DNN : Deep Neural Network with cito
packageModelsTable and OptionsBigBoss
tune function of cito
packageAdd the datatype multiclass. It is now possible to
model with a factorial response (not ordered).
Two new ensemble models have been added: EMmode and
EMfreq (for the mode of the response and the frequency of
that mode).
In biomod2_classes_0 file :
.BIOMOD.options.default.correct
functionIn biomod2_classes_1 file :
plot method for
BIOMOD.projection.out objectIn biomod2_classes_3 file :
.BIOMOD.formated.data.check_data,
.plot.BIOMOD.formated.data.abundance, summary
and show functionIn biomod2_classes_4 file :
factor_levels slot for biomod2_model
classpredict2 functions for single models to deal
with the different data typesIn bm_RunModelsLoop, BIOMOD_Projection,
BIOMOD_EnsembleModeling and
BIOMOD_EnsembleForecasting functions :
In biomod2_classes_5 file :
EMmode and EMfreq with
predict methodsdata.type in
bm_ModelingOptions
ROC to
AUCroc. It will switch automatically if you use
ROC.AUCprg : Area Under Curve of the
Precision-Recall-Gain curve.togglelayerselected,
maximumbackground, maximumiterations,
convergencethreshold, autofeature,
jackknife, writeclampgrid,
writemess, logfile and verbose
parameters in MAXENT parameters.check_calib.lines_names for
BIOMOD.formated.data.PA object in
bm_CrossValidation_user.defined
binary and nonbinary models into
OptionsBigboss
data.type = 'binary' by default in
bm_ModelingOptions
.BIOMOD_Modeling.summary
gam:: in bm_MakeFormula
call slot for BIOMOD.formated.data and
BIOMOD.formated.data.PA, BIOMOD.models.out,
BIOMOD.projection.out, and
BIOMOD.ensemble.models.out classesBIOMOD_Report function.Rmd templates in inst/rmd/ folder
(for report, ODMAP and code)BIOMOD_RangeSize to
bm_RangeSize
BIOMOD_RangeSize inputs to
BIOMOD.projection.out objectsBIOMOD.rangesize.out
bm_PlotRangeSize input to
BIOMOD.rangesize.out objects, and project maps with
coordinates of BIOMOD.projection.out objectshow outputs of BIOMOD.[...].out
objectsdata.type, has.filter.raster and
biomod2.version slot for BIOMOD.formated.data
and BIOMOD.formated.data.PA classes.BIOMOD.options.default.correct function (default
changes and corrections depending on the data type (mainly for
type, method, family and
distribution options) to
BIOMOD.options.dataset).BIOMOD.formated.data.check_data function (routine
that can be applied to both original and evaluation datasets).plot.BIOMOD.formated.data.abundance function (to
plot formated data when data type is not binary)BIOMOD.formated.data summary to the different
data typesdata.type slot for BIOMOD.models.out
classset_new_dirname and
set_new_dirname.models functions (to recursively modify the
dir.name slot in all biomod2 objects of an existing
simulation directory)model_type and thresholds_ordinal slot
for biomod2_model classpredict2 functions for single models to deal
with the different data types (mainly between binary,
ordinal and the others)bm_CrossValidation function :
.sample_mat function into
.sample_num and .sample_class functionsbm_FindOptimStat function :
k parameter.contingency_table_ordinal functionbm_CalculateStat function into
bm_CalculateStatBin and bm_CalculateStatAbun
functionsbm_VariablesImportance function :
on_0_1000 or on_1_1000 when
making projectionsbm_ModelAnalysis function (to analyse the
residuals of single models)ModelsTable data with nonbinary
data typeOptionsBigboss data with all nonbinary
single models optionsdir.name slot within
BIOMOD.formated.data into absolute one instead of relative
by defaultaes_string to [aes +
.data] in bm_Plot[...] functionsdigits and overwrite parameters in
BIOMOD_Projection functionget_var_type and get_var_range to
internal functions (.get_var_type and
.get_var_range)
bm_CrossValidation (partitions not being balanced
correctly)bm_PlotEvalMean
bm_Tuning (to be able
to run the step AIC with the user formula)scale.models = FALSE by default in
BIOMOD_Modeling and bm_RunModelsLoop
get_evaluations and
get_variables_importance message when data is not available
(extended to bm_Plot[...] functions)parallel::stopCluster and
foreach:::.foreachGlobals.errorhandling in foreach loop in
bm_RunModelsLoop (for MAXENT on Windows)bm_CrossValidation
bm_PlotEvalMean
MAXENT.partitions, MAXENT.kfolds,
MAXENT.user.grp and MAXENT.tune.args
parameters in bm_Tuning functiontif)OptionsBigboss have been
modified (concerns only ANN, CTA and RF models)RFd : Random Forest with a down-sampling
method.seed.val for bm_pseudoAbsences and
BIOMOD_FormatingData.fact.aggr argument for pseudo-absences selection
with the random and disk methods. It allows to reduce the resolution of
the environment.bm_ModelingOptions.BIOMOD_EnsembleModeling when multiple PA datasets
(obs and fit not matching when calling
bm_FindOptimStat)new.env
in BIOMOD_Projection and
BIOMOD_EnsembleForecasting
