])

# Get corresponding XY coordinates
myRespXY <- DataSpecies[, c('X_WGS84', 'Y_WGS84')]

# Load environmental variables extracted from BIOCLIM (bio_3, bio_4, bio_7, bio_11 & bio_12)
data(bioclim_current)
myExpl <- terra::rast(bioclim_current)



## ----------------------------------------------------------------------- #
file.out <- paste0(myRespName, "/", myRespName, ".AllModels.models.out")
if (file.exists(file.out)) {
  myBiomodModelOut <- get(load(file.out))
} else {

  # Format Data with true absences
  myBiomodData <- BIOMOD_FormatingData(resp.name = myRespName,
                                       resp.var = myResp,
                                       resp.xy = myRespXY,
                                       expl.var = myExpl)

  # Model single models
  myBiomodModelOut <- BIOMOD_Modeling(bm.format = myBiomodData,
                                      modeling.id = 'AllModels',
                                      models = c('RF', 'GLM'),
                                      CV.strategy = 'random',
                                      CV.nb.rep = 2,
                                      CV.perc = 0.8,
                                      OPT.strategy = 'bigboss',
                                      metric.eval = c('TSS', 'AUCroc'),
                                      var.import = 3,
                                      seed.val = 42)
}


## ----------------------------------------------------------------------- #
# Project single models
myBiomodProj <- BIOMOD_Projection(bm.mod = myBiomodModelOut,
                                  proj.name = 'Current',
                                  new.env = myExpl,
                                  models.chosen = 'all',
                                  metric.binary = 'all',
                                  metric.filter = 'all',
                                  build.clamping.mask = TRUE)
myBiomodProj
plot(myBiomodProj)

BIOMOD_RangeSize() output object class

Description

Class returned by BIOMOD_RangeSize, and used by bm_PlotRangeSize

Slots

Compt.By.Models

a data.frame containing the summary of range change for each comparison

Diff.By.Pixel

a SpatRaster or a data.frame containing a value for each point/pixel of each comparison

loss.gain

a SpatRaster or a data.frame containg for each point/pixel a value indicating a loss, a gain or a stable sitatution

data.type

a character corresponding to the data type

coord

a 2-columns matrix or data.frame containing the corresponding X and Y coordinates used to project the species distribution model(s)

row.names

A vector containing tags matching models names splitted by '_' character

Author(s)

Hélène Blancheteau

See Also

Other Toolbox objects: BIOMOD.ensemble.models.out, BIOMOD.formated.data, BIOMOD.formated.data.PA, BIOMOD.models.options, BIOMOD.models.out, BIOMOD.options.dataset, BIOMOD.options.default, BIOMOD.projection.out, BIOMOD.stored.data, biomod2_ensemble_model, biomod2_model

BIOMOD_Modeling and BIOMOD_EnsembleModeling output object class

Description

Classes used by BIOMOD_Modeling and BIOMOD_EnsembleModeling to build their output object (see BIOMOD.models.out objects)

Details

BIOMOD.stored.data is the basic object containing the slots inMemory and link.
All listed classes below are derived from BIOMOD.stored.data, and contain a val slot of specific type :

• BIOMOD.stored.data.frame : val is a data.frame

• BIOMOD.stored.SpatRaster : val is a PackedSpatRaster

• BIOMOD.stored.files : val is a character

• BIOMOD.stored.formated.data : val is a BIOMOD.formated.data object

• BIOMOD.stored.options : val is a BIOMOD.models.options object

• BIOMOD.stored.models.out : val is a BIOMOD.models.out object

Slots

inMemory

a logical defining whether the val slot has been loaded in memory or not

link

a character containing the file name of the saved val slot

val

an object of type depending on the BIOMOD.stored.[...] class (see Details)

Author(s)

Damien Georges

See Also

BIOMOD.formated.data, BIOMOD.models.out, BIOMOD_Modeling, BIOMOD_EnsembleModeling, BIOMOD_Projection, BIOMOD_EnsembleForecasting

Other Toolbox objects: BIOMOD.ensemble.models.out, BIOMOD.formated.data, BIOMOD.formated.data.PA, BIOMOD.models.options, BIOMOD.models.out, BIOMOD.options.dataset, BIOMOD.options.default, BIOMOD.projection.out, BIOMOD.rangesize.out, biomod2_ensemble_model, biomod2_model

Examples

showClass("BIOMOD.stored.data")
showClass("BIOMOD.stored.data.frame") 
showClass("BIOMOD.stored.SpatRaster") 
showClass("BIOMOD.stored.files") 
showClass("BIOMOD.stored.formated.data") 
showClass("BIOMOD.stored.options") 
showClass("BIOMOD.stored.models.out") 

BIOMOD_CrossValidation

Description

Deprecated function name for bm_CrossValidation

Usage

BIOMOD_CrossValidation(...)

Arguments

See Also

biomod2-deprecated

Project ensemble species distribution models onto new environment

Description

This function allows to project ensemble models built with the BIOMOD_EnsembleModeling function onto new environmental data (which can represent new areas, resolution or time scales for example).

Usage

BIOMOD_EnsembleForecasting(
  bm.em,
  bm.proj = NULL,
  proj.name = NULL,
  new.env = NULL,
  new.env.xy = NULL,
  models.chosen = "all",
  metric.binary = NULL,
  metric.filter = NULL,
  na.rm = TRUE,
  nb.cpu = 1,
  ...
)

Arguments

Details

... can take the following values :

digits

(optional, default 0) :
an integer value corresponding to the number of digits of the predictions

on_0_1000

(optional, default TRUE) :
a logical value defining whether 0 - 1 probabilities are to be converted to 0 - 1000 scale to save memory on backup

keep.in.memory

(optional, default TRUE) :
a logical value defining whether all projections are to be kept loaded at once in memory, or only links pointing to hard drive are to be returned

do.stack

(optional, default TRUE) :
a logical value defining whether all projections are to be saved as one SpatRaster object or several SpatRaster files (the default if projections are too heavy to be all loaded at once in memory)

output.format

(optional, default .RData or .tif) :
a character value corresponding to the projections saving format on hard drive, must be either .grd, .img, .tif or .RData (the default if new.env is given as matrix or data.frame)

compress

(optional, default TRUE) :
a logical or a character value defining whether and how objects should be compressed when saved on hard drive. Must be either TRUE, FALSE, gzip (for Windows OS) or xz (for other OS)

Value

A BIOMOD.projection.out object containing models projections, or links to saved outputs.
Models projections are stored out of R (for memory storage reasons) in proj.name folder created in the current working directory :

1. the output is a data.frame if new.env is a matrix or a data.frame

2. it is a SpatRaster if new.env is a SpatRaster (or several SpatRaster objects, if new.env is too large)

3. raw projections, as well as binary and filtered projections (if asked), are saved in the proj.name folder

Author(s)

Wilfried Thuiller, Damien Georges, Robin Engler

See Also

BIOMOD_FormatingData, bm_ModelingOptions, BIOMOD_Modeling, BIOMOD_EnsembleModeling, BIOMOD_RangeSize

Other Main functions: BIOMOD_EnsembleModeling(), BIOMOD_FormatingData(), BIOMOD_LoadModels(), BIOMOD_Modeling(), BIOMOD_Projection(), BIOMOD_RangeSize()

Examples

library(terra)
# Load species occurrences (6 species available)
data(DataSpecies)
head(DataSpecies)

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

# Get corresponding presence/absence data
myResp <- as.numeric(DataSpecies[, myRespName])

# Get corresponding XY coordinates
myRespXY <- DataSpecies[, c('X_WGS84', 'Y_WGS84')]

# Load environmental variables extracted from BIOCLIM (bio_3, bio_4, bio_7, bio_11 & bio_12)
data(bioclim_current)
myExpl <- terra::rast(bioclim_current)



 
# --------------------------------------------------------------- #
file.out <- paste0(myRespName, "/", myRespName, ".AllModels.models.out")
if (file.exists(file.out)) {
  myBiomodModelOut <- get(load(file.out))
} else {

  # Format Data with true absences
  myBiomodData <- BIOMOD_FormatingData(resp.name = myRespName,
                                       resp.var = myResp,
                                       resp.xy = myRespXY,
                                       expl.var = myExpl)

  # Model single models
  myBiomodModelOut <- BIOMOD_Modeling(bm.format = myBiomodData,
                                      modeling.id = 'AllModels',
                                      models = c('RF', 'GLM'),
                                      CV.strategy = 'random',
                                      CV.nb.rep = 2,
                                      CV.perc = 0.8,
                                      OPT.strategy = 'bigboss',
                                      metric.eval = c('TSS', 'AUCroc'),
                                      var.import = 3,
                                      seed.val = 42)
}


file.proj <- paste0(myRespName, "/proj_Current/", myRespName, ".Current.projection.out")
if (file.exists(file.proj)) {
  myBiomodProj <- get(load(file.proj))
} else {

  # Project single models
  myBiomodProj <- BIOMOD_Projection(bm.mod = myBiomodModelOut,
                                    proj.name = 'Current',
                                    new.env = myExpl,
                                    models.chosen = 'all',
                                    build.clamping.mask = TRUE)
}


file.EM <- paste0(myRespName, "/", myRespName, ".AllModels.ensemble.models.out")
if (file.exists(file.EM)) {
  myBiomodEM <- get(load(file.EM))
} else {

  # Model ensemble models
  myBiomodEM <- BIOMOD_EnsembleModeling(bm.mod = myBiomodModelOut,
                                        models.chosen = 'all',
                                        em.by = 'all',
                                        em.algo = c('EMmean', 'EMca'),
                                        metric.select = c('TSS'),
                                        metric.select.thresh = c(0.7),
                                        metric.eval = c('TSS', 'AUCroc'),
                                        var.import = 3,
                                        seed.val = 42)
}


# --------------------------------------------------------------- #
# Project ensemble models (from single projections)
myBiomodEMProj <- BIOMOD_EnsembleForecasting(bm.em = myBiomodEM, 
                                             bm.proj = myBiomodProj,
                                             models.chosen = 'all',
                                             metric.binary = 'all',
                                             metric.filter = 'all')

# Project ensemble models (building single projections)
myBiomodEMProj <- BIOMOD_EnsembleForecasting(bm.em = myBiomodEM,
                                             proj.name = 'CurrentEM',
                                             new.env = myExpl,
                                             models.chosen = 'all',
                                             metric.binary = 'all',
                                             metric.filter = 'all')
myBiomodEMProj
plot(myBiomodEMProj)

Create and evaluate an ensemble set of models and predictions

Description

This function allows to combine a range of models built with the BIOMOD_Modeling function in one (or several) ensemble model. Modeling uncertainty can be assessed as well as variables importance, ensemble predictions can be evaluated against original data, and created ensemble models can be projected over new conditions (see Details).

Usage

BIOMOD_EnsembleModeling(
  bm.mod,
  models.chosen = "all",
  em.by = "PA+run",
  em.algo,
  metric.select = "all",
  metric.select.thresh = NULL,
  metric.select.table = NULL,
  metric.select.dataset = NULL,
  metric.eval = c("KAPPA", "TSS", "AUCroc"),
  var.import = 0,
  EMci.alpha = 0.05,
  EMwmean.decay = "proportional",
  nb.cpu = 1,
  seed.val = NULL,
  do.progress = TRUE
)

Arguments

Details

Concerning models sub-selection (models.chosen) :
Applying get_built_models function to the bm.mod object gives the names of the single models created with the BIOMOD_Modeling function. The models.chosen argument can take either a sub-selection of these single model names, or the all default value, to decide which single models will be used for the ensemble model building.

Concerning models assembly rules (em.by) :
Single models built with the BIOMOD_Modeling function can be combined in 5 different ways to obtain ensemble models :

PA+run

each combination of pseudo-absence and repetition datasets is done, merging algorithms together

PA+algo

each combination of pseudo-absence and algorithm datasets is done, merging repetitions together

PA

pseudo-absence datasets are considered individually, merging algorithms and repetitions together

algo

algorithm datasets are considered individually, merging pseudo-absence and repetitions together

all

all single models are combined into one

Hence, depending on the chosen method, the number of ensemble models built will vary.
If no evaluation data was given to the BIOMOD_FormatingData function, some ensemble model evaluations may be biased due to difference in data used for single model evaluations.
Be aware that all of these combinations are allowed, but some may not make sense depending mainly on how pseudo-absence datasets have been built and whether all of them have been used for all single models or not (see PA.nb.absences and models.pa parameters in BIOMOD_FormatingData and BIOMOD_Modeling functions respectively).

