probably1.0.3 package

Tools for Post-Processing Predicted Values

append_class_pred

Add a class_pred column

as_class_pred

Coerce to a class_pred object

cal_apply

Applies a calibration to a set of existing predictions

cal_binary_tables

Probability Calibration table

cal_estimate_beta

Uses a Beta calibration model to calculate new probabilities

cal_estimate_isotonic

Uses an Isotonic regression model to calibrate model predictions.

cal_estimate_isotonic_boot

Uses a bootstrapped Isotonic regression model to calibrate probabiliti...

cal_estimate_linear

Uses a linear regression model to calibrate numeric predictions

cal_estimate_logistic

Uses a logistic regression model to calibrate probabilities

cal_estimate_multinomial

Uses a Multinomial calibration model to calculate new probabilities

cal_plot_breaks

Probability calibration plots via binning

cal_plot_logistic

Probability calibration plots via logistic regression

cal_plot_regression

Regression calibration plots

cal_plot_windowed

Probability calibration plots via moving windows

cal_validate_beta

Measure performance with and without using Beta calibration

cal_validate_isotonic

Measure performance with and without using isotonic regression calibra...

cal_validate_isotonic_boot

Measure performance with and without using bagged isotonic regression ...

cal_validate_linear

Measure performance with and without using linear regression calibrati...

cal_validate_logistic

Measure performance with and without using logistic calibration

cal_validate_multinomial

Measure performance with and without using multinomial calibration

class_pred

Create a class prediction object

collect_metrics.cal_rset

Obtain and format metrics produced by calibration validation

collect_predictions.cal_rset

Obtain and format predictions produced by calibration validation

control_conformal_full

Controlling the numeric details for conformal inference

int_conformal_cv

Prediction intervals via conformal inference CV+

int_conformal_full

Prediction intervals via conformal inference

int_conformal_quantile

Prediction intervals via conformal inference and quantile regression

int_conformal_split

Prediction intervals via split conformal inference

is_class_pred

Test if an object inherits from class_pred

levels.class_pred

Extract class_pred levels

locate-equivocal

Locate equivocal values

make_class_pred

Create a class_pred vector from class probabilities

predict.int_conformal_full

Prediction intervals from conformal methods

probably-package

probably: Tools for Post-Processing Class Probability Estimates

reexports

Objects exported from other packages

reportable_rate

Calculate the reportable rate

required_pkgs.cal_object

S3 methods to track which additional packages are needed for specific ...

threshold_perf

Generate performance metrics across probability thresholds

Models can be improved by post-processing class probabilities, by: recalibration, conversion to hard probabilities, assessment of equivocal zones, and other activities. 'probably' contains tools for conducting these operations as well as calibration tools and conformal inference techniques for regression models.

  • Maintainer: Max Kuhn
  • License: MIT + file LICENSE
  • Last published: 2024-02-23