tests/ folder (unused)new.env in
BIOMOD_Projection and
BIOMOD_EnsembleForecasting
MAXENT tuningBIOMOD.options.dataset and
BIOMOD.models.options classesGAM.binary.gam.gam and
GAM.binary.mgcv.bam in OptionsBigboss
datasetobs and fit parameters in
bm_FindOptimStat
bm_Tuning function :
MAXENT.algorithm and MAXENT.parallel
parametersXGBOOST and SRE
gamSpline to gamLoess method
to tune GAM.gam.gam model, and add GAM.span
and GAM.degree parametersBIOMOD_Modeling
BIOMOD.options.default and
BIOMOD.options.dataset classes, retrieving default
parameters and values with formalArgs functionModelsTable and OptionsBigboss
datasets containing single models informations and pre-defined modeling
optionsBIOMOD_ModelingOptions to
bm_ModelingOptions
bm_ModelingOptions directly in
BIOMOD_Modeling and add related OPT.[...]
parametersBIOMOD_Tuning to bm_Tuning and adapt
it to match with new modeling optionsbm_RunModelsLoop in
a more generalized way dealing with new modeling optionsBIOMOD_PresenceOnly function and add
BOYCE and MPA indices into
bm_FindOptimStat functionFLT4S datatype only when
EMcv is activated, otherwise use INT2S
wrap applied to a
data.frame in BIOMOD_Projection
predict method for RF with
do.classif = FALSE
bm_PlotEvalMean
do.classif ignored in
BIOMOD_ModelingOptions
BIOMOD_Projection
BIOMOD.formated.data
and BIOMOD.formated.data.PA
bm_PlotResponseCurves for ensemble
models merged by algo (for Maxent)point.size argument to
plot.BIOMOD.Formated.data
maxcell argument to
plot.BIOMOD.projected.out
verbose = 0 (from
verbose = 1)BIOMOD_FormatingData checks for
resp.xy
.get_data_mask
BIOMOD_Modeling when using
sampsize as a vector. argument strata was
badly formattedBIOMOD_EnsembleModeling for
additional projection with only one environmental variablesBIOMOD_EnsembleForecasting when
several projection are running simultaneously and using the same
temporary directorybm_CrossValidation with
user.defined tables badly formatted (TRUE/FALSE for data
not in the given PA dataset are now properly transformed into NA)models.pa
argument in BIOMOD_Modeling).BIOMOD_CrossValidation have been renamed
bm_CrossValidation and cross-validation with k-fold,
stratified and environmental strategy now work properly with
pseudo-absence dataset. All cross-validation strategy can now be called
directly through BIOMOD_Modeling.get_evaluations,
BIOMOD_EnsembleModeling, bm_RunModelsLoop,
bm_RunModel)save.output. output are now
automatically saved.terra and raster
CV.perc (formerly data.split.perc) now
uses a 0-1 range (instead of 0-100)BIOMOD_EnsembleModeling now
gives an error.metric.select.dataset to
BIOMOD_EnsembleModeling to choose the dataset which
evaluation metric should be used to filter and/or weigh the ensemble
models. Default value is now ‘validation’ instead of ‘evaluation’.na.rm to
BIOMOD_EnsembleModeling to harmonize the management of
NA among individual model predictions.RF$sampsize
parameter in BIOMOD_ModelingOptions
BIOMOD_Projection and
BIOMOD_EnsembleForecasting when
terraOption(todisk = TRUE) is activated (for large or
numerous raster).data.table object (that
are converted into standard data.frame).do.stack = FALSE and resp.name with
. inside.filter.raster = TRUE in bm_PseudoAbsence.BIOMOD.formated.data.PA
get_species_data and
get_eval_data
.BIOMOD_Modeling.prepare.data
bm_RunModelsLoop to do the PA loop within
the functioncalib.lines and eval.lines variable names
are standardised (no more calibLines or
eval_lines)data.table (removed use of
rbindlist).get_env_class to reduce code redundancycategorical_stack_to_terra into
.categorical_stack_to_terra
BIOMOD_FormatingData checks into
bm_PseudoAbsences
'.tif' is available as an output format for raster
projection'.tif' is the new default output format for raster
projectionplot and summary methods for
BIOMOD_FormatingData output. These method now support the
use of calib.lines to explore how the cross-validation
dataset are structured.plot methods for
BIOMOD.projection.out objects so that it uses
ggplot2 for nicer plots.BIOMOD.projection.out objects. They can be loaded from the
disk with get_predictions or represented through
BIOMOD.projection.out plot method.get_predictions now return a proper
data.frame (unless projection on spatial data) with many
additional information available. Old behavior can be reproduced by
using get_predictions(x, model.as.col = TRUE).get_evaluations now return a cleaner
data.frame with more consistent information available.maxent.jar); ‘MAXENT.Phillips.2’ -> ‘MAXNET’
(based on maxnet package).BIOMOD_FormatingData now gives warning when several
input data points are located in the same raster cellsfilter.raster in
BIOMOD_FormatingData to filter data points so that none are
located in the same raster cells.BIOMOD_EnsembleModeling now have an argument
em.algo to select the ensemble algorithm to be computed.
