Learn R Programming

⚠️There's a newer version (1.3.0) of this package.Take me there.

dials

Overview

This package contains infrastructure to create and manage values of tuning parameters for the tidymodels packages. If you are looking for how to tune parameters in tidymodels, please look at the tune package and tidymodels.org.

The name reflects the idea that tuning predictive models can be like turning a set of dials on a complex machine under duress.

Installation

You can install the released version of dials from CRAN with:

install.packages("dials")

You can install the development version from Github with:

# install.packages("pak")
pak::pak("tidymodels/dials")

Contributing

Please note that the dials project is released with a Contributor Code of Conduct. By contributing to this project, you agree to abide by its terms.

For questions and discussions about tidymodels packages, modeling, and machine learning, please post on RStudio Community.
If you think you have encountered a bug, please submit an issue.
Either way, learn how to create and share a reprex (a minimal, reproducible example), to clearly communicate about your code.
Check out further details on contributing guidelines for tidymodels packages and how to get help.

Copy Link

Version

Install

install.packages('dials')

Monthly Downloads

30,765

Version

1.2.1

License

MIT + file LICENSE

Issues

Pull Requests

Stars

Forks

Repository

https://github.com/tidymodels/dials

Homepage

https://dials.tidymodels.org

Maintainer

Last Published

February 22nd, 2024

Functions in dials (1.2.1)

grid_max_entropy

Space-filling parameter grids

Near-zero variance parameters

Class for converting parameter values back and forth to the unit range

Parameter for the effective minimum distance between embedded points

Degrees of freedom (integer)

Parameters for exponents

Maximum number of retained tokens

Number of unique values for pre-processing

Mixture of penalization terms

Number of knots (integer)

Number of Computation Runs

Parameters for possible engine parameters for earth models

Neural network parameters

Number of cut-points for binning

Tools for creating new parameter objects

prior_slab_dispersion

Bayesian PCA parameters

MARS pruning methods

Parameter to determine number of tokens in ngram

Parameters for class-imbalance sampling

Create grids of tuning parameters

Parameters for possible engine parameters for randomForest

Number of neighbors

Proportion of Randomly Selected Predictors

select_features

Parameter to enable feature selection

harmonic_frequency

Harmonic Frequency

pull_dials_object

Return a dials parameter object associated with parameters

shrinkage_correlation

Parameters for possible engine parameters for sda models

Functions to finalize data-specific parameter ranges

scheduler-param

Parameters for neural network learning rate schedulers These parameters are used for constructing neural network models.

regularization_method

Estimation methods for regularized models

Tools for working with parameter ranges

Gradient descent momentum parameter

Number of randomly sampled predictors

Initialization method for UMAP

Kernel parameters

Possible engine parameters for lightbgm

Objects exported from other packages

Term frequency weighting methods

Kernel functions for distance weighting

Parameter for "double normalization" when creating token counts

Amount of supervision parameter

vocabulary_size

Number of tokens in vocabulary

Survival Model Link Function

regularization_factor

Parameters for possible engine parameters for ranger

Text hashing parameters

Proportion of predictors

validation_set_prop

Proportion of data used for validation

Parameter functions related to tree- and rule-based models.

Amount of regularization/penalization

Word frequencies for removal

Number of new features

Number of Clusters

Amount of Trimming

Succinct summary of parameter objects

Tools for working with parameter values

General thresholding parameter

scale_pos_weight

Parameters for possible engine parameters for xgboost

Parameter for the moving window size

Kernel Smoothness

parameters_constr

Construct a new parameter set object

Information on tuning parameters within an object

Rolling summary statistic for moving windows

Placeholder for unknown parameter values

Early stopping parameter

Parametric distributions for censored data

update.parameters

Update a single parameter in a parameter set

confidence_factor

Parameters for possible engine parameters for C5.0

Laplace correction parameter

conditional_min_criterion

Parameters for possible engine parameters for partykit models

Activation functions between network layers

dials: Tools for working with tuning parameters

Support vector machine parameters

Parameter to determine which neighbors to use

Parameters for possible engine parameters for Cubist

Parameters for class weights for imbalanced problems

Minkowski distance parameter

Parameters for BART models These parameters are used for constructing Bayesian adaptive regression tree (BART) models.

adjust_deg_free

Parameters to adjust effective degrees of freedom