Concerning evaluation metrics :

metric.select

metric(s) must be chosen among the ones used within the BIOMOD_Modeling function to build the bm.mod object, unless metric.select = 'user.defined' and therefore values will be provided through the metric.select.table parameter.
Each selected metric will be used at different steps of the ensemble modeling function to :

1. remove low quality single models having a score lower than metric.select.thresh

2. perform the binary transformation if em.algo = 'EMca'

3. weight models if em.algo = 'EMwmean'

Note that metrics are not combined together, and one ensemble model is built for each metric provided.

metric.select.table

if metric.select = 'user.defined', this parameter allows to use evaluation metrics other than those calculated within biomod2. It must be a data.frame containing as many columns as models.chosen with matching names, and as many rows as evaluation metrics to be used. The number of rows must match the length of metric.select.thresh, and values will be compared to those defined in metric.select.thresh to remove low quality single models from the ensemble model building.

metric.select.dataset

by default, validation datasets will be used, unless no validation is available (no cross-validation) in which case calibration datasets will be used

Concerning ensemble algorithms :
6 modeling techniques are currently available :

EMmedian

median of probabilities over the selected models

Less sensitive to outliers than the mean

EMmean

mean of probabilities over the selected models

EMwmean

weighted mean of probabilities over the selected models

Probabilities are weighted according to their model evaluation scores obtained when building the bm.out object with the BIOMOD_Modeling function (better a model is, more importance it has in the ensemble) and summed.

The EMwmean.decay is the ratio between a weight and the next or previous one.
The formula is : W = W(-1) * EMwmean.decay.
For example, with the value of 1.6 and 4 weights wanted, the relative importance of the weights will be 1 / 1.6 / 2.56 (=1.6*1.6) / 4.096 (=2.56*1.6) from the weakest to the strongest, and gives 0.11 / 0.17 / 0.275 / 0.445 considering that the sum of the weights is equal to one. The lower the EMwmean.decay, the smoother the differences between the weights enhancing a weak discrimination between models.

If EMwmean.decay = 'proportional', the weights are assigned to each model proportionally to their evaluation scores. The discrimination is fairer than using the decay method where close scores can have strongly diverging weights, while the proportional method would assign them similar weights.

It is also possible to define the EMwmean.decay parameter as a function that will be applied to single models scores and transform them into weights.
For example, if EMwmean.decay = function(x) {x^2}, the squared of evaluation score of each model will be used to weight the models predictions.

EMca

committee averaging over the selected models

Probabilities are first transformed into binary data according to the threshold defined when building the bm.out object with the BIOMOD_Modeling function (maximizing the evaluation metric score over the calibration dataset). The committee averaging score is obtained by taking the average of these binary predictions.
It is built on the analogy of a simple vote :

• each single model votes for the species being either present (1) or absent (0)

• the sum of 1 is then divided by the number of single models voting

The interesting feature of this measure is that it gives both a prediction and a measure of uncertainty. When the prediction is close to 0 or 1, it means that all models agree to predict 0 or 1 respectively. When the prediction is around 0.5, it means that half the models predict 1 and the other half 0.
Note that this is for binary data only.

EMci

confidence interval around the mean of probabilities of the selected models

It creates 2 ensemble models :

• LOWER : there is less than 100 * EMci.alpha / 2 % of chance to get probabilities lower than the given ones

• UPPER : there is less than 100 * EMci.alpha / 2 % of chance to get probabilities upper than the given ones

These intervals are calculated with the following function :

I_c = [ \bar{x} - \frac{t_\alpha sd }{ \sqrt{n} }; \bar{x} + \frac{t_\alpha sd }{ \sqrt{n} }]

EMcv

coefficient of variation (sd / mean) of probabilities over the selected models

This is the only ensemble model that might not be over the same scale than the others, as CV is a measure of uncertainty rather a measure of probability of occurrence. It will be evaluated like all other ensemble models although its interpretation will be obviously different. If the CV gets a high evaluation score, it means that the uncertainty is high where the species is observed (which might not be a good feature of the model). The lower is the score, the better are the models.

EMmode

mode of the predictions over the selected models

For multiclass and ordinal data, EMmode will return the most frequent class found for each point. This is the only ensemble model that will return categorical data and not numeric values.

EMfreq

mode frequency of the predictions over the selected models

For multiclass and ordinal data, EMfreq will return the frequency of the mode found for each point. This is a way of assessing the uncertainty between models: the higher the frequency, the lower the uncertainty.

Value

A BIOMOD.ensemble.models.out object containing models outputs, or links to saved outputs.
Models outputs are stored out of R (for memory storage reasons) in 2 different folders created in the current working directory :

1. a models folder, named after the resp.name argument of BIOMOD_FormatingData, and containing all ensemble models

2. a hidden folder, named .BIOMOD_DATA, and containing outputs related files (original dataset, calibration lines, pseudo-absences selected, predictions, variables importance, evaluation values...), that can be retrieved with get_[...] or load functions, and used by other biomod2 functions, like BIOMOD_EnsembleForecasting

Author(s)

Wilfried Thuiller, Damien Georges, Robin Engler

See Also

BIOMOD_FormatingData, bm_ModelingOptions, bm_CrossValidation, bm_VariablesImportance, BIOMOD_Modeling, BIOMOD_EnsembleForecasting, bm_PlotEvalMean, bm_PlotEvalBoxplot, bm_PlotVarImpBoxplot, bm_PlotResponseCurves

Other Main functions: BIOMOD_EnsembleForecasting(), BIOMOD_FormatingData(), BIOMOD_LoadModels(), BIOMOD_Modeling(), BIOMOD_Projection(), BIOMOD_RangeSize()

Examples

library(terra)
# Load species occurrences (6 species available)
data(DataSpecies)
head(DataSpecies)

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

# Get corresponding presence/absence data
myResp <- as.numeric(DataSpecies[, myRespName])

# Get corresponding XY coordinates
myRespXY <- DataSpecies[, c('X_WGS84', 'Y_WGS84')]

# Load environmental variables extracted from BIOCLIM (bio_3, bio_4, bio_7, bio_11 & bio_12)
data(bioclim_current)
myExpl <- terra::rast(bioclim_current)



## ----------------------------------------------------------------------- #
file.out <- paste0(myRespName, "/", myRespName, ".AllModels.models.out")
if (file.exists(file.out)) {
  myBiomodModelOut <- get(load(file.out))
} else {

  # Format Data with true absences
  myBiomodData <- BIOMOD_FormatingData(resp.name = myRespName,
                                       resp.var = myResp,
                                       resp.xy = myRespXY,
                                       expl.var = myExpl)

  # Model single models
  myBiomodModelOut <- BIOMOD_Modeling(bm.format = myBiomodData,
                                      modeling.id = 'AllModels',
                                      models = c('RF', 'GLM'),
                                      CV.strategy = 'random',
                                      CV.nb.rep = 2,
                                      CV.perc = 0.8,
                                      OPT.strategy = 'bigboss',
                                      metric.eval = c('TSS', 'AUCroc'),
                                      var.import = 3,
                                      seed.val = 42)
}

## ----------------------------------------------------------------------- #
# Model ensemble models
myBiomodEM <- BIOMOD_EnsembleModeling(bm.mod = myBiomodModelOut,
                                      models.chosen = 'all',
                                      em.by = 'all',
                                      em.algo = c('EMmean', 'EMca'),
                                      metric.select = c('TSS'),
                                      metric.select.thresh = c(0.7),
                                      metric.eval = c('TSS', 'AUCroc'),
                                      var.import = 3,
                                      seed.val = 42)
myBiomodEM

# Get evaluation scores & variables importance
get_evaluations(myBiomodEM)
get_variables_importance(myBiomodEM)

# Represent evaluation scores
bm_PlotEvalMean(bm.out = myBiomodEM, dataset = 'calibration')
bm_PlotEvalBoxplot(bm.out = myBiomodEM, group.by = c('algo', 'algo'))

# # Represent variables importance
# bm_PlotVarImpBoxplot(bm.out = myBiomodEM, group.by = c('expl.var', 'algo', 'algo'))
# bm_PlotVarImpBoxplot(bm.out = myBiomodEM, group.by = c('expl.var', 'algo', 'merged.by.PA'))
# bm_PlotVarImpBoxplot(bm.out = myBiomodEM, group.by = c('algo', 'expl.var', 'merged.by.PA'))

# # Represent response curves
# bm_PlotResponseCurves(bm.out = myBiomodEM, 
#                       models.chosen = get_built_models(myBiomodEM),
#                       fixed.var = 'median')
# bm_PlotResponseCurves(bm.out = myBiomodEM, 
#                       models.chosen = get_built_models(myBiomodEM),
#                       fixed.var = 'min')
# bm_PlotResponseCurves(bm.out = myBiomodEM, 
#                       models.chosen = get_built_models(myBiomodEM, algo = 'EMmean'),
#                       fixed.var = 'median',
#                       do.bivariate = TRUE)

Format input data, and select pseudo-absences if wanted, for usage in biomod2

Description

This function gathers together all input data needed (xy, presences/absences, explanatory variables, and the same for evaluation data if available) to run biomod2 models. It allows to select pseudo-absences if no absence data is available, with different strategies (see Details).

Usage

BIOMOD_FormatingData(
  resp.name,
  resp.var,
  resp.xy = NULL,
  expl.var,
  dir.name = ".",
  data.type = "binary",
  eval.resp.var = NULL,
  eval.resp.xy = NULL,
  eval.expl.var = NULL,
  PA.nb.rep = 0,
  PA.nb.absences = 1000,
  PA.strategy = NULL,
  PA.dist.min = 0,
  PA.dist.max = NULL,
  PA.sre.quant = 0.025,
  PA.fact.aggr = NULL,
  PA.user.table = NULL,
  na.rm = TRUE,
  filter.raster = FALSE,
  seed.val = NULL
)

Arguments

Details

This function gathers and formats all input data needed to run biomod2 models. It supports different kind of inputs (e.g. matrix, SpatVector, SpatRaster) and provides different methods to select pseudo-absences if needed.

Concerning explanatory variables and XY coordinates :

• if SpatRaster, RasterLayer or RasterStack provided for expl.var or eval.expl.var,
  biomod2 will extract the corresponding values from XY coordinates provided :

  • either through resp.xy or eval.resp.xy respectively

  • or through resp.var or eval.resp.var, if provided as SpatVector or SpatialPointsDataFrame

  Be sure to give the objects containing XY coordinates in the same projection system than the raster objects !

• if data.frame or matrix provided for expl.var or eval.expl.var,
  biomod2 will simply merge it (cbind) with resp.var without considering XY coordinates.
  Be sure to give explanatory and response values in the same row order !

Concerning pseudo-absence selection (see bm_PseudoAbsences) :
Only in the case of binary data !

• if both presence and absence data are available : PA.nb.rep = 0 and no pseudo-absence will be selected.

• if no absence data is available, several pseudo-absence repetitions are recommended (to estimate the effect of pseudo-absence selection), as well as high number of pseudo-absence points.
  Be sure not to select more pseudo-absence points than maximum number of pixels in the studied area !

• it is possible to create several pseudo-absence repetitions with different number of points, BUT with the same sampling strategy. PA.nb.absences must contain as many values as the number of sets of pseudo-absences (PA.nb.rep).
  
  
  
  

Response variable

biomod2 models single species at a time (no multi-species).
Hence, resp.var must be an uni-dimensional object, either :

• a vector, a one-column matrix or data.frame, a SpatVector (without associated data - if presence-only)

• a SpatialPoints (if presence-only)

• a SpatialPointsDataFrame or SpatVector object

If resp.var is a non-spatial object (vector, matrix or data.frame), XY coordinates must be provided through resp.xy.

Different data types are available, and require different values :

binary

1 : presences, 0 : true absences or NA : no information point (can be used to select pseudo-absences)
If no true absences are available, pseudo-absence selection must be done.

count

positive integer values

multiclass

factor values

ordinal

ordered factor values

relative

numeric values between 0 and 1

abundance

positive numeric values

Explanatory variables

Factorial variables are allowed, but might lead to some pseudo-absence strategy or models omissions (e.g. sre).

Evaluation data

Although biomod2 provides tools to automatically divide dataset into calibration and validation parts through the modeling process (see CV.[..] parameters in BIOMOD_Modeling function ; or bm_CrossValidation function), it is also possible (and strongly advised) to directly provide two independent datasets, one for calibration/validation and one for evaluation

Pseudo-absence selection (see bm_PseudoAbsences)

Only in the case of binary data !
If no true absences are available, pseudo-absences must be selected from the background data, meaning data there is no information whether the species of interest occurs or not. It corresponds either to the remaining pixels of the expl.var (if provided as a SpatRaster or RasterStack) or to the points identified as NA in resp.var (if expl.var provided as a matrix or data.frame).

