{"Package":"brulee","Title":"High-Level Modeling Functions with 'torch'","Version":"0.4.0","Authors@R":"c(\nperson(\"Max\", \"Kuhn\", , \"max@posit.co\", role = c(\"aut\", \"cre\"),\ncomment = c(ORCID = \"0000-0003-2402-136X\")),\nperson(\"Daniel\", \"Falbel\", , \"daniel@posit.co\", role = \"aut\"),\nperson(given = \"Posit Software, PBC\", role = c(\"cph\", \"fnd\"))\n)","Description":"Provides high-level modeling functions to define and train\nmodels using the 'torch' R package. Models include linear, logistic,\nand multinomial regression as well as multilayer perceptrons.","License":"MIT + file LICENSE","URL":"https://github.com/tidymodels/brulee,\nhttps://brulee.tidymodels.org/","BugReports":"https://github.com/tidymodels/brulee/issues","Imports":"cli, coro (>= 1.0.1), dplyr, generics, ggplot2, glue, hardhat,\nrlang (>= 1.1.1), stats, tibble, torch (>= 0.13.0), utils","Suggests":"covr, modeldata, purrr, recipes, spelling, testthat,\nyardstick","Config/Needs/website":"tidyverse/tidytemplate","Config/testthat/edition":"3","Encoding":"UTF-8","Language":"en-US","RoxygenNote":"7.3.2","NeedsCompilation":"no","Packaged":"2025-01-30 21:36:04 UTC; max","Author":"Max Kuhn [aut, cre] (),\nDaniel Falbel [aut],\nPosit Software, PBC [cph, fnd]","Maintainer":"Max Kuhn ","Repository":"CRAN","Date/Publication":"2025-01-30 22:10:06 UTC","repoType":"cran","tarballUrl":"ftp://cran.r-project.org/pub/R/src/contrib/brulee_0.4.0.tar.gz","jsonAuthors":[{"name":"Max Kuhn","email":"max@posit.co","maintainer":true},{"name":"Daniel Falbel","email":"daniel@posit.co","maintainer":false},{"name":"Posit Software, PBC ","email":null,"maintainer":false}],"readme":"\n\n\n# brulee $\"a$ \n\n\n\n[![R-CMD-check](https://github.com/tidymodels/brulee/actions/workflows/R-CMD-check.yaml/badge.svg)](https://github.com/tidymodels/brulee/actions/workflows/R-CMD-check.yaml)\n[![Codecov test\ncoverage](https://codecov.io/gh/tidymodels/brulee/branch/main/graph/badge.svg)](https://app.codecov.io/gh/tidymodels/brulee?branch=main)\n[![Lifecycle:\nexperimental](https://img.shields.io/badge/lifecycle-experimental-orange.svg)](https://lifecycle.r-lib.org/articles/stages.html)\n\n\nThe R `brulee` package contains several basic modeling functions that\nuse the `torch` package infrastructure, such as:\n\n- [neural\n networks](https://brulee.tidymodels.org/reference/brulee_mlp.html)\n- [linear\n regression](https://brulee.tidymodels.org/reference/brulee_linear_reg.html)\n- [logistic\n regression](https://brulee.tidymodels.org/reference/brulee_logistic_reg.html)\n- [multinomial\n regression](https://brulee.tidymodels.org/reference/brulee_multinomial_reg.html)\n\n## Installation\n\nYou can install the released version of brulee from\n[CRAN](https://CRAN.R-project.org) with:\n\n``` r\ninstall.packages(\"brulee\")\n```\n\nAnd the development version from\n[GitHub](https://github.com/tidymodels/brulee) with:\n\n``` r\n# install.packages(\"pak\")\npak::pak(\"tidymodels/brulee\")\n```\n\n## Example\n\n`brulee` has formula, x/y, and recipe user interfaces for each function.\nFor example:\n\n``` r\nlibrary(brulee)\nlibrary(recipes)\nlibrary(yardstick)\n\ndata(bivariate, package = \"modeldata\")\nset.seed(20)\nnn_log_biv <- brulee_mlp(Class ~ log(A) + log(B), data = bivariate_train, \n epochs = 150, hidden_units = 3)\n\n# We use the tidymodels semantics to always return a tibble when predicting\npredict(nn_log_biv, bivariate_test, type = \"prob\") %>% \n bind_cols(bivariate_test) %>% \n roc_auc(Class, .pred_One)\n#> # A tibble: 1 × 3\n#> .metric .estimator .estimate\n#> \n#> 1 roc_auc binary 0.837\n```\n\nA recipe can also be used if the data require some sort of preprocessing\n(e.g., indicator variables, transformations, or standardization):\n\n``` r\nlibrary(recipes)\n\nrec <- \n recipe(Class ~ ., data = bivariate_train) %>% \n step_YeoJohnson(all_numeric_predictors()) %>% \n step_normalize(all_numeric_predictors())\n\nset.seed(20)\nnn_rec_biv <- brulee_mlp(rec, data = bivariate_train, \n epochs = 150, hidden_units = 3)\n\n# A little better\npredict(nn_rec_biv, bivariate_test, type = \"prob\") %>% \n bind_cols(bivariate_test) %>% \n roc_auc(Class, .pred_One)\n#> # A tibble: 1 × 3\n#> .metric .estimator .estimate\n#> \n#> 1 roc_auc binary 0.866\n```\n\n## Code of Conduct\n\nPlease note that the brulee project is released with a [Contributor Code\nof\nConduct](https://contributor-covenant.org/version/2/0/CODE_OF_CONDUCT.html).\nBy contributing to this project, you agree to abide by its terms.","jobInfo":{"package":"brulee","version":"0.4.0","parsingStatus":"success","parserVersion":2,"parsedAt":"2025-01-30T23:30:12+0000"}}

Description

Usage

Arguments

Value