% Generated by roxygen2: do not edit by hand
% Please edit documentation in R/tag_smartly.R
\name{check_model}
\alias{check_model}
\title{Check model performance}
\usage{
check_model(mod, indices, tags, modfam, xdat, cutoff)
}
\arguments{
\item{mod}{a glmnet model}

\item{indices}{a numeric vector of documents belonging to a subset used to fit the model}

\item{tags}{a character vector of document classifications or tags for documents at indices}

\item{modfam}{a string of length 1; options are "binomial" or "multinomial"}

\item{xdat}{a sparse document-feature matrix of class ngCMatrix}

\item{cutoff}{What probability should be used as a cutoff for topic classification? Options are proportions between 0-1, or NULL to find an optimal cutoff based on model accuracy.}
}
\value{
a vector containing the cutoff used and model accuracy at that cutoff
}
\description{
Given model inputs, checks model performance based on optimal or specified cutoff in classification probability
}