Package: sperrorest 3.0.5

Alexander Brenning

sperrorest: Perform Spatial Error Estimation and Variable Importance Assessment

Implements spatial error estimation and permutation-based variable importance measures for predictive models using spatial cross-validation and spatial block bootstrap.

Authors:Alexander Brenning [aut, cre], Patrick Schratz [aut], Tobias Herrmann [ctb]

sperrorest_3.0.5.tar.gz
sperrorest_3.0.5.zip(r-4.7)sperrorest_3.0.5.zip(r-4.6)sperrorest_3.0.5.zip(r-4.5)
sperrorest_3.0.5.tgz(r-4.6-any)sperrorest_3.0.5.tgz(r-4.5-any)
sperrorest_3.0.5.tar.gz(r-4.7-any)sperrorest_3.0.5.tar.gz(r-4.6-any)
sperrorest_3.0.5.tgz(r-4.6-emscripten)
manual.pdf |manual.html✨
DESCRIPTION |NEWS
card.svg |card.png
sperrorest/json (API)

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

Bug tracker:https://github.com/giscience-fsu/sperrorest/issues

Pkgdown/docs site:https://giscience-fsu.github.io

On CRAN:

Conda:

cross-validationmachine-learningspatial-statisticsspatio-temporal-modelingstatistical-learning

7.11 score 20 stars 58 scripts 3.7k downloads 3 mentions 32 exports 30 dependencies

Last updated from:b4d2a1426b. Checks:7 NOTE, 2 OK. Indexed: yes.

TargetResultTimeFilesSyslog
linux-devel-x86_64NOTE210
source / vignettesOK296
linux-release-x86_64NOTE200
macos-release-arm64NOTE193
macos-oldrel-arm64NOTE173
windows-develNOTE141
windows-releaseNOTE153
windows-oldrelNOTE168
wasm-releaseOK134

Exports:add.distanceas.represamplingas.resamplingas.tilenamedataset_distanceerr_defaultget_small_tilesis_represamplingis.resamplingpartition_cvpartition_cv_stratpartition_discpartition_factorpartition_factor_cvpartition_kmeanspartition_loopartition_tilesremove_missing_levelsrepresampling_bootstraprepresampling_disc_bootstraprepresampling_factor_bootstraprepresampling_kmeans_bootstraprepresampling_tile_bootstrapresample_factorresample_strat_uniformresample_uniformrunfoldsrunrepssperroresttile_neighborstransfer_parallel_outputvalidate.resampling

Dependencies:bitopscaToolsclicodetoolsdigestdplyrfuturefuture.applygenericsglobalsgluegplotsgtoolsKernSmoothlifecyclelistenvmagrittrparallellypillarpkgconfigR6rlangROCRstringistringrtibbletidyselectutf8vctrswithr

Spatial Modeling Using Statistical Learning Techniques
Introduction | Data and Packages | Modeling | Linear Discriminant Analysis (LDA) | Classification Tree | RandomForest | Cross-Validation Estimation of Predictive Performance | Usage Advice | References
Rendered fromspatial-modeling-use-case.Rmdusingknitr::rmarkdown

Last update: 2021-11-19
Started: 2017-06-11

Custom Predict and Model Functions
Introduction | User-defined Model Functions | Problem | Solution | User-defined Predict Functions | randomForest | svm
Rendered fromcustom-pred-and-model-functions.Rmdusingknitr::rmarkdown

Last update: 2020-03-15
Started: 2017-06-11

Readme and manuals

Help Manual

Help pageTopics
Spatial Error Estimation and Variable Importancesperrorest-package
Add distance information to resampling objectsadd.distance add.distance.represampling add.distance.resampling
Resampling objects with repetition, i.e. sets of partitionings or bootstrap samplesas.represampling as.represampling.list as.represampling_list is_represampling print.represampling represampling
Resampling objects such as partitionings or bootstrap samplesas.resampling as.resampling.default as.resampling.factor as.resampling.list as.resampling_default as.resampling_list is.resampling print.resampling resampling validate.resampling
Alphanumeric tile namesas.character.tilename as.numeric.tilename as.tilename as.tilename.character as.tilename.numeric as.tilename_character as.tilename_numeric print.tilename tilename
Calculate mean nearest-neighbour distance between point datasetsdataset_distance
Default error functionerr_default
Identify small partitions that need to be fixed.get_small_tiles
Partition the data for a (non-spatial) cross-validationpartition_cv
Partition the data for a stratified (non-spatial) cross-validationpartition_cv_strat
Leave-one-disc-out cross-validation and leave-one-out cross-validationpartition_disc partition_loo
Partition the data for a (non-spatial) leave-one-factor-out cross-validation based on a given, fixed partitioningpartition_factor
Partition the data for a (non-spatial) k-fold cross-validation at the group levelpartition_factor_cv
Partition samples spatially using k-means clustering of the coordinatespartition_kmeans
Partition the study area into rectangular tilespartition_tiles
Plot spatial resampling objectsplot.represampling plot.resampling
Non-spatial bootstrap resamplingrepresampling_bootstrap
Overlapping spatial block bootstrap using circular blocksrepresampling_disc_bootstrap
Bootstrap at an aggregated levelrepresampling_factor_bootstrap
Spatial block bootstrap using rectangular blocksrepresampling_tile_bootstrap
Draw uniform random (sub)sample at the group levelresample_factor
Draw stratified random sampleresample_strat_uniform
Draw uniform random (sub)sampleresample_uniform
Perform spatial error estimation and variable importance assessmentsperrorest
title Summary statistics for a resampling objectssummary.represampling summary.resampling
Summarize error statistics obtained by sperrorestsummary.sperroresterror
Summarize variable importance statistics obtained by sperrorestsummary.sperrorestimportance
Summary and print methods for sperrorest resultsprint.sperrorest print.sperrorestbenchmarks print.sperroresterror print.sperrorestimportance print.sperrorestpackageversion print.sperrorestreperror summary.sperrorest summary.sperrorestreperror
Determine the names of neighbouring tiles in a rectangular patterntile_neighbors