% Generated by roxygen2: do not edit by hand
% Please edit documentation in R/xgb.Booster.R
\name{xgb.attr}
\alias{xgb.attr}
\alias{xgb.attr<-}
\title{Accessors for serializable attributes of a model.}
\usage{
xgb.attr(object, which)

xgb.attr(object, which) <- value
}
\arguments{
\item{object}{Object of class \code{xgb.Booster} or  \code{xgb.Booster.handle}.}

\item{which}{a non-empty character string specifying which attribute is to be accessed.}

\item{value}{a value of an attribute. Non-character values are converted to character.
When length of a \code{value} vector is more than one, only the first element is used.}
}
\value{
\code{xgb.attr} returns either a string value of an attribute 
or \code{NULL} if an attribute wasn't stored in a model.
}
\description{
These methods allow to manipulate key-value attribute strings of an xgboost model.
}
\details{
Note that the xgboost model attributes are a separate concept from the attributes in R.
Specifically, they refer to key-value strings that can be attached to an xgboost model
and stored within the model's binary representation.
In contrast, any R-attribute assigned to an R-object of \code{xgb.Booster} class
would not be saved by \code{xgb.save}, since xgboost model is an external memory object
and its serialization is handled extrnally.

Also note that the attribute setter would usually work more efficiently for \code{xgb.Booster.handle}
than for \code{xgb.Booster}, since only just a handle would need to be copied.
}
\examples{
data(agaricus.train, package='xgboost')
train <- agaricus.train

bst <- xgboost(data = train$data, label = train$label, max.depth = 2,
               eta = 1, nthread = 2, nround = 2, objective = "binary:logistic")

xgb.attr(bst, "my_attribute") <- "my attribute value"
print(xgb.attr(bst, "my_attribute"))

xgb.save(bst, 'xgb.model')
bst1 <- xgb.load('xgb.model')
print(xgb.attr(bst1, "my_attribute"))

}