Package: cutpointr 1.2.0

cutpointr: Determine and Evaluate Optimal Cutpoints in Binary Classification Tasks

Estimate cutpoints that optimize a specified metric in binary classification tasks and validate performance using bootstrapping. Some methods for more robust cutpoint estimation are supported, e.g. a parametric method assuming normal distributions, bootstrapped cutpoints, and smoothing of the metric values per cutpoint using Generalized Additive Models. Various plotting functions are included. For an overview of the package see Thiele and Hirschfeld (2021) <doi:10.18637/jss.v098.i11>.

Authors:Christian Thiele [cre, aut]

cutpointr_1.2.0.tar.gz
cutpointr_1.2.0.zip(r-4.7)cutpointr_1.2.0.zip(r-4.6)cutpointr_1.2.0.zip(r-4.5)
cutpointr_1.2.0.tgz(r-4.6-x86_64)cutpointr_1.2.0.tgz(r-4.6-arm64)cutpointr_1.2.0.tgz(r-4.5-x86_64)cutpointr_1.2.0.tgz(r-4.5-arm64)
cutpointr_1.2.0.tar.gz(r-4.7-arm64)cutpointr_1.2.0.tar.gz(r-4.7-x86_64)cutpointr_1.2.0.tar.gz(r-4.6-arm64)cutpointr_1.2.0.tar.gz(r-4.6-x86_64)
cutpointr_1.2.0.tgz(r-4.6-emscripten)
manual.pdf |manual.html✨
DESCRIPTION |NEWS
card.svg |card.png
cutpointr/json (API)

# Install 'cutpointr' in R:
install.packages('cutpointr', repos = c('https://thie1e.r-universe.dev', 'https://cloud.r-project.org'))

Bug tracker:https://github.com/thie1e/cutpointr/issues

Uses libs:
  • c++– GNU Standard C++ Library v3
Datasets:
  • prostate_nodal - Nodal involvement and acid phosphatase levels in 53 prostate cancer patients
  • suicide - Suicide attempts and DSI sum scores of 532 subjects

On CRAN:

Conda:

bootstrappingcutpoint-optimizationroc-curvecpp

10.74 score 92 stars 2 packages 432 scripts 1.3k downloads 11 mentions 70 exports 34 dependencies

Last updated from:0a044b9448. Checks:11 ERROR, 2 OK. Indexed: yes.

TargetResultTimeFilesSyslog
linux-devel-arm64ERROR216
linux-devel-x86_64ERROR214
source / vignettesOK326
linux-release-arm64ERROR204
linux-release-x86_64ERROR240
macos-release-arm64ERROR192
macos-release-x86_64ERROR346
macos-oldrel-arm64ERROR160
macos-oldrel-x86_64ERROR417
windows-develERROR176
windows-releaseERROR179
windows-oldrelERROR203
wasm-releaseOK162

Exports:abs_d_ppv_npvabs_d_sens_specacc_constrainaccuracyadd_metricaucboot_ciboot_testcohens_kappacutpointcutpointrcutpointr_cutpointsF1_scorefalse_discovery_ratefalse_omission_ratefnfnrfpfprJaccardmaximize_boot_metricmaximize_gam_metricmaximize_loess_metricmaximize_metricmetric_constrainminimize_boot_metricminimize_gam_metricminimize_loess_metricminimize_metricmisclassification_costmulti_cutpointrnlrnpvoc_manualoc_meanoc_medianoc_youden_kerneloc_youden_normalodds_ratiop_chisquaredplot_cut_bootplot_cutpointrplot_metricplot_metric_bootplot_precision_recallplot_rocplot_sensitivity_specificityplot_xplrppvprecisionprod_ppv_npvprod_sens_specrecallrisk_ratiorocroc01sens_constrainsensitivityspec_constrainspecificitysum_ppv_npvsum_sens_spectntnrtotal_utilitytptpryouden

Dependencies:clicodetoolscpp11dplyrfarverforeachgenericsggplot2gluegridExtragtableisobanditeratorslabelinglifecyclemagrittrpillarpkgconfigpurrrR6RColorBrewerRcpprlangS7scalesstringistringrtibbletidyrtidyselectutf8vctrsviridisLitewithr

An introduction to cutpointr
Installation | Example | Features | Calculating cutpoints | Method functions for cutpoint estimation | Metric functions | Separate subgroups and bootstrapping | Parallelized bootstrapping | Additional features | Finding all cutpoints with acceptable performance | Manual and mean / median cutpoints | Nonstandard evaluation via tidyeval | Midpoints
Rendered fromcutpointr.Rmdusingknitr::rmarkdown

Last update: 2024-12-10
Started: 2017-04-07

Robust cutpoint estimation
More robust cutpoint estimation methods | Bootstrapped cutpoints | LOESS smoothing for selecting a cutpoint | Smoothing via Generalized Additive Models for selecting a cutpoint | Parametric method assuming normality | Nonparametric kernel method
Rendered fromcutpointr_estimation.Rmdusingknitr::rmarkdown

Last update: 2024-12-10
Started: 2022-01-18

cutpointr: Bootstrapping
Determine optimal cutpoints | Without bootstrapping: maximize_metric | Bootstrap cutpoints: maximize_boot_metric | Validate cutpoint optimization with bootstrapping | maximize_metric | maximize_boot_metric
Rendered fromcutpointr_bootstrapping.Rmdusingknitr::rmarkdown

