Package: dials 1.4.0.9000

Hannah Frick

dials: Tools for Creating Tuning Parameter Values

Many models contain tuning parameters (i.e. parameters that cannot be directly estimated from the data). These tools can be used to define objects for creating, simulating, or validating values for such parameters.

Authors:Max Kuhn [aut], Hannah Frick [aut, cre], Posit Software, PBC [cph, fnd]

dials_1.4.0.9000.tar.gz
dials_1.4.0.9000.zip(r-4.5)dials_1.4.0.9000.zip(r-4.4)dials_1.4.0.9000.zip(r-4.3)
dials_1.4.0.9000.tgz(r-4.5-any)dials_1.4.0.9000.tgz(r-4.4-any)dials_1.4.0.9000.tgz(r-4.3-any)
dials_1.4.0.9000.tar.gz(r-4.5-noble)dials_1.4.0.9000.tar.gz(r-4.4-noble)
dials_1.4.0.9000.tgz(r-4.4-emscripten)dials_1.4.0.9000.tgz(r-4.3-emscripten)
dials.pdf |dials.html✨
dials/json (API)
NEWS

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

Bug tracker:https://github.com/tidymodels/dials/issues

Pkgdown site:https://dials.tidymodels.org

On CRAN:

Conda:

14.22 score 114 stars 52 packages 426 scripts 29k downloads 7 mentions 166 exports 28 dependencies

Last updated 25 days agofrom:d42b138b91. Checks:4 OK, 4 NOTE. Indexed: yes.

TargetResultLatest binary
Doc / VignettesOKFeb 14 2025
R-4.5-winOKFeb 14 2025
R-4.5-macOKFeb 14 2025
R-4.5-linuxOKFeb 14 2025
R-4.4-winNOTEFeb 14 2025
R-4.4-macNOTEFeb 14 2025
R-4.3-winNOTEFeb 14 2025
R-4.3-macNOTEFeb 14 2025

Exports:activationactivation_2adjust_deg_freeall_neighborsbatch_sizebufferclass_weightsconditional_min_criterionconditional_test_statisticconditional_test_typeconfidence_factorcostcost_complexitydeg_freedegreedegree_intdiagonal_covariancedist_powerdropoutencode_unitepochsextract_parameter_dialsextrapolationfinalizefreq_cutfuzzy_thresholdingget_batch_sizesget_log_pget_nget_n_fracget_n_frac_rangeget_pget_rbf_rangegrid_latin_hypercubegrid_max_entropygrid_randomgrid_regulargrid_space_fillingharmonic_frequencyhas_unknownshidden_unitshidden_units_2initial_umapis_unknownkernel_offsetLaplacelearn_rateloss_reductionlower_limitlower_quantilemax_nodesmax_num_termsmax_rulesmax_timesmax_tokensmin_distmin_nmin_timesmin_uniquemixturemomentummtrymtry_longmtry_propneighborsnew_qual_paramnew_quant_paramno_global_pruningnum_breaksnum_clustersnum_compnum_hashnum_knotsnum_leavesnum_random_splitsnum_runsnum_termsnum_tokensover_ratioparametersparameters_constrpenaltypenalty_L1penalty_L2predictor_proppredictor_winnowingprior_mixture_thresholdprior_outcome_rangeprior_slab_dispersionprior_terminal_node_coefprior_terminal_node_expoprod_degreepruneprune_methodrange_getrange_setrange_validateranger_class_rulesranger_reg_rulesranger_split_rulesrate_decayrate_initialrate_largestrate_reductionrate_schedulerate_step_sizerate_stepsrbf_sigmaregularization_factorregularization_methodregularize_depthrule_bandssample_propsample_sizescale_factorscale_pos_weightselect_featuresshrinkage_correlationshrinkage_frequenciesshrinkage_variancesigned_hashsignificance_thresholdsmoothnessspline_degreesplitting_rulestop_itersummary_statsurv_distsurvival_linksvm_margintarget_weightthresholdtokentree_depthtreestrim_amountunbiased_rulesunder_ratiounique_cutunknownupper_limitvalidation_set_propvalue_inversevalue_samplevalue_seqvalue_setvalue_transformvalue_validatevalues_activationvalues_initial_umapvalues_prune_methodvalues_regularization_methodvalues_schedulervalues_summary_statvalues_surv_distvalues_survival_linkvalues_test_statisticvalues_test_typevalues_tokenvalues_weight_funcvalues_weight_schemevocabulary_sizeweightweight_funcweight_schemewindow_size

Dependencies:clicolorspaceDiceDesigndplyrfansifarvergenericsgluehardhatlabelinglifecyclemagrittrmunsellpillarpkgconfigpurrrR6RColorBrewerrlangscalessfdsparsevctrstibbletidyselectutf8vctrsviridisLitewithr

Working with Tuning Parameters

Rendered fromdials.Rmdusingknitr::rmarkdownon Feb 14 2025.