Separate arguments are now deprecated (prob.mean,
prob.median, prob.cv, prob.ci,
committee.averaging, prob.mean.weight).
Building all possible ensemble models can now be done with
em.algo = c('EMmean','EMmedian','EMcv','EMci','EMca','EMwmean').em.by have slightly changed:
‘PA_dataset’ -> ‘PA’, ‘PA_dataset+repet’ -> ‘PA+run’ and
‘PA_dataset+algo’ -> ‘PA+algo’BIOMOD_Modeling and
BIOMOD_EnsembleModeling.MAXENT.Phillips.2 and single variable
models.BIOMOD_CrossValidation for block-stratified
samplingBIOMOD_CrossValidation for
pseudo-absencessp is back into Imports due to the
need to use sp::read.asciigrid
terra version number (>= 1.6-33)
as terra 1.6-41 was released on CRAN.do.stack = TRUE, only stacked projection are now
saved to the disk.initial_heap_size and max_heap_size
in MAXENT.Phillips modeling optionsMAXENT.Phillips.MAXENT.Phillips predict method for large dataset
(require sp::read.asciigrid).em.by = 'all' or 'algo'.BIOMOD_EnsembleModeling.BIOMOD_EnsembleForecasting when a single
evaluation metric was available and binary/filtered transformation were
asked for.BIOMOD.formated.data.PA
object.MAXENT.Phillips for Windows.do.stack = FALSE with
BIOMOD_Projection.EMcv ensemble modeling for
data.frame by removing dependency to
raster::cv.free method with
PackedSpatRaster
BIOMOD_FormatingData in case where no coordinates
are givenMAXENT.Phillips predict2 method
for SpatRaster so that it saves environmental data as
.asc and do not use the data.frame
method.BIOMOD.formated.data@data.mask slot.
data.mask can now be safely saved and re-opened ;
data.mask can now store a different extent for evaluation
datasetterra
(> 1.6.33) and do not automatically import
raster and sp.raster and sp package into
SUGGESTS rather than DEPENDS.raster and sp input data type are still
supported.data() .bm_BinaryTransformation now always returns
0/1 and never
TRUE/FALSE
bm_PlotResponseCurves for
new.env possible data types.BIOMOD_Projection and
BIOMOD_EnsembleForecasting now properly support matrix as
new.env
get_prediction on biomod.projection.out
generated from BIOMOD_Projection based on
SpatRaster with arg as.data.frame = TRUE are
now possible.bm_BinaryTransformation now return same type of object
as its inputBIOMOD_RangeSize, indicating
how comparison are done depending on the number of models in current vs
future.BIOMOD_CrossValidation
bm_BinaryTransformation with
data.frame/matrix and
do.filtering = TRUE
bm_PlotResponseCurves now work with factors in
univariate representationbm_PlotResponseCurves properly handles
SpatRaster and Raster as
new.env
MAXENT.Phillips and a
single environmental variableBIOMOD_EnsembleForecasting so that it
properly accounts for new.env.xy when projecting on
matrix or data.frame.BIOMOD_EnsembleModeling now works when called for a
single ensemble modelBIOMOD_RangeSize.
