Package: SDMtune 1.3.3

Sergio Vignali

SDMtune: Species Distribution Model Selection

User-friendly framework that enables the training and the evaluation of species distribution models (SDMs). The package implements functions for data driven variable selection and model tuning and includes numerous utilities to display the results. All the functions used to select variables or to tune model hyperparameters have an interactive real-time chart displayed in the 'RStudio' viewer pane during their execution.

Authors:Sergio Vignali [aut, cre], Arnaud Barras [aut], Veronika Braunisch [aut], Conservation Biology - University of Bern [fnd]

SDMtune_1.3.3.tar.gz
SDMtune_1.3.3.zip(r-4.7)SDMtune_1.3.3.zip(r-4.6)SDMtune_1.3.3.zip(r-4.5)
SDMtune_1.3.3.tgz(r-4.6-x86_64)SDMtune_1.3.3.tgz(r-4.6-arm64)SDMtune_1.3.3.tgz(r-4.5-x86_64)SDMtune_1.3.3.tgz(r-4.5-arm64)
SDMtune_1.3.3.tar.gz(r-4.7-arm64)SDMtune_1.3.3.tar.gz(r-4.7-x86_64)SDMtune_1.3.3.tar.gz(r-4.6-arm64)SDMtune_1.3.3.tar.gz(r-4.6-x86_64)
SDMtune_1.3.3.tgz(r-4.6-emscripten)
manual.pdf |manual.html✨
DESCRIPTION |NEWS
card.svg |card.png
SDMtune/json (API)

# Install 'SDMtune' in R:
install.packages('SDMtune', repos = c('https://consbiol-unibern.r-universe.dev', 'https://cloud.r-project.org'))

Bug tracker:https://github.com/consbiol-unibern/sdmtune/issues

Pkgdown/docs site:https://consbiol-unibern.github.io

Uses libs:
  • c++– GNU Standard C++ Library v3
Datasets:

On CRAN:

Conda:

hyperparameter-tuningspecies-distribution-modellingvariable-selectioncpp

7.79 score 30 stars 171 scripts 482 downloads 3 mentions 46 exports 41 dependencies

Last updated from:0daa7dedc8. Checks:13 OK. Indexed: yes.

TargetResultTimeFilesSyslog
linux-devel-arm64OK360
linux-devel-x86_64OK393
source / vignettesOK392
linux-release-arm64OK386
linux-release-x86_64OK376
macos-release-arm64OK368
macos-release-x86_64OK562
macos-oldrel-arm64OK233
macos-oldrel-x86_64OK421
windows-develOK375
windows-releaseOK344
windows-oldrelOK367
wasm-releaseOK184

Exports:addSamplesToBgaiccANNaucBRTcheckMaxentInstallationcombineCVconfMatrixcorVardoJkgetTunableArgsgridSearchMaxentmaxentThmaxentVarImpMaxnetmergeSWDmodelReportoptimizeModelplotplotCorplotJkplotPAplotPredplotResponseplotROCplotVarImppredictprepareSWDrandomFoldsrandomSearchreduceVarRFSDMmodelSDMmodel2MaxEntSDMmodelCVSDMtuneSWDswd2csvthinDatathresholdstraintrainValTesttssvarImpvarSel

Dependencies:clicodetoolscpp11dismofarverforeachgbmggplot2glmnetgluegtableisobanditeratorsjsonlitelabelinglatticelifecyclemagrittrMatrixmaxnetnnetR6randomForestrasterRColorBrewerRcppRcppEigenrlangrstudioapiS7scalesshapespstringistringrsurvivalterravctrsviridisLitewhiskerwithr

SDMtune - basic use
Prepare data for the analysis | Acquire environmental variables | Prepare presence and background locations | Create an SWD object | Explore the SWD object | Save an SWD object | Train a model | Train a model with default settings | Explore an SDMmodel object | Train a model changing the default settings | Make prediction | Create a distribution map | Plot a distribution map | Plot a presence/absence map | Evaluate a model | AUC | TSS | AICc | Training and testing | Cross validation | Spatial Cross Validation | References
Rendered frombasic-use.Rmdusingknitr::rmarkdown

Last update: 2023-05-01
Started: 2020-06-09

SDMtune - hyperparameter tuning
Training, validation and testing split | Check the effect of varying one hyperparameter | Tune hyperparameters | Random search | Optimize Model | Evaluate final model | Hyperparameters tuning with cross validation
Rendered fromhyper-tuning.Rmdusingknitr::rmarkdown

Last update: 2023-05-01
Started: 2020-06-09

SDMtune - presence absence models
Train the model | Tune model hyperparameters | Evaluate the final model | References
Rendered frompresence-absence.Rmdusingknitr::rmarkdown

Last update: 2023-05-01
Started: 2020-06-09

SDMtune - variable selection
Load data and train model | Variable importance | Permutation importance | Jackknife test for variable importance | Response curves | Model report | Data-driven variable selection | Remove highly correlated variables | Remove variables with low importance | References
Rendered fromvar-selection.Rmdusingknitr::rmarkdown

Last update: 2023-05-01
Started: 2020-06-09

Readme and manuals

Help Manual

Help pageTopics
Add Samples to BackgroundaddSamplesToBg
AICcaicc
Artificial Neural NetworkANN ANN-class show,ANN-method
AUCauc
Boosted Regression TreeBRT BRT-class show,BRT-method
Check Maxent InstallationcheckMaxentInstallation
Combine Cross Validation modelscombineCV
Confusion MatrixconfMatrix
Print Correlated VariablescorVar
Jackknife TestdoJk
Get Tunable ArgumentsgetTunableArgs
Grid SearchgridSearch
MaxentMaxent Maxent-class show,Maxent-method
MaxEnt ThresholdsmaxentTh
Maxent Variable ImportancemaxentVarImp
MaxnetMaxnet Maxnet-class show,Maxnet-method
Merge SWD ObjectsmergeSWD
Model ReportmodelReport
Optimize ModeloptimizeModel
Plot CorrelationplotCor
Plot Jackknife TestplotJk
Plot Presence Absence MapplotPA
Plot PredictionplotPred
Plot Response CurveplotResponse
Plot ROC curveplotROC
Plot Variable ImportanceplotVarImp
Predict ANNpredict,ANN-method
Predict BRTpredict,BRT-method
Predict Maxentpredict,Maxent-method
Predict Maxnetpredict,Maxnet-method
Predict RFpredict,RF-method
Predictpredict,SDMmodel-method
Predict for Cross Validationpredict,SDMmodelCV-method
Prepare an SWD objectprepareSWD
Create Random FoldsrandomFolds
Random SearchrandomSearch
Reduce VariablesreduceVar
Random ForestRF RF-class show,RF-method
SDMmodelSDMmodel SDMmodel-class show,SDMmodel-method
SDMmodel2MaxEntSDMmodel2MaxEnt
SDMmodelCVSDMmodelCV SDMmodelCV-class show,SDMmodelCV-method
SDMtune classplot,SDMtune,missing-method SDMtune SDMtune-class show,SDMtune-method
Sample With Datashow,SWD-method SWD SWD-class
SWD to csvswd2csv
Thin DatathinData
Thresholdsthresholds
Traintrain
Train, Validation and Test datasetstrainValTest
True Skill Statisticstss
Variable ImportancevarImp
Variable SelectionvarSel
Virtual SpeciesvirtualSp