Several methods are available to do this selection :

random

all points of initial background are pseudo-absence candidates. PA.nb.absences are drawn randomly, for each PA.nb.rep requested.

sre

pseudo-absences have to be selected in conditions (combination of explanatory variables) that differ in a defined proportion (PA.sre.quant) from those of presence points. A Surface Range Envelop model is first run over the species of interest (see bm_SRE), and pseudo-absences are selected outside this envelop.
This case is appropriate when all the species climatic niche has been sampled, otherwise it may lead to over-optimistic model evaluations and predictions !

disk

pseudo-absences are selected within circles around presence points defined by PA.dist.min and PA.dist.max distance values (in the same projection system units as coord and expl.var). It allows to select pseudo-absence points that are not too close to (avoid same niche and pseudo-replication) or too far (localized sampling strategy) from presences.

user.defined

pseudo-absences are defined in advance and given as data.frame through the PA.user.table parameter.

Value

A BIOMOD.formated.data or BIOMOD.formated.data.PA object that can be used to build species distribution model(s) with the BIOMOD_Modeling function.
print/show, plot and summary functions are available to have a summary of the created object.

Author(s)

Damien Georges, Wilfried Thuiller

See Also

bm_PseudoAbsences, BIOMOD_Modeling

Other Main functions: BIOMOD_EnsembleForecasting(), BIOMOD_EnsembleModeling(), BIOMOD_LoadModels(), BIOMOD_Modeling(), BIOMOD_Projection(), BIOMOD_RangeSize()

Examples

library(terra)

# Load species occurrences (6 species available)
data(DataSpecies)
head(DataSpecies)

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

# Get corresponding presence/absence data
myResp <- as.numeric(DataSpecies[, myRespName])

# Get corresponding XY coordinates
myRespXY <- DataSpecies[, c('X_WGS84', 'Y_WGS84')]

# Load environmental variables extracted from BIOCLIM (bio_3, bio_4, bio_7, bio_11 & bio_12)
data(bioclim_current)
myExpl <- terra::rast(bioclim_current)



# ---------------------------------------------------------------#
# Format Data with true absences
myBiomodData <- BIOMOD_FormatingData(resp.name = myRespName,
                                     resp.var = myResp,
                                     resp.xy = myRespXY,
                                     expl.var = myExpl)
myBiomodData
summary(myBiomodData)
plot(myBiomodData)


# ---------------------------------------------------------------#
# # Transform true absences into potential pseudo-absences
# myResp.PA <- ifelse(myResp == 1, 1, NA)
# 
# # Format Data with pseudo-absences : random method
# myBiomodData.r <- BIOMOD_FormatingData(resp.name = myRespName,
#                                        resp.var = myResp.PA,
#                                        resp.xy = myRespXY,
#                                        expl.var = myExpl,
#                                        PA.nb.rep = 4,
#                                        PA.nb.absences = 1000,
#                                        PA.strategy = 'random')
# 
# # Format Data with pseudo-absences : disk method
# myBiomodData.d <- BIOMOD_FormatingData(resp.name = myRespName,
#                                        resp.var = myResp.PA,
#                                        resp.xy = myRespXY,
#                                        expl.var = myExpl,
#                                        PA.nb.rep = 4,
#                                        PA.nb.absences = 500,
#                                        PA.strategy = 'disk',
#                                        PA.dist.min = 5,
#                                        PA.dist.max = 35)
# 
# # Format Data with pseudo-absences : SRE method
# myBiomodData.s <- BIOMOD_FormatingData(resp.name = myRespName,
#                                        resp.var = myResp.PA,
#                                        resp.xy = myRespXY,
#                                        expl.var = myExpl,
#                                        PA.nb.rep = 4,
#                                        PA.nb.absences = 1000,
#                                        PA.strategy = 'sre',
#                                        PA.sre.quant = 0.025)
# 
# # Format Data with pseudo-absences : user.defined method
# myPAtable <- data.frame(PA1 = ifelse(myResp == 1, TRUE, FALSE),
#                         PA2 = ifelse(myResp == 1, TRUE, FALSE))
# for (i in 1:ncol(myPAtable)) myPAtable[sample(which(myPAtable[, i] == FALSE), 500), i] = TRUE
# myBiomodData.u <- BIOMOD_FormatingData(resp.name = myRespName,
#                                        resp.var = myResp.PA,
#                                        resp.xy = myRespXY,
#                                        expl.var = myExpl,
#                                        PA.strategy = 'user.defined',
#                                        PA.user.table = myPAtable)
# 
# myBiomodData.r
# myBiomodData.d
# myBiomodData.s
# myBiomodData.u
# plot(myBiomodData.r)
# plot(myBiomodData.d)
# plot(myBiomodData.s)
# plot(myBiomodData.u)


# ---------------------------------------------------------------#
# # Select multiple sets of pseudo-absences
#
# # Transform true absences into potential pseudo-absences
# myResp.PA <- ifelse(myResp == 1, 1, NA)
# 
# # Format Data with pseudo-absences : random method
# myBiomodData.multi <- BIOMOD_FormatingData(resp.name = myRespName,
#                                            resp.var = myResp.PA,
#                                            resp.xy = myRespXY,
#                                            expl.var = myExpl,
#                                            PA.nb.rep = 4,
#                                            PA.nb.absences = c(1000, 500, 500, 200),
#                                            PA.strategy = 'random')
# myBiomodData.multi
# summary(myBiomodData.multi)
# plot(myBiomodData.multi)

Load species distribution models built with biomod2

Description

This function loads individual models built with BIOMOD_Modeling or BIOMOD_EnsembleModeling functions.

Usage

BIOMOD_LoadModels(
  bm.out,
  full.name = NULL,
  PA = NULL,
  run = NULL,
  algo = NULL,
  merged.by.PA = NULL,
  merged.by.run = NULL,
  merged.by.algo = NULL,
  filtered.by = NULL
)

Arguments

Details

This function might be of particular use to load models and make response plot analyses.

Running the function providing only bm.out argument will load all models built by the BIOMOD_Modeling or BIOMOD_EnsembleModeling function, but a subselection of models can be done using the additional arguments (full.name, PA, run, algo, merged.by.PA, merged.by.run, merged.by.algo, filtered.by).

Value

A vector containing the names of the loaded models.

Author(s)

Damien Georges

See Also

BIOMOD_Modeling, BIOMOD_EnsembleModeling

Other Main functions: BIOMOD_EnsembleForecasting(), BIOMOD_EnsembleModeling(), BIOMOD_FormatingData(), BIOMOD_Modeling(), BIOMOD_Projection(), BIOMOD_RangeSize()

Examples

library(terra)

# Load species occurrences (6 species available)
data(DataSpecies)
head(DataSpecies)

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

# Get corresponding presence/absence data
myResp <- as.numeric(DataSpecies[, myRespName])

# Get corresponding XY coordinates
myRespXY <- DataSpecies[, c('X_WGS84', 'Y_WGS84')]

# Load environmental variables extracted from BIOCLIM (bio_3, bio_4, bio_7, bio_11 & bio_12)
data(bioclim_current)
myExpl <- terra::rast(bioclim_current)



# ---------------------------------------------------------------
file.out <- paste0(myRespName, "/", myRespName, ".AllModels.models.out")
if (file.exists(file.out)) {
  myBiomodModelOut <- get(load(file.out))
} else {

  # Format Data with true absences
  myBiomodData <- BIOMOD_FormatingData(resp.name = myRespName,
                                       resp.var = myResp,
                                       resp.xy = myRespXY,
                                       expl.var = myExpl)

  # Model single models
  myBiomodModelOut <- BIOMOD_Modeling(bm.format = myBiomodData,
                                      modeling.id = 'AllModels',
                                      models = c('RF', 'GLM'),
                                      CV.strategy = 'random',
                                      CV.nb.rep = 2,
                                      CV.perc = 0.8,
                                      OPT.strategy = 'bigboss',
                                      metric.eval = c('TSS','AUCroc'),
                                      var.import = 3,
                                      seed.val = 42)
}


# ---------------------------------------------------------------
# Loading some models built
BIOMOD_LoadModels(bm.out = myBiomodModelOut, algo = 'RF')

Run a range of species distribution models

Description

This function allows to calibrate and evaluate a range of modeling techniques for a given species distribution. The dataset can be split up in calibration/validation parts, and the predictive power of the different models can be estimated using a range of evaluation metrics (see Details).

Usage

BIOMOD_Modeling(
  bm.format,
  modeling.id = as.character(format(Sys.time(), "%s")),
  models = c("ANN", "CTA", "FDA", "GAM", "GBM", "GLM", "MARS", "MAXENT", "MAXNET", "RF",
    "RFd", "SRE", "XGBOOST"),
  models.pa = NULL,
  CV.strategy = "random",
  CV.nb.rep = 1,
  CV.perc = NULL,
  CV.k = NULL,
  CV.balance = NULL,
  CV.env.var = NULL,
  CV.strat = NULL,
  CV.user.table = NULL,
  CV.do.full.models = TRUE,
  OPT.strategy = "default",
  OPT.user.val = NULL,
  OPT.user.base = "bigboss",
  OPT.user = NULL,
  metric.eval = c("KAPPA", "TSS", "AUCroc"),
  var.import = 0,
  weights = NULL,
  prevalence = 0.5,
  scale.models = FALSE,
  nb.cpu = 1,
  seed.val = NULL,
  do.progress = TRUE
)

Arguments

Details

bm.format

If pseudo absences have been added to the original dataset (see BIOMOD_FormatingData),
PA.nb.rep *(nb.rep + 1) models will be created.

models

The set of models to be calibrated on the data. 12 modeling techniques are currently available :

• ANN : Artificial Neural Network (nnet)

• CTA : Classification Tree Analysis (rpart)

• DNN : Deep Neural Network (cito)

• FDA : Flexible Discriminant Analysis (fda)

• GAM : Generalized Additive Model (gam, gam or bam)
  (see bm_ModelingOptions for details on algorithm selection)

• GBM : Generalized Boosting Model, or usually called Boosted Regression Trees (gbm)

• GLM : Generalized Linear Model (glm)

• MARS : Multiple Adaptive Regression Splines (earth)

• MAXENT : Maximum Entropy (see Maxent website)

• MAXNET : Maximum Entropy (maxnet)

• RF : Random Forest (randomForest)

• RFd : Random Forest downsampled (randomForest)

• SRE : Surface Range Envelop or usually called BIOCLIM (bm_SRE)

• XGBOOST : eXtreme Gradient Boosting Training (xgboost)

ANN

CTA

DNN

FDA

GAM

GBM

GLM

MARS

MAXENT

MAXNET

RF

RFd

SRE

XGBOOST

binary

x

x

x

x

x

x

x

x

x

x

x

x

x

x

multiclass

x

x

x

x

x

x

ordinal

x

x

x

x

x

x

x

x

abundance / count / relative

x

x

x

x

x

x

x

x

models.pa

Different models might respond differently to different numbers of pseudo-absences. It is possible to create sets of pseudo-absences with different numbers of points (see BIOMOD_FormatingData) and to assign only some of these datasets to each single model.

CV.[...] parameters

Different methods are available to calibrate/validate the single models (see bm_CrossValidation).

OPT.[...] parameters

Different methods are available to parameterize the single models (see bm_ModelingOptions and BIOMOD.options.dataset).

• default : only default parameter values of default parameters of the single models functions are retrieved. Nothing is changed so it might not give good results.

• bigboss : uses parameters pre-defined by biomod2 team and that are available in the dataset OptionsBigboss.
  to be optimized in near future

• user.defined : updates default or bigboss parameters with some parameters values defined by the user (but matching the format of a BIOMOD.models.options object)

• tuned : calling the bm_Tuning function to try and optimize some default values

metric.eval

Please refer to CAWRC website ("Methods for dichotomous forecasts") to get detailed description (simple/complex metrics).
Several evaluation metrics can be selected.
Optimal value of each method can be obtained with the get_optim_value function.

simple

• POD : Probability of detection (hit rate)

• FAR : False alarm ratio

• POFD : Probability of false detection (fall-out)

• SR : Success ratio

• ACCURACY : Accuracy (fraction correct)

• BIAS : Bias score (frequency bias)

complex

• AUCroc : Area Under Curve of Relative operating characteristic

• AUCprg : Area Under Curve of Precision-Recall-Gain curve

• TSS : True skill statistic (Hanssen and Kuipers discriminant, Peirce's skill score)

• KAPPA : Cohen's Kappa (Heidke skill score)

• OR : Odds Ratio

• ORSS : Odds ratio skill score (Yule's Q)

• CSI : Critical success index (threat score)

• ETS : Equitable threat score (Gilbert skill score)

presence-only

• BOYCE : Boyce index

• MPA : Minimal predicted area (cutoff optimizing MPA to predict 90% of presences)

abundance / count / relative data

• RMSE : Root Mean Square Error

• MSE : Mean Square Error

• MAE : Mean Absolute Error

• Rsquared : R squared

• Rsquared_aj : R squared adjusted

• Max_error : Maximum error

multiclass/ordinal data

• Accuracy : Accuracy

• Recall : Macro average Recall

• Precision : Macro average Precision

• F1 : Macro F1 score

Results after modeling can be obtained through the get_evaluations function.
Evaluation metric are calculated on the calibrating data (column calibration), on the cross-validation data (column validation) or on the evaluation data (column evaluation).
For cross-validation data, see CV.[...] parameters in BIOMOD_Modeling function.
For evaluation data, see eval.[...] parameters in BIOMOD_FormatingData.

var.import

A value characterizing how much each variable has an impact on each model predictions can be calculated by randomizing the variable of interest and computing the correlation between original and shuffled variables (see bm_VariablesImportance).

weights & prevalence

More or less weight can be given to some specific observations.
Automatically created weights will be integer values to prevent some modeling issues.
Note that MAXENT, MAXNET, RF, RFd and SRE models do not take weights into account.

