User related messages

• improve and generalize biomod2 test functions (prefixed .fun_test[...])
• standardize / uniformize all cat, message, stop and warning messages

Other improvements

• update default parameters in functions (set to NULL if possible)
• add seed.val parameter in bm_CrossValidation
• add check in BIOMOD_Projection and BIOMOD_EnsembleForecasting to deal with single model projection failing
• fix kfold selection for non-binary data in bm_CrossValidation
• add check in bm_PseudoAbsences to select presences in PA.user.table when user.defined strategy
• fix bm_Tuning for DNN : multiple sets (PA x RUN), hidden width and depth parameters
• fix EMwmean for abundance models using RMSE, MSE, MAE or Max_error as selection metrics

Full changelog can found here

Abundance

• data.type parameter (binary, count, multiclass, ordinal, relative, abundance)
• associated abundance models
• add DataSTOC data set for examples

Other improvements

• update BIOMOD_RangeSize function with auxiliary function
• add BIOMOD_Report function
• add DNN model
• update xgboost package

Full changelog can found here

Add RFd : Random Forest with a down-sampling method.

Optimizing of Bigboss and Tuning
Some options for OptionsBigboss have been modified (concerns only ANN, CTA and RF models)
Some changes for the tuning ranges.

Few others changes, notably the possibility to give the same options to all datasets with “for_all_datasets” in bm_ModelingOptions.

Full changelog can found here

Modeling options are now automatically retrieved from single models functions,
normally allowing the use of all arguments taken into account by these functions,
with the help of ModelsTable and OptionsBigboss datasets

Tuning has been cleaned up

In consequence, BIOMOD_ModelingOptions and BIOMOD_Tuning functions become secundary functions (bm_ModelingOptions and bm_Tuning), and modeling options can be directly built through BIOMOD_Modeling function

Full changelog can found here

Adding xgboost to the available models

Other Changes

• Changed CV.do.full.models default value to FALSE
• Bugfix

Full changelog can found here

Cleaning and expanding features for pseudo-absences and cross-validation

• Improved pseudo-absence management: it is now possible to have pseudo-absence dataset of different size and algorithm can be setup to run on different pseudo-absence dataset (with models.pa argument in BIOMOD_Modeling).
• Rework and harmonization of cross-validation function. 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.
• Lots of internal changes improving package functioning and solving bugs

Full changelog can found here

Improvement release

• Added support for .tif
• Improved plot and summary for output of BIOMOD_FormatingData
• Improved plot for output of BIOMOD_Projection
• Binary and filtered transformation are now properly stored and can be access with get_predictions(x, metric.binary = "TSS") (resp. metric.filter)
• Lots of internal changes improving package functioning and solving bugs

Full changelog can found here

First release with primary support for terra
This is the first release that have a primary support for package terra. Older version are based on raster.
Full changelog can found here from version 4.2-0 to 4.2-1

Last stable release for 2022
This is the last release that have a primary support for package raster. Newer version are based on terra.
Full changelog can found here from version 4.1-2 to 4.1-3

Major Release

• clean all functions, reorganize files, remove old / unused functions
• standardize function names and parameter names
• update roxygen2 documentation for all functions, including examples
• create github website to host documentation, examples, vignettes, news

Full changelog can found here for version 4.0 to 4.1-1