{"Package":"cvwrapr","Type":"Package","Title":"Tools for Cross Validation","Version":"1.0","Authors@R":"person(\"Kenneth\", \"Tay\", role = c(\"aut\", \"cre\"), email = \"kjytay@gmail.com\")","Description":"Tools for performing cross-validation (CV). The main function is a\ngeneral purpose wrapper that performs k-fold CV for any tuning parameter in\nany supervised learning method. The package also has a function that\ncomputes the loss incurred by a set of predictions for a variety of loss\nfunctions and model families.","License":"GPL (>= 3)","Encoding":"UTF-8","RoxygenNote":"7.1.1","Suggests":"doParallel, gbm, glmnet, knitr, Matrix, parallel, pls,\nrmarkdown, testthat","Imports":"survival, foreach","VignetteBuilder":"knitr","NeedsCompilation":"no","Packaged":"2021-06-10 21:37:52 UTC; kjytay","Author":"Kenneth Tay [aut, cre]","Maintainer":"Kenneth Tay ","Repository":"CRAN","Date/Publication":"2021-06-11 10:10:02 UTC","repoType":"cran","tarballUrl":"ftp://cran.r-project.org/pub/R/src/contrib/cvwrapr_1.0.tar.gz","jsonAuthors":[{"name":"Kenneth Tay","email":"kjytay@gmail.com","maintainer":true}],"readme":"\n\n\n# cvwrapr\n\n\n\n\nThe goal of `cvwrapr` is to make cross-validation (CV) easy. The main\nfunction in the package is `kfoldcv`. It performs K-fold CV for a\nhyperparameter, returning the CV error for a path of hyperparameter\nvalues along with other useful information. The `computeError` function\nallows the user to compute the CV error for a range of loss functions\nfrom a matrix of out-of-fold predictions. See the package vignettes for\nmore examples.\n\n## Installation\n\nYou can install the development version from\n[GitHub](https://github.com/) with:\n\n``` r\n# install.packages(\"devtools\")\ndevtools::install_github(\"kjytay/cvwrapr\")\n```\n\n## Example\n\nThis is a basic example showing how to perform cross-validation for the\n`lambda` parameter in the lasso (Tibshirani 1996).\n\n``` r\n# simulate data\nset.seed(1)\nnobs <- 100; nvars <- 10\nx <- matrix(rnorm(nobs * nvars), nrow = nobs)\ny <- rowSums(x[, 1:2]) + rnorm(nobs)\n\nlibrary(cvwrapr)\nlibrary(glmnet)\n\nset.seed(1)\ncv_fit <- kfoldcv(x, y, train_fun = glmnet, predict_fun = predict)\n```\n\nThe returned output contains information on the CV procedure and can be\nplotted.\n\n``` r\nnames(cv_fit)\n#> [1] \"lambda\" \"cvm\" \"cvsd\" \"cvup\" \"cvlo\" \n#> [6] \"lambda.min\" \"lambda.1se\" \"index\" \"name\" \"overallfit\"\nplot(cv_fit)\n```\n\n

","jobInfo":{"package":"cvwrapr","version":"1.0","parsingStatus":"success","parserVersion":1,"parsedAt":"2021-06-11T12:29:48+0000"}}

Description

Usage

Arguments

Value