• If prevalence = 0.5 (the default), presences and absences will be weighted equally (i.e. the weighted sum of presences equals the weighted sum of absences).

• If prevalence is set below (above) 0.5, more weight will be given to absences (presences).

• If weights is defined, prevalence argument will be ignored (EXCEPT for MAXENT).

scale.models

A binomial GLM is created to scale predictions from 0 to 1.
SRE is never scaled, and ANN and FDA categorical models always are.
Note that it may lead to reduction in projected scale amplitude.
This parameter is quite experimental and it is recommended not to use it. It was developed in the idea to ensure comparable predictions by removing the scale prediction effect (the more extended projections are, the more they influence ensemble forecasting results).

Value

A BIOMOD.models.out object containing models outputs, or links to saved outputs.
Models outputs are stored out of R (for memory storage reasons) in 2 different folders created in the current working directory :

1. a models folder, named after the resp.name argument of BIOMOD_FormatingData, and containing all calibrated models for each repetition and pseudo-absence run

2. a hidden folder, named .BIOMOD_DATA, and containing outputs related files (original dataset, calibration lines, pseudo-absences selected, predictions, variables importance, evaluation values...), that can be retrieved with get_[...] or load functions, and used by other biomod2 functions, like BIOMOD_Projection or BIOMOD_EnsembleModeling

Author(s)

Wilfried Thuiller, Damien Georges, Robin Engler

See Also

glm, gam, gam, bam, gbm, rpart, nnet, cito, fda, earth, randomForest, maxnet, xgboost, BIOMOD_FormatingData, bm_ModelingOptions, bm_Tuning, bm_CrossValidation, bm_VariablesImportance, BIOMOD_Projection, BIOMOD_EnsembleModeling, bm_PlotEvalMean, bm_PlotEvalBoxplot, bm_PlotVarImpBoxplot, bm_PlotResponseCurves

Other Main functions: BIOMOD_EnsembleForecasting(), BIOMOD_EnsembleModeling(), BIOMOD_FormatingData(), BIOMOD_LoadModels(), BIOMOD_Projection(), BIOMOD_RangeSize()

Examples

library(terra)

# Load species occurrences (6 species available)
data(DataSpecies)
head(DataSpecies)

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

# Get corresponding presence/absence data
myResp <- as.numeric(DataSpecies[, myRespName])

# Get corresponding XY coordinates
myRespXY <- DataSpecies[, c('X_WGS84', 'Y_WGS84')]

# Load environmental variables extracted from BIOCLIM (bio_3, bio_4, bio_7, bio_11 & bio_12)
data(bioclim_current)
myExpl <- terra::rast(bioclim_current)



# ---------------------------------------------------------------------------- #
# Format Data with true absences
myBiomodData <- BIOMOD_FormatingData(resp.name = myRespName,
                                     resp.var = myResp,
                                     resp.xy = myRespXY,
                                     expl.var = myExpl)


# ---------------------------------------------------------------------------- #
# Model single models
myBiomodModelOut <- BIOMOD_Modeling(bm.format = myBiomodData,
                                    modeling.id = 'AllModels',
                                    models = c('RF', 'GLM'),
                                    CV.strategy = 'random',
                                    CV.nb.rep = 2,
                                    CV.perc = 0.8,
                                    OPT.strategy = 'bigboss',
                                    metric.eval = c('TSS','AUCroc'),
                                    var.import = 2,
                                    seed.val = 42)
myBiomodModelOut

# Get evaluation scores & variables importance
get_evaluations(myBiomodModelOut)
get_variables_importance(myBiomodModelOut)

# Represent evaluation scores 
bm_PlotEvalMean(bm.out = myBiomodModelOut, dataset = 'calibration')
bm_PlotEvalMean(bm.out = myBiomodModelOut, dataset = 'validation')
bm_PlotEvalBoxplot(bm.out = myBiomodModelOut, group.by = c('algo', 'run'))

# # Represent variables importance 
# bm_PlotVarImpBoxplot(bm.out = myBiomodModelOut, group.by = c('expl.var', 'algo', 'algo'))
# bm_PlotVarImpBoxplot(bm.out = myBiomodModelOut, group.by = c('expl.var', 'algo', 'run'))
# bm_PlotVarImpBoxplot(bm.out = myBiomodModelOut, group.by = c('algo', 'expl.var', 'run'))

# # Represent response curves 
# mods <- get_built_models(myBiomodModelOut, run = 'RUN1')
# bm_PlotResponseCurves(bm.out = myBiomodModelOut, 
#                       models.chosen = mods,
#                       fixed.var = 'median')
# bm_PlotResponseCurves(bm.out = myBiomodModelOut, 
#                       models.chosen = mods,
#                       fixed.var = 'min')
# mods <- get_built_models(myBiomodModelOut, full.name = 'GuloGulo_allData_RUN2_RF')
# bm_PlotResponseCurves(bm.out = myBiomodModelOut, 
#                       models.chosen = mods,
#                       fixed.var = 'median',
#                       do.bivariate = TRUE)

BIOMOD_ModelingOptions

Description

Deprecated function name for bm_ModelingOptions

Usage

BIOMOD_ModelingOptions(...)

Arguments

See Also

biomod2-deprecated

BIOMOD_PresenceOnly

Description

Deprecated function name for bm_FindOptimStat

Usage

BIOMOD_PresenceOnly(...)

Arguments

See Also

biomod2-deprecated

Project a range of calibrated species distribution models onto new environment

Description

This function allows to project a range of models built with the BIOMOD_Modeling function onto new environmental data (which can represent new areas, resolution or time scales for example).

Usage

BIOMOD_Projection(
  bm.mod,
  proj.name,
  new.env,
  new.env.xy = NULL,
  models.chosen = "all",
  metric.binary = NULL,
  metric.filter = NULL,
  build.clamping.mask = TRUE,
  nb.cpu = 1,
  seed.val = NULL,
  ...
)

Arguments

Details

If models.chosen = 'all', projections are done for all calibration and pseudo absences runs if applicable.
These projections may be used later by the BIOMOD_EnsembleForecasting function.

If build.clamping.mask = TRUE, a raster file will be saved within the projection folder. This mask values will correspond to the number of variables in each pixel that are out of their calibration / validation range, identifying locations where predictions are uncertain.

... can take the following values :

omit.na

(optional, default TRUE) :
a logical value defining whether all not fully referenced environmental points will get NA as predictions or not

digits

(optional, default 0) :
an integer value corresponding to the number of digits of the predictions

on_0_1000

(optional, default TRUE) :
a logical value defining whether 0 - 1 probabilities are to be converted to 0 - 1000 scale to save memory on backup

keep.in.memory

(optional, default TRUE) :
a logical value defining whether all projections are to be kept loaded at once in memory, or only links pointing to hard drive are to be returned

do.stack

(optional, default TRUE) :
a logical value defining whether all projections are to be saved as one SpatRaster object or several SpatRaster files (the default if projections are too heavy to be all loaded at once in memory)

output.format

(optional, default .RData or .tif) :
a character value corresponding to the projections saving format on hard drive, must be either .grd, .img, .tif or .RData (the default if new.env is given as matrix or data.frame)

compress

(optional, default TRUE) :
a logical or a character value defining whether and how objects should be compressed when saved on hard drive. Must be either TRUE, FALSE, gzip (for Windows OS) or xz (for other OS)

overwrite

(optional, default do.stack) :
a logical defining whether pre-existing projections with same modeling ID and project name should be replaced or not

Value

A BIOMOD.projection.out object containing models projections, or links to saved outputs.
Models projections are stored out of R (for memory storage reasons) in proj.name folder created in the current working directory :

1. the output is a data.frame if new.env is a matrix or a data.frame

2. it is a SpatRaster if new.env is a SpatRaster (or several SpatRaster objects, if new.env is too large)

3. raw projections, as well as binary and filtered projections (if asked), are saved in the proj.name folder

Author(s)

Wilfried Thuiller, Damien Georges

See Also

BIOMOD_Modeling, BIOMOD_EnsembleModeling, BIOMOD_RangeSize

Other Main functions: BIOMOD_EnsembleForecasting(), BIOMOD_EnsembleModeling(), BIOMOD_FormatingData(), BIOMOD_LoadModels(), BIOMOD_Modeling(), BIOMOD_RangeSize()

Examples

library(terra)

# Load species occurrences (6 species available)
data(DataSpecies)
head(DataSpecies)

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

# Get corresponding presence/absence data
myResp <- as.numeric(DataSpecies[, myRespName])

# Get corresponding XY coordinates
myRespXY <- DataSpecies[, c('X_WGS84', 'Y_WGS84')]

# Load environmental variables extracted from BIOCLIM (bio_3, bio_4, bio_7, bio_11 & bio_12)
data(bioclim_current)
myExpl <- terra::rast(bioclim_current)



# ---------------------------------------------------------------#
file.out <- paste0(myRespName, "/", myRespName, ".AllModels.models.out")
if (file.exists(file.out)) {
  myBiomodModelOut <- get(load(file.out))
} else {

  # Format Data with true absences
  myBiomodData <- BIOMOD_FormatingData(resp.name = myRespName,
                                       resp.var = myResp,
                                       resp.xy = myRespXY,
                                       expl.var = myExpl)

  # Model single models
  myBiomodModelOut <- BIOMOD_Modeling(bm.format = myBiomodData,
                                      modeling.id = 'AllModels',
                                      models = c('RF', 'GLM'),
                                      CV.strategy = 'random',
                                      CV.nb.rep = 2,
                                      CV.perc = 0.8,
                                      OPT.strategy = 'bigboss',
                                      metric.eval = c('TSS','AUCroc'),
                                      var.import = 3,
                                      seed.val = 42)
}


# ---------------------------------------------------------------#
# Project single models
file.proj <- paste0(myRespName, "/proj_Current/", myRespName, ".Current.projection.out")
if (file.exists(file.proj)) {
  myBiomodProj <- get(load(file.proj))
} else {
myBiomodProj <- BIOMOD_Projection(bm.mod = myBiomodModelOut,
                                  proj.name = 'Current',
                                  new.env = myExpl,
                                  models.chosen = 'all')
}
myBiomodProj
plot(myBiomodProj)

Analyze the range size differences between projections of species distribution models

Description

This function allows to calculate the absolute number of locations (pixels) lost, stable and gained, as well as the corresponding relative proportions, between two (or more) binary projections of (ensemble) species distribution models (which can represent new time scales or environmental scenarios for example).

Usage

BIOMOD_RangeSize(
  proj.current,
  proj.future,
  models.chosen = "all",
  metric.binary = NULL
)

Arguments

Value

A BIOMOD.rangesize.out containing principaly two objects :

Compt.By.Species

a data.frame containing the summary of range change for each comparison

• Loss : number of pixels predicted to be lost

• Stable_Abs : number of pixels not currently occupied and not predicted to be

• Stable_Pres : number of pixels currently occupied and predicted to remain occupied

• Gain : number of pixels predicted to be gained

• PercLoss : percentage of pixels currently occupied and predicted to be lost (Loss / (Loss + Stable_Pres))

• PercGain : percentage of pixels predicted to be gained compare to the number of pixels currently occupied (Gain / (Loss + Stable_Pres))

• SpeciesRangeChange : percentage of pixels predicted to change (loss or gain) compare to the number of pixels currently occupied (PercGain - PercLoss)

• CurrentRangeSize : number of pixels currently occupied

• FutureRangeSize0Disp : number of pixels predicted to be occupied, assuming no migration

• FutureRangeSize1Disp : number of pixels predicted to be occupied, assuming migration

loss.gain

an object in the same form than the input data (proj.current and proj.future) and containing a value for each point/pixel of each comparison among :

• -2 : predicted to be lost

• -1 : predicted to remain occupied

• 0 : predicted to remain unoccupied

• 1 : predicted to be gained

Diff.By.Pixel

an object in the same form than the input data (proj.current and proj.future) and containing a value for each point/pixel of each comparison obtain with :

• Future - 2* Current for binary data

• Future - Current for nonbinary after rescaling Future and Current from 0 to 1.

