validation: validate regression bnet models
In gherardovarando/Rbnet: Define and Manage Bayesian Networks

validate regression bnet models

1
2
3

validation(x = naive_bayes.bnet, data = NULL, targets = NULL,
  predictors = NULL, folds = NULL, numfolds = 10, randomfolds = F,
  train = NULL, test = NULL, evalfun = MSE, mc.cores = 1)

`data`	data.frame a dataset of observation
`targets`	names of target variables
`predictors`	names of predictor variables
`folds`	folds for validation
`numfolds`	number of folds
`randomfolds`	logical
`train`	indx of train set
`test`	indx of test set
`evalfun`	function, must works as MSE or ARE
`mc.cores`	positive integer
`object`	optional value, a bnet object
`modelfun`	function to compute a bnet regression model, optional
`mle`	logica

mean and standard deviation of the numfolds evaluations, each one of the evaluations is computed using evalfun (usally MSE) over the prediction and the true value for the given fold (ot train/test set).

gherardovarando/Rbnet documentation built on May 17, 2019, 4:18 a.m.