Tidy Tuning Tools
Augment data with holdout predictions
Plot tuning search results
Tools for selecting metrics and evaluation times
Obtain and format results produced by tuning functions
Calculate and format metrics from tuning functions
Compute average confusion matrix across resamples
Control aspects of the Bayesian search process
Control aspects of the grid search process
Control aspects of the last fit process
Use same scale for plots of observed vs predicted values
Save most recent results to search path
Determine if case weights should be passed on to yardstick
Get colors for tune text.
Exponential decay function
Extract elements of tune objects
Remove some tuning parameter results
Splice final parameters into objects
Fit a model to the numerically optimal configuration
Fit multiple models via resampling
Get time for analysis of dynamic survival metrics
Bootstrap confidence intervals for performance metrics
Internal functions to help use parallel processing
Fit the final best model to the training set and evaluate the test set
Quietly load package namespace
Merge parameter grid values into objects
Write a message that respects the line width
Determine the minimum set of model fits
Determine names of the outcome data in a workflow
Support for parallel processing in tune
Acquisition function for scoring parameter combinations
Objects exported from other packages
Schedule a grid
Investigate best tuning parameters
Display distinct errors from tune objects
Various accessor functions
Bayesian optimization of model parameters.
Model tuning via grid search
Internal functions used by other tidymodels packages
The ability to tune models is important. 'tune' contains functions and classes to be used in conjunction with other 'tidymodels' packages for finding reasonable values of hyper-parameters in models, pre-processing methods, and post-processing steps.
Useful links