Author(s)

Maya Guéguen, Hélène Blancheteau

See Also

BIOMOD_Projection, BIOMOD_EnsembleForecasting, bm_PlotRangeSize

Other Main functions: BIOMOD_EnsembleForecasting(), BIOMOD_EnsembleModeling(), BIOMOD_FormatingData(), BIOMOD_LoadModels(), BIOMOD_Modeling(), BIOMOD_Projection()

Examples

library(terra)

# Load species occurrences (6 species available)
data(DataSpecies)
head(DataSpecies)

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

# Get corresponding presence/absence data
myResp <- as.numeric(DataSpecies[, myRespName])

# Get corresponding XY coordinates
myRespXY <- DataSpecies[, c('X_WGS84', 'Y_WGS84')]

# Load environmental variables extracted from BIOCLIM (bio_3, bio_4, bio_7, bio_11 & bio_12)
data(bioclim_current)
myExpl <- terra::rast(bioclim_current)



# --------------------------------------------------------------- #
file.out <- paste0(myRespName, "/", myRespName, ".AllModels.models.out")
if (file.exists(file.out)) {
  myBiomodModelOut <- get(load(file.out))
} else {

  # Format Data with true absences
  myBiomodData <- BIOMOD_FormatingData(resp.name = myRespName,
                                       resp.var = myResp,
                                       resp.xy = myRespXY,
                                       expl.var = myExpl)

  # Model single models
  myBiomodModelOut <- BIOMOD_Modeling(bm.format = myBiomodData,
                                      modeling.id = 'AllModels',
                                      models = c('RF', 'GLM'),
                                      CV.strategy = 'random',
                                      CV.nb.rep = 2,
                                      CV.perc = 0.8,
                                      OPT.strategy = 'bigboss',
                                      metric.eval = c('TSS', 'AUCroc'),
                                      var.import = 3,
                                      seed.val = 42)
}

models.proj <- get_built_models(myBiomodModelOut, algo = "RF")
  # Project single models
  myBiomodProj <- BIOMOD_Projection(bm.mod = myBiomodModelOut,
                                    proj.name = 'CurrentRangeSize',
                                    new.env = myExpl,
                                    models.chosen = models.proj,
                                    metric.binary = 'all',
                                    build.clamping.mask = TRUE)


# --------------------------------------------------------------- #
# Load environmental variables extracted from BIOCLIM (bio_3, bio_4, bio_7, bio_11 & bio_12)
data(bioclim_future)
myExplFuture <- terra::rast(bioclim_future)

# Project onto future conditions
myBiomodProjectionFuture <- BIOMOD_Projection(bm.mod = myBiomodModelOut,
                                              proj.name = 'FutureRangeSize',
                                              new.env = myExplFuture,
                                              models.chosen = models.proj,
                                              metric.binary = 'TSS')
                                              
# Compute differences
myBiomodRangeSize <- BIOMOD_RangeSize(proj.current = myBiomodProj,
                                      proj.future = myBiomodProjectionFuture,
                                      metric.binary = "TSS")


# Represent main results
bm_PlotRangeSize(bm.range = myBiomodRangeSize)

Produce summary outputs from a simulation folder

Description

This function allows to produce summary report or ODMAP table from a biomod2 simulation folder.

Usage

BIOMOD_Report(
  bm.out,
  strategy = "report",
  params.color = list(color1 = "#eb4034", color2 = "#e0918b", color3 = "#658f70"),
  params.ODMAP = list(O.mod.objective = NULL, O.boundary = NULL, O.obs.type = NULL,
    O.pred.type = NULL, D.eco.level = NULL, D.samp.design = NULL)
)

Arguments

Details

This function gathers and formats all objects contained in one biomod2 modeling folder to produce, based on Rmarkdown templates, standardized reports to help the user :

• summarize its modeling results

• share them through standardized informations through ODMAP protocol

• provide reproducible code
  
  

Type of report

Different data types are available, and require different values :

report

biomod2 provides functions to summarize the information, such as print, plot or summary methods adapted to BIOMOD.[...].out objects, as well as get_[...] and bm_Plot[...] functions. All these are called here and applied to objects contained in the provided modeling folder.

ODMAP

following Zurell et al. 2020, ODMAP (Overview, Data, Model, Assessment and Prediction) protocol aims to standardize documentation of modeling to help improve both transparency and reproducibility of results. ODMAP v1.0 website provides an application to fill this type of report. biomod2 tries here to help one user to pre-fill the fields of this protocol.

code

call slot contained within BIOMOD.formated.data, BIOMOD.models.out, BIOMOD.projection.out and BIOMOD.ensemble.models.out objects keep in memory the R command used to obtain them. All these calls are gathered here in one summary file.

Value

A standardized .html file obtained from an Rmarkdown template, and a .csv table in the case of ODMAP report.

Author(s)

Maya Guéguen

References

• Zurell D, Franklin J, König C, Bouchet PJ, Serra-Diaz JM, Dormann CF, Elith J, Fandos Guzman G, Feng X, Guillera-Arroita G, Guisan A, Leitão PJ, Lahoz-Monfort JJ, Park DS, Peterson AT, Rapacciuolo G, Schmatz DR, Schröder B, Thuiller W, Yates KL, Zimmermann NE, Merow C (2020) A standard protocol for describing species distribution models. Ecography 43: 1261-1277. doi:10.1111/ecog.04960

See Also

ODMAP, match.call

Examples

library(terra)

BIOMOD_Tuning

Description

Deprecated function name for bm_Tuning

Usage

BIOMOD_Tuning(...)

Arguments

See Also

biomod2-deprecated

BinaryTransformation

Description

Deprecated function name for bm_BinaryTransformation

Usage

BinaryTransformation(...)

Arguments

See Also

biomod2-deprecated

Abundance to build test SDM

Description

A dataset covering France with abundance data for 20 bird species. Data comes from the French Breeding Bird Survey (STOC program) of Vigie-Nature (Fontaine et al. 2020). The original dataset contains 2,948 sites in which bird species have been observed from 2001 to 2024. Original raw data is not provided here, but might be made available on request by contacting Benoît Fontaine (benoit.fontaine [at] mnhn.fr).

Usage

DataSTOC

Format

A data.frame object with 2006 rows and 30 columns (10 variables and 20 species):

site

Observation site ID

period

yearly average headcounts have been computed over 2 times periods: 2006-2011 and 2012-2017

X_WGS84

Longitude

Y_WGS84

Latitude

temp

annual mean temperature (CHELSA)

precip

annual mean precipitation (CHELSA)

cover_agri

percentage of agricultural habitat cover around each point (European Union’s Copernicus Land Monitoring Service information)

cover_water

percentage of water habitat cover around each point (European Union’s Copernicus Land Monitoring Service information)

cover_wet

percentage of wetland habitat cover around each point (European Union’s Copernicus Land Monitoring Service information)

sdiv_hab

Shannon habitat diversity (CORINE)

Alauda.arvensis

abundance data for Eurasian skylark

Cettia.cetti

abundance data for Cetti's warbler

Coccothraustes.coccothraustes

abundance data for hawfinch

Cuculus.canorus

abundance data for common cuckoo

Emberiza.calandra

abundance data for corn bunting

Emberiza.citrinella

abundance data for yellowhammer

Erithacus.rubecula

abundance data for European robin

Fulica.atra

abundance data for Eurasian coot

Luscinia.megarhynchos

abundance data for common nightingale

Passer.domesticus

abundance data for house sparrow

Perdix.perdix

abundance data for grey partridge

Periparus.ater

abundance data for coal tit

Phylloscopus.bonelli

abundance data for western Bonelli's warbler

Phylloscopus.trochilus

abundance data for willow warbler

Regulus.regulus

abundance data for goldcrest

Serinus.serinus

abundance data for European serin

Sitta.europaea

abundance data for Eurasian nuthatch

Streptopelia.decaocto

abundance data for Eurasian collared dove

Sylvia.melanocephala

abundance data for Sardinian warbler

Turdus.philomelos

abundance data for song thrush

References

• STOC EPS - Vigie-Nature (2025). French Breeding Bird Monitoring Scheme. Muséum National d’Histoire Naturelle (MNHN) - Office Français pour la Biodiversité (OFB) - Ligue pour la Protection des Oiseaux (LPO).

• Fontaine B, Moussy C, Chiffard Carricaburu J, Dupuis J, Corolleur E, Schmaltz L, Lorrillière R, Loïs G, Gaudard C. (2020) Suivi des oiseaux communs en France 1989-2019 : 30 ans de suivis participatifs. MNHN - Centre d'Ecologie et des Sciences de la Conservation, LPO BirdLife France - Service Connaissance, Ministère de la Transition écologique et solidaire. 46 pp.

• Brun P, Zimmermann NE, Hari C, Pellissier L, Karger DN. (2022) CHELSA-BIOCLIM+ A novel set of global climate-related predictors at kilometre-resolution. EnviDat. DOI: 10.16904/envidat.332

• CORINE Land Cover 2018 (raster 100 m), Europe, 6-yearly - version 2020_20u1, May 2020. European Environment Agency. DOI: doi:10.2909/960998c1-1870-4e82-8051-6485205ebbac

Presence-Absence data to build test SDM

Description

A dataset covering all the continent with presence/absence data for 6 mammal species. Presence/absence were derived from range maps downloaded at IUCN.

Usage

DataSpecies

Format

A data.frame object with 2488 rows and 10 variables:

X_WGS84

Longitude

Y_WGS84

Latitude

ConnochaetesGnou

Presence (1) or Absence (0) for black wildebeest

GuloGulo

Presence (1) or Absence (0) for wolverine

PantheraOnca

Presence (1) or Absence (0) for jaguar

PteropusGiganteus

Presence (1) or Absence (0) for Indian flying fox

TenrecEcaudatus

Presence (1) or Absence (0) for tailless tenrec

VulpesVulpes

Presence (1) or Absence (0) for red fox

Find.Optim.Stat

Description

Deprecated function name for bm_FindOptimStat

Usage

Find.Optim.Stat(...)

Arguments

See Also

biomod2-deprecated

Single models package and functions

Description

A data.frame containing for each single model available in biomod2 the package and functions to be called.

Usage

ModelsTable

Format

A data.frame object with 12 rows and 5 variables:

model

all single models that can be computed in biomod2

type

data type associated to the models

package

R package used

func

function used in the R package

train

function called by caret for the tuning

All single models available are the following :

• ANN (nnet)

• CTA (rpart)

• DNN (cito)

• FDA (fda)

• GAM (gam, gam or bam)

• GBM (gbm)

• GLM (glm)

• MARS (earth)

• MAXENT (see Maxent website)

• MAXNET (maxnet)

• RF (randomForest)

• RFd (randomForest downsampled)

• SRE (bm_SRE)

• XGBOOST (xgboost)

ODMAP empty table

Description

A data.frame containing ODMAP (Overview, Data, Model, Assessment, Prediction) protocol components.

Usage

ODMAP

Format

A data.frame object with 84 rows and 4 variables:

section

Overview, Data, Model, Assessment or Prediction step

subsection

corresponding field

element

corresponding field

value

to be filled with BIOMOD_Report function

Bigboss pre-defined parameter values for single models

Description

A BIOMOD.models.options object containing for each single model available in biomod2 the parameter values pre-defined by biomod2 team.

Usage

OptionsBigboss

Format

A BIOMOD.models.options object with some changed values :

ANN.nnet.nnet

• size = 5

• decay = 0.1

• trace = FALSE

• rang = 0.1

• maxit = 200

CTA.rpart.rpart

• control = list(xval = 5, minbucket = 5, minsplit = 5, cp = 0.001, maxdepth = 10)

• cost = NULL

DNN.cito.dnn

• batchsize = 100L

• epochs = 150L

• hidden = c(100L, 100L)

• lr = 0.05

• optimizer = "adam"

• lambda = 0.001

• alpha = 1.0

• validation = 0.2

• lr_scheduler = config_lr_scheduler("reduce_on_plateau", patience = 7)

• early_stopping = 14

FDA.mda.fda

GAM.gam.gam

GAM.mgcv.bam

GAM.mgcv.gam

• method = 'GCV.Cp'

• control = list(epsilon = 1e-06, trace = FALSE, maxit = 100)

GBM.gbm.gbm

• n.trees = 2500

• interaction.depth = 7

• n.minobsinnode = 5

• shrinkage = 0.001

• cv.folds = 3

• keep.data = FALSE

• n.cores = 1

GLM.stats.glm

• mustart = 0.5

• control = glm.control(maxit = 50)

MARS.earth.earth

• ncross = 0

• nk = NULL

• penalty = 2

• thresh = 0.001

• nprune = NULL

• pmethod = 'backward'

MAXENT.MAXENT.MAXENT

• path_to_maxent.jar = '.'