Last update: 2022-04-13
Started: 2022-01-18

cutpointr benchmarks
Rendered fromcutpointr_benchmarks.Rmdusingknitr::rmarkdown

Last update: 2022-04-13
Started: 2022-01-18

Plotting with cutpointr
Flexible plotting function | Manual plotting
Rendered fromcutpointr_plotting.Rmdusingknitr::rmarkdown

Last update: 2022-04-13
Started: 2022-01-18

ROC curves with cutpointr
Calculating only the ROC curve | ROC curve and optimal cutpoint for multiple variables | Accessing data, roc_curve, and boot | Adding metrics to the result of cutpointr() or roc()
Rendered fromcutpointr_roc.Rmdusingknitr::rmarkdown

Last update: 2022-04-13
Started: 2022-01-18

User-defined functions for estimation methods and metrics
User-defined functions | method | metric
Rendered fromcutpointr_user_functions.Rmdusingknitr::rmarkdown

Last update: 2022-04-13
Started: 2022-01-18

Readme and manuals

Help Manual

Help pageTopics
Calculate the absolute difference of positive and negative predictive valueabs_d_ppv_npv
Calculate the absolute difference of sensitivity and specificityabs_d_sens_spec
Calculate accuracyaccuracy
Add metrics to a cutpointr or roc_cutpointr objectadd_metric add_metric.cutpointr add_metric.multi_cutpointr add_metric.roc_cutpointr
Calculate AUC from a roc_cutpointr or cutpointr objectauc auc.cutpointr auc.roc_cutpointr
Calculate bootstrap confidence intervals from a cutpointr objectboot_ci
Test for equivalence of a metricboot_test
Calculate Cohen's Kappacohens_kappa
Extract the cutpoints from a ROC curve generated by cutpointrcutpoint cutpoints
Determine and evaluate optimal cutpointscutpointr cutpointr.default cutpointr.numeric
The standard evaluation version of cutpointr (deprecated)cutpointr_
Calculate the F1-scoreF1_score
Calculate the false omission and false discovery ratefalse_discovery_rate false_omission_rate
Calculate the Jaccard IndexJaccard
Optimize a metric function in binary classification after bootstrappingmaximize_boot_metric minimize_boot_metric
Optimize a metric function in binary classification after smoothing via generalized additive modelsmaximize_gam_metric minimize_gam_metric
Optimize a metric function in binary classification after LOESS smoothingmaximize_loess_metric minimize_loess_metric
Optimize a metric function in binary classificationmaximize_metric minimize_metric
Metrics that are constrained by another metricacc_constrain metric_constrain sens_constrain spec_constrain
Calculate the misclassification costmisclassification_cost
Calculate optimal cutpoints and further statistics for multiple predictorsmulti_cutpointr
Calculate the negative predictive valuenpv
Set a manual cutpoint for use with cutpointroc_manual
Use the sample mean as cutpointoc_mean
Use the sample median as cutpointoc_median
Determine an optimal cutpoint maximizing the Youden-Index based on kernel smoothed densitiesoc_youden_kernel
Determine an optimal cutpoint for the Youden-Index assuming normal distributionsoc_youden_normal
Calculate the odds ratioodds_ratio
Calculate the p-value of a chi-squared testp_chisquared
Plot the bootstrapped distribution of optimal cutpoints from a cutpointr objectplot_cut_boot
General purpose plotting function for cutpointr or roc_cutpointr objectsplot_cutpointr
Plot a metric over all possible cutoffs from a cutpointr objectplot_metric
Plot the bootstrapped metric distribution from a cutpointr objectplot_metric_boot
Precision recall plot from a cutpointr objectplot_precision_recall
Plot ROC curve from a cutpointr or roc_cutpointr objectplot_roc plot_roc.cutpointr plot_roc.roc_cutpointr
Sensitivity and specificity plot from a cutpointr objectplot_sensitivity_specificity
Plot the distribution of the independent variable per class from a cutpointr objectplot_x
Plot cutpointr objectsplot.cutpointr
Plotting multi_cutpointr objects is currently not supportedplot.multi_cutpointr
Plot ROC curve from a cutpointr or roc_cutpointr objectplot.roc_cutpointr
Calculate the positive or negative likelihood rationlr plr
Calculate the positive predictive valueppv
Calculate precisionprecision
Predict using a cutpointr objectpredict.cutpointr
Print cutpointr objectsprint.cutpointr
Print multi_cutpointr objectsprint.multi_cutpointr
Calculate the product of positive and negative predictive valueprod_ppv_npv
Calculate the product of sensitivity and specificityprod_sens_spec
Nodal involvement and acid phosphatase levels in 53 prostate cancer patientsprostate_nodal
Calculate recallrecall
Calculate the risk ratio (relative risk)risk_ratio
Calculate a ROC curveroc
Calculate the distance between points on the ROC curve and (0,1)roc01
Calculate sensitivitysensitivity
Calculate specificityspecificity
Suicide attempts and DSI sum scores of 532 subjectssuicide
Calculate the sum of positive and negative predictive valuesum_ppv_npv
Calculate the sum of sensitivity and specificitysum_sens_spec
Calculate the total utilitytotal_utility
Extract number true / false positives / negativesfn fp tn tp
Calculate true / false positive / negative ratefnr fpr tnr tpr
Calculate bandwidth for LOESS smoothing of metric functions by rule of thumbuser_span_cutpointr
Calculate the Youden-Indexyouden