Last update: 2024-10-17
Started: 2021-11-09

Readme and manuals

Help Manual

Help pageTopics
Activation functions between network layersactivation activation_2 values_activation
Parameters to adjust effective degrees of freedomadjust_deg_free
Parameter to determine which neighbors to useall_neighbors
Parameters for BART models These parameters are used for constructing Bayesian adaptive regression tree (BART) models.bart-param prior_outcome_range prior_terminal_node_coef prior_terminal_node_expo
Buffer sizebuffer
Parameters for class weights for imbalanced problemsclass_weights
Parameters for possible engine parameters for partykit modelsconditional_min_criterion conditional_test_statistic conditional_test_type values_test_statistic values_test_type
Parameters for possible engine parameters for C5.0confidence_factor fuzzy_thresholding no_global_pruning predictor_winnowing rule_bands
Support vector machine parameterscost svm_margin
Degrees of freedom (integer)deg_free
Parameters for exponentsdegree degree_int prod_degree spline_degree
Minkowski distance parameterdist_power
Neural network parametersbatch_size dropout epochs hidden_units hidden_units_2
Parameters for possible engine parameters for Cubistextrapolation max_rules unbiased_rules
Functions to finalize data-specific parameter rangesfinalize finalize.default finalize.list finalize.logical finalize.param finalize.parameters get_batch_sizes get_log_p get_n get_n_frac get_n_frac_range get_p get_rbf_range
Near-zero variance parametersfreq_cut unique_cut
Create grids of tuning parametersgrid_random grid_random.list grid_random.param grid_random.parameters grid_regular grid_regular.list grid_regular.param grid_regular.parameters
Space-filling parameter gridsgrid_space_filling grid_space_filling.list grid_space_filling.param grid_space_filling.parameters
Harmonic Frequencyharmonic_frequency
Initialization method for UMAPinitial_umap values_initial_umap
Laplace correction parameterLaplace
Learning ratelearn_rate
Parameters for possible engine parameters for randomForestmax_nodes
Parameters for possible engine parameters for earth modelsmax_num_terms
Word frequencies for removalmax_times min_times
Maximum number of retained tokensmax_tokens
Parameter for the effective minimum distance between embedded pointsmin_dist
Number of unique values for pre-processingmin_unique
Mixture of penalization termsmixture
Gradient descent momentum parametermomentum
Number of randomly sampled predictorsmtry mtry_long
Proportion of Randomly Selected Predictorsmtry_prop
Number of neighborsneighbors
Tools for creating new parameter objectsnew-param new_qual_param new_quant_param
Number of cut-points for binningnum_breaks
Number of Clustersnum_clusters
Number of new featuresnum_comp num_terms
Text hashing parametersnum_hash signed_hash
Number of knots (integer)num_knots
Possible engine parameters for lightbgmnum_leaves
Number of Computation Runsnum_runs
Parameter to determine number of tokens in ngramnum_tokens
Parameters for class-imbalance samplingover_ratio under_ratio
Information on tuning parameters within an objectparameters parameters.default parameters.list parameters.param
Amount of regularization/penalizationpenalty
Proportion of predictorspredictor_prop
Bayesian PCA parametersprior_mixture_threshold prior_slab_dispersion
MARS pruning methodsprune_method values_prune_method
Limits for the range of predictionslower_limit range_limits upper_limit
Tools for working with parameter rangesrange_get range_set range_validate
Kernel parameterskernel_offset rbf_sigma scale_factor
Parameters for possible engine parameters for rangerlower_quantile num_random_splits ranger_class_rules ranger_reg_rules ranger_split_rules regularization_factor regularize_depth significance_threshold splitting_rule
Estimation methods for regularized modelsregularization_method values_regularization_method
Parameters for possible engine parameters for xgboostpenalty_L1 penalty_L2 scale_pos_weight
Parameters for neural network learning rate schedulers These parameters are used for constructing neural network models.rate_decay rate_initial rate_largest rate_reduction rate_schedule rate_steps rate_step_size scheduler-param values_scheduler
Parameter to enable feature selectionselect_features
Parameters for possible engine parameters for sda modelsdiagonal_covariance shrinkage_correlation shrinkage_frequencies shrinkage_variance
Kernel Smoothnesssmoothness
Early stopping parameterstop_iter
Rolling summary statistic for moving windowssummary_stat values_summary_stat
Parametric distributions for censored datasurv_dist values_surv_dist
Survival Model Link Functionsurvival_link values_survival_link
Amount of supervision parametertarget_weight
General thresholding parameterthreshold
Token typestoken values_token
Parameter functions related to tree- and rule-based models.cost_complexity loss_reduction min_n prune sample_prop sample_size trees tree_depth
Amount of Trimmingtrim_amount
Placeholder for unknown parameter valueshas_unknowns is_unknown unknown
Update a single parameter in a parameter setupdate.parameters
Proportion of data used for validationvalidation_set_prop
Tools for working with parameter valuesvalue_inverse value_sample value_seq value_set value_transform value_validate
Number of tokens in vocabularyvocabulary_size
Parameter for '"double normalization"' when creating token countsweight
Kernel functions for distance weightingvalues_weight_func weight_func
Term frequency weighting methodsvalues_weight_scheme weight_scheme
Parameter for the moving window sizewindow_size