RF.randomForest.randomForest

• ntree = 500

• mtry = 2

• sampsize = NULL

• nodesize = 5

• maxnodes = NULL

RFd.randomForest.randomForest

• type = 'classification'

• ntree = 500

• mtry = 2

• strata = factor(c(0, 1))

• sampsize = NULL

• nodesize = 5

• maxnodes = NULL

SRE.biomod2.bm_SRE

• do.extrem = TRUE

XGBOOST.xgboost.xgboost

• params = list(max_depth = 2, eta = 1)

• nthread = 2

• nrounds = 4

ProbDensFunc

Description

Deprecated function name for bm_PlotRangeSize

Usage

ProbDensFunc(...)

Arguments

See Also

biomod2-deprecated

Bioclimatic variables for SDM based on current condition

Description

A SpatRaster with 5 bioclimatic variables commonly used for SDM and describing current climate. Additional information available at worldclim.

Usage

bioclim_current

Format

A SpatRaster with 5 layers:

bio3

Isothermality

bio4

Temperature Seasonality

bio7

Temperature Annual Range

bio11

Mean Temperature of Coldest Quarter

bio12

Annual Precipitation

Bioclimatic variables for SDM based on future condition

Description

A SpatRaster with 5 bioclimatic variables commonly used for SDM and describing future climate based on old RCP scenarios at the horizon 2080.

Usage

bioclim_future

Format

A SpatRaster with 5 layers:

bio3

Isothermality

bio4

Temperature Seasonality

bio7

Temperature Annual Range

bio11

Mean Temperature of Coldest Quarter

bio12

Annual Precipitation

Deprecated functions in package biomod2.

Description

The functions listed below are deprecated and were removed in the current version. When possible, alternative functions with similar functionality are mentioned. Help pages for deprecated functions are available at help("<function>-deprecated").

Usage

BIOMOD_CrossValidation(...)

BIOMOD_ModelingOptions(...)

BIOMOD_PresenceOnly(...)

BIOMOD_Tuning(...)

BinaryTransformation(...)

calculate.stat(...)

Find.Optim.Stat(...)

getStatOptimValue(...)

makeFormula(...)

models_scores_graph(...)

ProbDensFunc(...)

response.plot2(...)

sample.factor.levels(...)

sre(...)

.transform.outputs.list(...)

variables_importance(...)

BIOMOD_CrossValidation

For BIOMOD_CrossValidation use bm_CrossValidation.

BIOMOD_ModelingOptions

For BIOMOD_ModelingOptions use bm_ModelingOptions.

BIOMOD_PresenceOnly

For BIOMOD_PresenceOnly use bm_FindOptimStat.

BIOMOD_Tuning

For BIOMOD_Tuning use bm_Tuning.

BinaryTransformation

For BinaryTransformation use bm_BinaryTransformation.

calculate.stat

For calculate.stat use bm_CalculateStatBin.

Find.Optim.Stat

For Find.Optim.Stat use bm_FindOptimStat.

getStatOptimValue

For getStatOptimValue use get_optim_value.

makeFormula

For makeFormula use bm_MakeFormula.

models_scores_graph

For models_scores_graph use bm_PlotEvalMean.

ProbDensFunc

For ProbDensFunc use bm_PlotRangeSize.

response.plot2

For response.plot2 use bm_PlotResponseCurves.

sample.factor.levels

For sample.factor.levels use bm_SampleFactorLevels.

sre

For sre use bm_SRE.

.transform.outputs.list

For .transform.outputs.list use .transform_outputs_list.

variables_importance

For variables_importance use bm_VariablesImportance.

Ensemble model output object class (when running BIOMOD_EnsembleModeling())

Description

Class created by BIOMOD_EnsembleModeling

Usage

## S4 method for signature 'biomod2_ensemble_model'
show(object)

Arguments

Details

biomod2_model is the basic object for biomod2 ensemble species distribution models.
All listed classes below are derived from biomod2_model, and have a model_class slot specific value :

• biomod2_ensemble_model : model_class is EM

• EMmean_biomod2_model : model_class is EMmean

• EMmedian_biomod2_model : model_class is EMmedian

• EMcv_biomod2_model : model_class is EMcv

• EMci_biomod2_model : model_class is EMci

• EMca_biomod2_model : model_class is EMca

• EMwmean_biomod2_model : model_class is EMwmean

• EMmode_biomod2_model : model_class is EMmode

• EMfreq_biomod2_model : model_class is EMfreq

Slots

modeling.id

a character corresponding to the name (ID) of the simulation set

model_name

a character corresponding to the model name

model_class

a character corresponding to the model class

model_options

a list containing the model options

model

the corresponding model object

scaling_model

the corresponding scaled model object

dir_name

a character corresponding to the modeling folder

resp_name

a character corresponding to the species name

expl_var_names

a vector containing names of explanatory variables

expl_var_type

a vector containing classes of explanatory variables

expl_var_range

a list containing ranges of explanatory variables

model_evaluation

a data.frame containing the model evaluations

model_variables_importance

a data.frame containing the model variables importance

Author(s)

Damien Georges

See Also

biomod2_model, BIOMOD_EnsembleModeling

Other Toolbox objects: BIOMOD.ensemble.models.out, BIOMOD.formated.data, BIOMOD.formated.data.PA, BIOMOD.models.options, BIOMOD.models.out, BIOMOD.options.dataset, BIOMOD.options.default, BIOMOD.projection.out, BIOMOD.rangesize.out, BIOMOD.stored.data, biomod2_model

Examples

showClass("biomod2_ensemble_model")
showClass("EMmean_biomod2_model")
showClass("EMmedian_biomod2_model")
showClass("EMcv_biomod2_model")
showClass("EMci_biomod2_model")
showClass("EMca_biomod2_model")
showClass("EMwmean_biomod2_model")
showClass("EMmode_biomod2_model")
showClass("EMfreq_biomod2_model")

Single model output object class (when running BIOMOD_Modeling())

Description

Class created by BIOMOD_Modeling and bm_RunModel

Usage

## S4 method for signature 'biomod2_model'
show(object)

Arguments

Details

biomod2_model is the basic object for biomod2 single species distribution models.
All listed classes below are derived from biomod2_model, and have a model_class slot specific value :

• ANN_biomod2_model : model_class is ANN

• CTA_biomod2_model : model_class is CTA

• DNN_biomod2_model : model_class is DNN

• FDA_biomod2_model : model_class is FDA

• GBM_biomod2_model : model_class is GBM

• GLM_biomod2_model : model_class is GLM

• MARS_biomod2_model : model_class is MARS

• MAXENT_biomod2_model : model_class is MAXENT

• MAXNET_biomod2_model : model_class is MAXNET

• RF_biomod2_model : model_class is RF

• RFd_biomod2_model : model_class is RFd

• SRE_biomod2_model : model_class is SRE

Slots

model_name

a character corresponding to the model name

model_class

a character corresponding to the model class

model_type

a character corresponding to the type of data

model_options

a list containing the model options

model

the corresponding model object

scaling_model

the corresponding scaled model object

levels_factor

the levels of the original data for ordinal and multiclass

dir_name

a character corresponding to the modeling folder

resp_name

a character corresponding to the species name

expl_var_names

a vector containing names of explanatory variables

expl_var_type

a vector containing classes of explanatory variables

expl_var_range

a list containing ranges of explanatory variables

model_evaluation

a data.frame containing the model evaluations

model_variables_importance

a data.frame containing the model variables importance

Author(s)

Damien Georges

See Also

BIOMOD_Modeling, bm_RunModel

Other Toolbox objects: BIOMOD.ensemble.models.out, BIOMOD.formated.data, BIOMOD.formated.data.PA, BIOMOD.models.options, BIOMOD.models.out, BIOMOD.options.dataset, BIOMOD.options.default, BIOMOD.projection.out, BIOMOD.rangesize.out, BIOMOD.stored.data, biomod2_ensemble_model

Examples

showClass("biomod2_model")
showClass("ANN_biomod2_model")
showClass("CTA_biomod2_model")
showClass("DNN_biomod2_model")
showClass("FDA_biomod2_model")
showClass("GAM_biomod2_model")
showClass("GBM_biomod2_model")
showClass("GLM_biomod2_model")
showClass("MARS_biomod2_model")
showClass("MAXENT_biomod2_model")
showClass("MAXNET_biomod2_model")
showClass("RF_biomod2_model")
showClass("RFd_biomod2_model")
showClass("SRE_biomod2_model")

Convert probability values into binary values using a predefined threshold

Description

This internal biomod2 function allows to convert probability (not necessary between 0 and 1) values into binary presence-absence (0 or 1) values according to a predefined threshold (see Details).

Usage

bm_BinaryTransformation(data, threshold, do.filtering = FALSE)

## S4 method for signature 'data.frame'
bm_BinaryTransformation(data, threshold, do.filtering = FALSE)

## S4 method for signature 'matrix'
bm_BinaryTransformation(data, threshold, do.filtering = FALSE)

## S4 method for signature 'numeric'
bm_BinaryTransformation(data, threshold, do.filtering = FALSE)

## S4 method for signature 'SpatRaster'
bm_BinaryTransformation(data, threshold, do.filtering = FALSE)

Arguments

Details

If data is a vector, threshold should be a single numeric value.
If data is a matrix, data.frame or SpatRaster, threshold should be a vector containing as many values as the number of columns or layers contained in data. If only one numeric value is given, the same threshold will be applied to all columns or layers.

If do.filtering = FALSE, binary (0 or 1) values are returned.
If do.filtering = TRUE, values will be filtered according to threshold, meaning that :

• data < threshold will return 0

• data >= threshold will return the actual values of data (not transformed in 1)

Value

An object of the same class than data and containing either binary (0 or 1) values, or filtered values.

Author(s)

Wilfried Thuiller, Damien Georges

See Also

BIOMOD_Projection, BIOMOD_EnsembleForecasting

Other Secondary functions: bm_CrossValidation(), bm_FindOptimStat(), bm_MakeFormula(), bm_ModelAnalysis(), bm_ModelingOptions(), bm_PlotEvalBoxplot(), bm_PlotEvalMean(), bm_PlotRangeSize(), bm_PlotResponseCurves(), bm_PlotVarImpBoxplot(), bm_PseudoAbsences(), bm_RangeSize(), bm_RunModelsLoop(), bm_SRE(), bm_SampleBinaryVector(), bm_SampleFactorLevels(), bm_Tuning(), bm_VariablesImportance()

Examples

## Generate a 0-1000 vector (normal distribution)
vec.d <- rnorm(100, 500, 100)

## From continuous to binary / filtered vector
vec.d_bin <- bm_BinaryTransformation(data = vec.d, threshold = 500)
vec.d_filt <- bm_BinaryTransformation(data = vec.d, threshold = 500, do.filtering = TRUE)
cbind(vec.d, vec.d_bin, vec.d_filt)

Build cross-validation table

Description

This internal biomod2 function allows to build a cross-validation table according to 6 different methods : random, kfold, block, strat, env or user.defined (see Details).

Usage

bm_CrossValidation(
  bm.format,
  strategy = "random",
  nb.rep = 0,
  perc = 0.8,
  k = 0,
  balance = "presences",
  env.var = NULL,
  strat = "both",
  user.table = NULL,
  do.full.models = FALSE
)

bm_CrossValidation_user.defined(bm.format, ...)

## S4 method for signature 'BIOMOD.formated.data'
bm_CrossValidation_user.defined(bm.format, user.table)

## S4 method for signature 'BIOMOD.formated.data.PA'
bm_CrossValidation_user.defined(bm.format, user.table)

bm_CrossValidation_random(bm.format, ...)

## S4 method for signature 'BIOMOD.formated.data'
bm_CrossValidation_random(bm.format, nb.rep, perc)

## S4 method for signature 'BIOMOD.formated.data.PA'
bm_CrossValidation_random(bm.format, nb.rep, perc)

bm_CrossValidation_kfold(bm.format, ...)

## S4 method for signature 'BIOMOD.formated.data'
bm_CrossValidation_kfold(bm.format, nb.rep, k)

## S4 method for signature 'BIOMOD.formated.data.PA'
bm_CrossValidation_kfold(bm.format, nb.rep, k)

bm_CrossValidation_block(bm.format, ...)

## S4 method for signature 'BIOMOD.formated.data'
bm_CrossValidation_block(bm.format)

## S4 method for signature 'BIOMOD.formated.data.PA'
bm_CrossValidation_block(bm.format)

bm_CrossValidation_strat(bm.format, ...)

