Package: kindling 0.3.1.9000

kindling: Higher-Level Interface of 'torch' Package to Auto-Train Neural Networks

Provides a higher-level interface to the 'torch' package for defining, training, and fine-tuning neural networks through code generation. The package supports several architectures, including feedforward (multi-layer perceptron) and recurrent neural networks (RNN, LSTM, GRU), while reducing boilerplate 'torch' code. Model training methods also bridge to machine learning frameworks in R, particularly the 'tidymodels' ecosystem, including 'parsnip' model specifications, workflows, recipes, and tuning tools.

Authors:Joshua Marie [aut, cre], Antoine Soetewey [aut]

kindling_0.3.1.9000.tar.gz
kindling_0.3.1.9000.zip(r-4.7)kindling_0.3.1.9000.zip(r-4.6)kindling_0.3.1.9000.zip(r-4.5)
kindling_0.3.1.9000.tgz(r-4.6-any)kindling_0.3.1.9000.tgz(r-4.5-any)
kindling_0.3.1.9000.tar.gz(r-4.7-any)kindling_0.3.1.9000.tar.gz(r-4.6-any)
kindling_0.3.1.9000.tgz(r-4.6-emscripten)
manual.pdf |manual.html
DESCRIPTION |NEWS
card.svg |card.png
kindling/json (API)

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

Bug tracker:https://github.com/joshuamarie/kindling/issues

Pkgdown/docs site:https://kindling.joshuamarie.com

On CRAN:

Conda:

neural-networkstidymodelstorch

7.21 score 27 stars 11 scripts 278 downloads 36 exports 115 dependencies

Last updated from:0edd360f6a. Checks:9 OK. Indexed: yes.

TargetResultTimeFilesSyslog
linux-devel-x86_64OK220
source / vignettesOK253
linux-release-x86_64OK210
macos-release-arm64OK174
macos-oldrel-arm64OK136
windows-develOK151
windows-releaseOK158
windows-oldrelOK155
wasm-releaseOK171

Exports:.i.in.is_output.layer.outact_funsactivationsargsautoplot_diagnosticsbiasbidirectionalearly_stopffnnffnn_generatorffnn_wrappergarsongrid_depthhidden_neuronsmlp_kindlingn_hlayersnew_act_fnnn_archnn_module_generatoroldenoptimizeroutput_activationplot_diagnosticsrnnrnn_generatorrnn_kindlingrnn_wrappertable_summarytrain_nntrain_nn_wrappertrain_nnsnipvalidation_split

Dependencies:base64encbitbit64bslibcachemcallrclasscliclockcodetoolscorocpp11data.tabledescdiagramdialsDiceDesigndigestdplyrevaluatefarverfastmapfontawesomefsfurrrfuturefuture.applyGauProgenericsggplot2globalsgluegowergtablehardhathighrhtmltoolsipredisobandjquerylibjsonliteKernSmoothknitrlabelinglatticelavalbfgslifecyclelistenvlubridatemagrittrMASSMatrixmemoisemimemixoptmodelenvNeuralNetToolsnnetnumDerivotelparallellyparsnippillarpkgconfigplyrprettyunitsprocessxprodlimprogressrpspurrrR6rappdirsRColorBrewerRcppRcppArmadillorecipesreshape2rlangrmarkdownrpartrsampleS7safetensorssassscalessfdshapeslidersparsevctrssplitfngrSQUAREMstringistringrsurvivaltailortibbletidyrtidyselecttimechangetimeDatetinytextorchtunetzdbutf8vctrsviridisLitewarpwithrworkflowsxfunyamlyardstick

Custom Activation Function
Rationale | Function to use | Basic Usage | Using Custom Activations in a Model | Skipping the Dry-Run Probe | Naming Your Custom Activation | Error Handling | Summary

Last update: 2026-03-03
Started: 2026-03-01

Special Cases: Linear and Logistic Regression
What's so special about | Setup | Linear Regression as a Special Case | Data | Fitting the model | Evaluation | Comparison with lm() | Logistic Regression as a Special Case | Binary Logistic Regression | Comparison with glm() / nnet::multinom()

Last update: 2026-03-02
Started: 2026-02-24

Tuning Capabilities
Rationale | Custom grid creation | Setup | Usage | Data Preparation | Using grid_depth() | Tuning | Inspect | Visualizing Results | Finalizing the Model | Evaluating on the test set | A Note on Parametric Activations

Last update: 2026-03-02
Started: 2026-02-12

Getting Started with kindling
Introduction | Installation | Before using | Four Levels of Interaction | Level 1: Code Generation | Level 2: Direct Training | Level 3: tidymodels Integration | Learn More

Last update: 2026-01-26
Started: 2026-01-16

Similar packages and comparison
Similar packages | Package Comparison | Complementary Use

Last update: 2026-01-25
Started: 2026-01-25

Readme and manuals

Help Manual

Help pageTopics
Activation Functions Specification Helperact_funs
Activation Function Arguments Helperargs
Plot prediction diagnostics for a fitted neural networkautoplot_diagnostics plot_diagnostics
Early Stopping Specificationearly_stop
Generalized Neural Network Trainergen-nn-train train_nn train_nn.data.frame train_nn.dataset train_nn.default train_nn.formula train_nn.matrix
Depth-Aware Grid Generation for Neural Networksgrid_depth grid_depth.default grid_depth.list grid_depth.model_spec grid_depth.param grid_depth.parameters grid_depth.workflow
Base models for Neural Network Training in kindlingffnn kindling-basemodels rnn
Variable Importance Methods for kindling Modelsgarson.ffnn_fit kindling-varimp olden.ffnn_fit vi_model.ffnn_fit
Layer argument pronouns for formula-based specifications.i .in .is_output .layer .out layer_prs
Multi-Layer Perceptron (Feedforward Neural Network) via kindlingmlp_kindling
Custom Activation Function Constructornew_act_fn
Architecture specification for train_nn()nn_arch
Functions to generate 'nn_module' (language) expressionffnn_generator nn_gens rnn_generator
Generalized Neural Network Module Expression Generatornn_module_generator
Ordinal Suffixes Generatorordinal_gen
Recurrent Neural Network via kindlingrnn_kindling
Summarize and Display a Two-Column Data Frame as a Formatted Tabletable_summary
Parsnip Interface of 'train_nn()'train_nnsnip