Comparisons with non-binary values throw errors.BIOMOD_RangeSize and data.frame methodBIOMOD_RangeSize data.frame method now
handles 1 current vs n future projectionBIOMOD_PresenceOnly that can now work when
evaluation data are providedBIOMOD_PresenceOnly that can now work when
only the EM have been providedBIOMOD_PresenceOnly to
SpatRaster and SpatVector.build_clamping_mask now support categorical
variables.categorical2numeric to transform
categorical variables into numeric within a
data.frame..get_categorical_names to
retrieve categorical variable names from a data.frame.load_stored_object method into a method for
BIOMOD.stored.SpatRaster and a method for all other
BIOMOD.stored.data.BIOMOD.stored.SpatRaster stores
PackedSpatraster and not SpatRaster..CompteurSp based on old function
CompteurSp that was defined within a function.check_data_range().BIOMOD_EnsembleForecasting.metric.select.BIOMOD_EnsembleModeling to generate
warnings when ensemble models are expected to be run with <= 1
models.data.frame instead of matrix.dir.name can now be provided as project argument so
that results may be saved in a custom folder.predict with CTA algorithm and categorical
variables on raster is now possible.em.by = "algo" or em.by = "all") so that
evaluation uses the union of PA data sets instead of the whole
environmental space supplied.INT2S data format when on_0_1000 is set to
TRUE.get_[...], load_stored_object
and BIOMOD_LoadModels, instead of
get(load(...))) and the workflow within
get_[...] functions (use
load_stored_object and similar arguments such as
as.data.frame, full.name,
…).BIOMOD.ensemble.models.out and
BIOMOD.models.out objects
BIOMOD.ensemble.models.out object for
evaluations, variables importance and predictions..Models.save.objects in
BIOMOD_modeling to .fill_BIOMOD.models.out in
biomod2_internal.R.BIOMOD.ensemble.models.out and use
load_stored_object to directly get them within
get_[...] functions.BIOMOD_FormatingData, instead of throwing an error linked
to data.mask.on_0_1000 can now be passed without errors so
that projection may either be on a range from 0 to 1 or from 0 to 1000.
The latter option being more effective memory-wise.BIOMOD_EnsembleModeling so that em.by can not
be of length > 1..get_models_assembling so that it
did not confound MAXENT.Phillips2 with
MAXENT.Phillips when grouping models by algorithm in
BIOMOD_EnsembleModeling.get_predictions method for
BIOMOD.ensemble.models.out now accepts an
evaluation arg. Evaluation values, variables’ importance
and Calibration/Evaluation predictions for ensemble models are now
properly saved by BIOMOD_EnsembleModeling().prob.ci.inf et prob.ci.sup.BIOMOD_PresenceOnly now properly manage
NA.bm_PlotResponseCurves to only plot
show.variables.get_predictions.BIOMOD.projection.out now properly
works when asked for a subset of model.gbm package to its development version at rpatin/gbm can be
used. (see issue https://github.com/biomodhub/biomod2/issues/102)biomod2_classes files)BIOMOD_FormatingData functionBIOMOD_ModelingOptions functionBIOMOD_FormatingData : test class condition only a
first element (to deal with matrix / array
objects)BIOMOD_EnsembleForecasting for EMcv
model when only one single model was keptbetamultiplier parameter to tune MAXENT.Phillips
(Frank B. request)as.data.frame argument for
get_evaluations() function to enable formal and ensemble
models evaluation scores mergingMAXENT calculations (via java) (thanks to Burke
G.)do.stack argument is set to FALSE
biomod2 objects from a version
to the current oneFALSE by
defaultbiomod2 models objects (should
be predicted, evaluated, and you can do variables importance) the same
way than all formal biomod2 modelsbiomod2_projection
object: should be plotted…variable_importance functionmgcv to
gam to deal with memory (cache) over-consuming
(thanks to Burke G.)response.plot2 function (optimization + deal
with factorial variables)ProbDensFunc() function to package to produce nice
plots that show inter-models variabilityrasterVis dependency for nicer biomod2
plotsPA.dist.min and PA.dist.max are now
defined in meters when you work with unprojected rasters in disk pseudo
absences selectionBIOMOD_Modeling
col,
lty, data_species…)modeling.id arg (BIOMOD_Modeling)
for prevent from no wanted models overwriting and facilitate models
tests and comparisons (thanks Frank B.)biomod2 datasetpROC package dependencyRemoveProperly()
BIOMOD_LoadModels supports multiple models inputNA in evaluation table issue (*thanks
Frank B.)biomod2 are now defined as “biomod2
models objects” (own scaling models, own predict function, …).grd or .img)