## S4 method for signature 'BIOMOD.formated.data'
bm_CrossValidation_strat(bm.format, balance, strat, k)

## S4 method for signature 'BIOMOD.formated.data.PA'
bm_CrossValidation_strat(bm.format, balance, strat, k)

bm_CrossValidation_env(bm.format, ...)

## S4 method for signature 'BIOMOD.formated.data'
bm_CrossValidation_env(bm.format, balance, k, env.var)

## S4 method for signature 'BIOMOD.formated.data.PA'
bm_CrossValidation_env(bm.format, balance, k, env.var)

Arguments

Details

Several parameters are available within the function and some of them can be used with different cross-validation strategies :

| ....... | random | kfold | block | strat | env |
__________________________________________________
| nb.rep. | x..... | x.... | ..... | ..... | ... |
| perc... | x..... | ..... | ..... | ..... | ... |
| k...... | ...... | x.... | ..... | x.... | x.. |
| balance | ...... | ..... | ..... | x.... | x.. |
| strat.. | ...... | ..... | ..... | x.... | ... |

Concerning column names of matrix output :

The number of columns depends on the strategy selected. The column names are given a posteriori of the selection, ranging from 1 to the number of columns. If do.full.models = TRUE, columns merging runs (and/or pseudo-absence datasets) are added at the end.

Concerning cross-validation strategies :

random

Most simple method to calibrate and validate a model is to split the original dataset in two datasets : one to calibrate the model and the other one to validate it. The splitting can be repeated nb.rep times.

k-fold

The k-fold method splits the original dataset in k datasets of equal sizes : each part is used successively as the validation dataset while the other k-1 parts are used for the calibration, leading to k calibration/validation ensembles. This multiple splitting can be repeated nb.rep times.

block

It may be used to test for model overfitting and to assess transferability in geographic space. block stratification was described in Muscarella et al. 2014 (see References). Four bins of equal size are partitioned (bottom-left, bottom-right, top-left and top-right).

stratified

It may be used to test for model overfitting and to assess transferability in geographic space. x and y stratification was described in Wenger and Olden 2012 (see References). y stratification uses k partitions along the y-gradient, x stratification does the same for the x-gradient. both returns 2k partitions: k partitions stratified along the x-gradient and k partitions stratified along the y-gradient.

environmental

It may be used to test for model overfitting and to assess transferability in environmental space. It returns k partitions for each variable given in env.var.

user-defined

Allow the user to give its own crossvalidation table. For a presence-absence dataset, column names must be formatted as: _allData_RUNx with x an integer. For a presence-only dataset for which several pseudo-absence dataset were generated, column names must be formatted as: _PAx_RUNy with x an integer and PAx an existing pseudo-absence dataset and y an integer

Concerning balance parameter :

If balance = 'presences', presences are divided (balanced) equally over the partitions (e.g. Fig. 1b in Muscarelly et al. 2014). Absences or pseudo-absences will however be unbalanced over the partitions especially if the presences are clumped on an edge of the study area.

If balance = 'absences', absences (resp. pseudo-absences or background) are divided (balanced) as equally as possible between the partitions (geographical balanced bins given that absences are spread over the study area equally, approach similar to Fig. 1 in Wenger et Olden 2012). Presences will however be unbalanced over the partitions especially if the presences are clumped on an edge of the study area.

Value

A matrix or data.frame defining for each repetition (in columns) which observation lines should be used for models calibration (TRUE) and validation (FALSE).

Author(s)

Maya Guéguen

References

• Muscarella, R., Galante, P.J., Soley-Guardia, M., Boria, R.A., Kass, J.M., Uriarte, M. & Anderson, R.P. (2014). ENMeval: An R package for conducting spatially independent evaluations and estimating optimal model complexity for Maxent ecological niche models. Methods in Ecology and Evolution, 5, 1198-1205.

• Wenger, S.J. & Olden, J.D. (2012). Assessing transferability of ecological models: an underappreciated aspect of statistical validation. Methods in Ecology and Evolution, 3, 260-267.

See Also

get.block, kfold, BIOMOD_FormatingData, BIOMOD_Modeling

Other Secondary functions: bm_BinaryTransformation(), bm_FindOptimStat(), bm_MakeFormula(), bm_ModelAnalysis(), bm_ModelingOptions(), bm_PlotEvalBoxplot(), bm_PlotEvalMean(), bm_PlotRangeSize(), bm_PlotResponseCurves(), bm_PlotVarImpBoxplot(), bm_PseudoAbsences(), bm_RangeSize(), bm_RunModelsLoop(), bm_SRE(), bm_SampleBinaryVector(), bm_SampleFactorLevels(), bm_Tuning(), bm_VariablesImportance()

Examples

library(terra)

# Load species occurrences (6 species available)
data(DataSpecies)
head(DataSpecies)

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

# Get corresponding presence/absence data
myResp <- as.numeric(DataSpecies[, myRespName])

# Get corresponding XY coordinates
myRespXY <- DataSpecies[, c('X_WGS84', 'Y_WGS84')]

# Load environmental variables extracted from BIOCLIM (bio_3, bio_4, bio_7, bio_11 & bio_12)
data(bioclim_current)
myExpl <- terra::rast(bioclim_current)



# --------------------------------------------------------------- #
# Format Data with true absences
myBiomodData <- BIOMOD_FormatingData(resp.name = myRespName,
                                     resp.var = myResp,
                                     resp.xy = myRespXY,
                                     expl.var = myExpl)

# --------------------------------------------------------------- #
# Create the different validation datasets

# random selection
cv.r <- bm_CrossValidation(bm.format = myBiomodData,
                           strategy = "random",
                           nb.rep = 3,
                           k = 0.8)

# k-fold selection
cv.k <- bm_CrossValidation(bm.format = myBiomodData,
                           strategy = "kfold",
                           nb.rep = 2,
                           k = 3)

# block selection
cv.b <- bm_CrossValidation(bm.format = myBiomodData,
                           strategy = "block")

# stratified selection (geographic)
cv.s <- bm_CrossValidation(bm.format = myBiomodData,
                           strategy = "strat",
                           k = 2,
                           balance = "presences",
                           strat = "x")

# stratified selection (environmental)
cv.e <- bm_CrossValidation(bm.format = myBiomodData,
                           strategy = "env",
                           k = 2,
                           balance = "presences")

head(cv.r)
apply(cv.r, 2, table)
head(cv.k)
apply(cv.k, 2, table)
head(cv.b)
apply(cv.b, 2, table)
head(cv.s)
apply(cv.s, 2, table)
head(cv.e)
apply(cv.e, 2, table)

Calculate the best score according to a given evaluation method

Description

This internal biomod2 function allows the user to find the threshold to convert continuous values into binary ones leading to the best score for a given evaluation metric.

Usage

bm_FindOptimStat(
  metric.eval = "TSS",
  obs,
  fit,
  nb.thresh = 100,
  threshold = NULL,
  boyce.bg.env = NULL,
  mpa.perc = 0.9,
  k = NULL
)

get_optim_value(metric.eval)

bm_CalculateStatBin(misc, metric.eval = "TSS")

bm_CalculateStatAbun(metric.eval, obs, fit, k)

Arguments

Details

Please refer to CAWRC website ("Methods for dichotomous forecasts") to get detailed description (simple/complex metrics).
Optimal value of each method can be obtained with the get_optim_value function.

simple

• POD : Probability of detection (hit rate)

• FAR : False alarm ratio

• POFD : Probability of false detection (fall-out)

• SR : Success ratio

• ACCURACY : Accuracy (fraction correct)

• BIAS : Bias score (frequency bias)

complex

• AUCroc : Area Under Curve of Relative operating characteristic

• AUCprg : Area Under Curve of Precision-Recall-Gain curve

• TSS : True skill statistic (Hanssen and Kuipers discriminant, Peirce's skill score)

• KAPPA : Cohen's Kappa (Heidke skill score)

• OR : Odds Ratio

• ORSS : Odds ratio skill score (Yule's Q)

• CSI : Critical success index (threat score)

• ETS : Equitable threat score (Gilbert skill score)

presence-only

• BOYCE : Boyce index

• MPA : Minimal predicted area (cutoff optimizing MPA to predict 90% of presences)

abundance / count / relative data

• RMSE : Root Mean Square Error

• MSE : Mean Square Error

• MAE : Mean Absolute Error

• Rsquared : R squared

• Rsquared_aj : R squared adjusted

• Max_error : Maximum error

multiclass/ordinal data

• Accuracy : Accuracy

• Recall : Macro average Recall

• Precision : Macro average Precision

• F1 : Macro F1 score

Note that if a value is given to threshold, no optimization will be done, and only the score for this threshold will be returned.

The Boyce index returns NA values for SRE models because it can not be calculated with binary predictions.
This is also the reason why some NA values might appear for GLM models if they did not converge.

Value

A 1 row x 5 columns data.frame containing :

• metric.eval : the chosen evaluation metric

• cutoff : the associated cut-off used to transform the continuous values into binary

• sensitivity : the sensibility obtained on fitted values with this threshold

• specificity : the specificity obtained on fitted values with this threshold

• best.stat : the best score obtained for the chosen evaluation metric

Note

In order to break dependency loop between packages biomod2 and ecospat, code of ecospat.boyce() and ecospat.mpa() in ecospat) functions have been copied within this file from version 3.2.2 (august 2022).

Author(s)

Damien Georges

References

• Engler, R., Guisan, A., and Rechsteiner L. 2004. An improved approach for predicting the distribution of rare and endangered species from occurrence and pseudo-absence data. Journal of Applied Ecology, 41(2), 263-274.

• Hirzel, A. H., Le Lay, G., Helfer, V., Randin, C., and Guisan, A. 2006. Evaluating the ability of habitat suitability models to predict species presences. Ecological Modelling, 199(2), 142-152.

See Also

ecospat.boyce() and ecospat.mpa() in ecospat, BIOMOD_Modeling, bm_RunModelsLoop, BIOMOD_EnsembleModeling

Other Secondary functions: bm_BinaryTransformation(), bm_CrossValidation(), bm_MakeFormula(), bm_ModelAnalysis(), bm_ModelingOptions(), bm_PlotEvalBoxplot(), bm_PlotEvalMean(), bm_PlotRangeSize(), bm_PlotResponseCurves(), bm_PlotVarImpBoxplot(), bm_PseudoAbsences(), bm_RangeSize(), bm_RunModelsLoop(), bm_SRE(), bm_SampleBinaryVector(), bm_SampleFactorLevels(), bm_Tuning(), bm_VariablesImportance()

Examples

## Generate a binary vector
vec.a <- sample(c(0, 1), 100, replace = TRUE)

## Generate a 0-1000 vector (random drawing)
vec.b <- runif(100, min = 0, max = 1000)

## Generate a 0-1000 vector (biased drawing)
BiasedDrawing <- function(x, m1 = 300, sd1 = 200, m2 = 700, sd2 = 200) {
  return(ifelse(x < 0.5, rnorm(1, m1, sd1), rnorm(1, m2, sd2)))
}
vec.c <- sapply(vec.a, BiasedDrawing)
vec.c[which(vec.c < 0)] <- 0
vec.c[which(vec.c > 1000)] <- 1000

## Find optimal threshold for a specific evaluation metric
bm_FindOptimStat(metric.eval = 'TSS', fit = vec.b, obs = vec.a)
bm_FindOptimStat(metric.eval = 'TSS', fit = vec.c, obs = vec.a, nb.thresh = 100)
bm_FindOptimStat(metric.eval = 'TSS', fit = vec.c, obs = vec.a, threshold = 280)

Standardized formula maker

Description

This internal biomod2 function allows the user to create easily a standardized formula that can be used later by statistical models.

Usage

bm_MakeFormula(
  resp.name,
  expl.var,
  type = "simple",
  interaction.level = 0,
  k = NULL
)

Arguments

Details

It is advised to give only a subset of expl.var table to avoid useless memory consuming.
If some explanatory variables are factorial, expl.var must be a data.frame whose corresponding columns are defined as factor.

Value

A formula class object that can be directly given to most of R statistical models.

Author(s)

Damien Georges

See Also

formula, s, s, bm_ModelingOptions, bm_Tuning, bm_RunModelsLoop

Other Secondary functions: bm_BinaryTransformation(), bm_CrossValidation(), bm_FindOptimStat(), bm_ModelAnalysis(), bm_ModelingOptions(), bm_PlotEvalBoxplot(), bm_PlotEvalMean(), bm_PlotRangeSize(), bm_PlotResponseCurves(), bm_PlotVarImpBoxplot(), bm_PseudoAbsences(), bm_RangeSize(), bm_RunModelsLoop(), bm_SRE(), bm_SampleBinaryVector(), bm_SampleFactorLevels(), bm_Tuning(), bm_VariablesImportance()

Examples

## Create simple simulated data
myResp.s <- sample(c(0, 1), 20, replace = TRUE)
myExpl.s <- data.frame(var1 = sample(c(0, 1), 100, replace = TRUE),
                       var2 = rnorm(100),
                       var3 = 1:100)

## Generate automatic formula
bm_MakeFormula(resp.name = 'myResp.s',
               expl.var = head(myExpl.s),
               type = 'quadratic',
               interaction.level = 0)

Analyze the residuals of the single models

Description

This function return several graphs to help analyse the single models.

Usage

bm_ModelAnalysis(
  bm.mod,
  models.chosen = "all",
  color.by = "full.name",
  do.plot = TRUE
)

Arguments

Details

5 plots can be obtained with this function :

residuals ~ observations number

to detect outliers. The x-axis only helps to find the outlier number.

residuals Q-Q plot

to check if residuals follow a normal distribution. Points should follow the black line.

residuals histogram

to check if residuals follow a normal distribution. Histogram should represent a gaussian distribution.

residuals ~ fitted values

to detect an heteroscedasticity of the residuals. It corresponds to the Tukey-Anscombe plot.

R scores

to detect overfitting. Representing Rsquared and Rsquared_aj values, there should not be a big gap between calibration and validation values.

Please not the all plots are made for all models, independently to the different models assumptions. It is up to the user to interpret the graphics.

Value

A list containing :

1. a data.frame with variables, predicted values and model residuals

2. a data.frame with the R score values

3. 5 ggplot objects (from A to E) representing the different analyses.

Author(s)

Hélène Blancheteau

See Also

Other Secondary functions: bm_BinaryTransformation(), bm_CrossValidation(), bm_FindOptimStat(), bm_MakeFormula(), bm_ModelingOptions(), bm_PlotEvalBoxplot(), bm_PlotEvalMean(), bm_PlotRangeSize(), bm_PlotResponseCurves(), bm_PlotVarImpBoxplot(), bm_PseudoAbsences(), bm_RangeSize(), bm_RunModelsLoop(), bm_SRE(), bm_SampleBinaryVector(), bm_SampleFactorLevels(), bm_Tuning(), bm_VariablesImportance()

Other Plot functions: bm_PlotEvalBoxplot(), bm_PlotEvalMean(), bm_PlotRangeSize(), bm_PlotResponseCurves(), bm_PlotVarImpBoxplot()

Examples

library(terra)

# Load species occurrences (6 species available)
data(DataSpecies)
head(DataSpecies)

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

# Get corresponding presence/absence data
myResp <- as.numeric(DataSpecies[, myRespName])

# Get corresponding XY coordinates
myRespXY <- DataSpecies[, c('X_WGS84', 'Y_WGS84')]

# Load environmental variables extracted from BIOCLIM (bio_3, bio_4, bio_7, bio_11 & bio_12)
data(bioclim_current)
myExpl <- terra::rast(bioclim_current)



# ---------------------------------------------------------------#
file.out <- paste0(myRespName, "/", myRespName, ".AllModels.models.out")
if (file.exists(file.out)) {
  myBiomodModelOut <- get(load(file.out))
} else {

  # Format Data with true absences
  myBiomodData <- BIOMOD_FormatingData(resp.name = myRespName,
                                       resp.var = myResp,
                                       resp.xy = myRespXY,
                                       expl.var = myExpl)

  # Model single models
  myBiomodModelOut <- BIOMOD_Modeling(bm.format = myBiomodData,
                                      modeling.id = 'AllModels',
                                      models = c('RF', 'GLM'),
                                      CV.strategy = 'random',
                                      CV.nb.rep = 2,
                                      CV.perc = 0.8,
                                      OPT.strategy = 'bigboss',
                                      metric.eval = c('TSS','AUCroc'),
                                      var.import = 3,
                                      seed.val = 42)
}


# ---------------------------------------------------------------#
# Explore single models
myBiomodAnalysis <- bm_ModelAnalysis(bm.mod = myBiomodModelOut, color.by = "run")

plot(myBiomodAnalysis$plot.outliers)

Configure the modeling options for each selected model

Description

Parameterize and/or tune biomod2's single models options.

Usage

bm_ModelingOptions(
  data.type = "binary",
  models,
  strategy,
  user.val = NULL,
  user.base = "bigboss",
  bm.format = NULL,
  calib.lines = NULL
)

Arguments

Details

This function creates a BIOMOD.models.options object containing parameter values for each single model that can be run within biomod2 through BIOMOD_Modeling function.

11 different algorithms are currently available, with different versions for some (GAM, MAXENT, RF). These models are associated with binary or nonbinary datatypes, but all combinations are not available. They are all listed within the ModelsTable dataset.

Different strategies are available to set those parameters, through the strategy argument :

default

all parameters names and values are directly retrieve from functions to be called through formalArgs and formals functions

bigboss

default parameter values are updated with values predefined by biomod2 team (see OptionsBigboss)

user.defined

default parameter values are updated with values provided by the user

tuned

default parameter values are updated by calling bm_Tuning function

To define the same options for all datasets of a model, options can be provided through the user.val parameter as a list whose name is for_all_datasets.

Value

A BIOMOD.models.options of object that can be used to build species distribution model(s) with the BIOMOD_Modeling function.

Note

MAXENT being the only external model (not called through a R package), default parameters, and their values, are the following :

Author(s)

Damien Georges, Wilfried Thuiller, Maya Guéguen

See Also

ModelsTable, OptionsBigboss, BIOMOD.models.options, bm_Tuning, BIOMOD_Modeling

Other Secondary functions: bm_BinaryTransformation(), bm_CrossValidation(), bm_FindOptimStat(), bm_MakeFormula(), bm_ModelAnalysis(), bm_PlotEvalBoxplot(), bm_PlotEvalMean(), bm_PlotRangeSize(), bm_PlotResponseCurves(), bm_PlotVarImpBoxplot(), bm_PseudoAbsences(), bm_RangeSize(), bm_RunModelsLoop(), bm_SRE(), bm_SampleBinaryVector(), bm_SampleFactorLevels(), bm_Tuning(), bm_VariablesImportance()

Examples

library(terra)

# Load species occurrences (6 species available)
data(DataSpecies)
head(DataSpecies)

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

# Get corresponding presence/absence data
myResp <- as.numeric(DataSpecies[, myRespName])

# Get corresponding XY coordinates
myRespXY <- DataSpecies[, c('X_WGS84', 'Y_WGS84')]

# Load environmental variables extracted from BIOCLIM (bio_3, bio_4, bio_7, bio_11 & bio_12)
data(bioclim_current)
myExpl <- terra::rast(bioclim_current)



# ---------------------------------------------------------------#
# Format Data with true absences
myBiomodData <- BIOMOD_FormatingData(resp.name = myRespName,
                                     resp.var = myResp,
                                     resp.xy = myRespXY,
                                     expl.var = myExpl)

# k-fold selection
cv.k <- bm_CrossValidation(bm.format = myBiomodData,
                           strategy = 'kfold',
                           nb.rep = 2,
                           k = 3)


# ---------------------------------------------------------------#
allModels <- c('ANN', 'CTA', 'DNN', 'FDA', 'GAM', 'GBM', 'GLM', 'MARS'
               , 'MAXENT', 'MAXNET', 'RF', 'RFd', 'SRE', 'XGBOOST')

# default parameters
opt.d <- bm_ModelingOptions(data.type = 'binary',
                            models = allModels,
                            strategy = 'default')

# providing formated data
opt.df <- bm_ModelingOptions(data.type = 'binary',
                             models = allModels,
                             strategy = 'default',
                             bm.format = myBiomodData,
                             calib.lines = cv.k)

opt.d
opt.d@models
opt.d@options$ANN.binary.nnet.nnet
names(opt.d@options$ANN.binary.nnet.nnet@args.values)

opt.df@options$ANN.binary.nnet.nnet
names(opt.df@options$ANN.binary.nnet.nnet@args.values)


# ---------------------------------------------------------------#
# bigboss parameters
opt.b <- bm_ModelingOptions(data.type = 'binary',
                            models = allModels,
                            strategy = 'bigboss')

# user defined parameters
user.SRE <- list('_allData_allRun' = list(quant = 0.01))
user.XGBOOST <- list('_allData_allRun' = list(nrounds = 10))
user.val <- list(SRE.binary.biomod2.bm_SRE = user.SRE
                 , XGBOOST.binary.xgboost.xgboost = user.XGBOOST)

opt.u <- bm_ModelingOptions(data.type = 'binary',
                            models = c('SRE', 'XGBOOST'),
                            strategy = 'user.defined',
                            user.val = user.val)

opt.b
opt.u

## Not run: 
# tuned parameters with formated data
opt.t <- bm_ModelingOptions(data.type = 'binary',
                            models = c('SRE', 'XGBOOST'),
                            strategy = 'tuned',
                            bm.format = myBiomodData)
opt.t

## End(Not run)

Plot boxplot of evaluation scores

Description

This function represents boxplot of evaluation scores of species distribution models, from BIOMOD.models.out or BIOMOD.ensemble.models.out objects that can be obtained from BIOMOD_Modeling or BIOMOD_EnsembleModeling functions. Scores are represented according to 2 grouping methods (see Details).

Usage

bm_PlotEvalBoxplot(
  bm.out,
  dataset = "calibration",
  group.by = c("algo", "run"),
  do.plot = TRUE,
  ...
)

Arguments

Details

... can take the following values :

• main : a character corresponding to the graphic title

• scales : a character corresponding to the scales argument of the facet_wrap function, must be either fixed, free_x, free_y or free

Value

A list containing a data.frame with evaluation scores and the corresponding ggplot object representing them in boxplot.

Author(s)

Damien Georges, Maya Guéguen

See Also

BIOMOD.models.out, BIOMOD.ensemble.models.out, BIOMOD_Modeling, BIOMOD_EnsembleModeling, get_evaluations

Other Secondary functions: bm_BinaryTransformation(), bm_CrossValidation(), bm_FindOptimStat(), bm_MakeFormula(), bm_ModelAnalysis(), bm_ModelingOptions(), bm_PlotEvalMean(), bm_PlotRangeSize(), bm_PlotResponseCurves(), bm_PlotVarImpBoxplot(), bm_PseudoAbsences(), bm_RangeSize(), bm_RunModelsLoop(), bm_SRE(), bm_SampleBinaryVector(), bm_SampleFactorLevels(), bm_Tuning(), bm_VariablesImportance()

Other Plot functions: bm_ModelAnalysis(), bm_PlotEvalMean(), bm_PlotRangeSize(), bm_PlotResponseCurves(), bm_PlotVarImpBoxplot()

Examples

library(terra)

# Load species occurrences (6 species available)
data(DataSpecies)
head(DataSpecies)

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

# Get corresponding presence/absence data
myResp <- as.numeric(DataSpecies[, myRespName])

# Get corresponding XY coordinates
myRespXY <- DataSpecies[, c('X_WGS84', 'Y_WGS84')]

# Load environmental variables extracted from BIOCLIM (bio_3, bio_4, bio_7, bio_11 & bio_12)
data(bioclim_current)
myExpl <- terra::rast(bioclim_current)



# ---------------------------------------------------------------
file.out <- paste0(myRespName, "/", myRespName, ".AllModels.models.out")
if (file.exists(file.out)) {
  myBiomodModelOut <- get(load(file.out))
} else {

  # Format Data with true absences
  myBiomodData <- BIOMOD_FormatingData(resp.name = myRespName,
                                       resp.var = myResp,
                                       resp.xy = myRespXY,
                                       expl.var = myExpl)

  # Model single models
  myBiomodModelOut <- BIOMOD_Modeling(bm.format = myBiomodData,
                                      modeling.id = 'AllModels',
                                      models = c('RF', 'GLM'),
                                      CV.strategy = 'random',
                                      CV.nb.rep = 2,
                                      CV.perc = 0.8,
                                      OPT.strategy = 'bigboss',
                                      metric.eval = c('TSS', 'ROC'),
                                      var.import = 3,
                                      seed.val = 42)
}


# ---------------------------------------------------------------
# Get evaluation scores
get_evaluations(myBiomodModelOut)

# Represent evaluation scores
bm_PlotEvalBoxplot(bm.out = myBiomodModelOut, group.by = c('algo', 'run'))

Plot mean evaluation scores

Description

This function represents mean evaluation scores (and their standard deviation) of species distribution models, from BIOMOD.models.out or BIOMOD.ensemble.models.out objects that can be obtained from BIOMOD_Modeling or BIOMOD_EnsembleModeling functions. Scores are represented according to 2 different evaluation methods, and models can be grouped (see Details).

Usage