1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
|
\name{summary.optimx}
\alias{summary.optimx}
\encoding{UTF-8}
\title{Summarize optimx object}
\concept{minimization}
\concept{maximization}
\description{
Summarize an \code{"optimx"} object.
}
\usage{
\method{summary}{optimx}(object, order = NULL, par.select = TRUE, ...)
}
\arguments{
\item{object}{Object returned by \code{optimx}.}
\item{order}{A column name, character vector of columns names, R
expression in terms of column names or a list of R expressions in terms of
column names. \code{NULL}, the default, means no re-ordering.
\code{rownames} can be used to
alphabetic ordering by method name.
\code{NULL}, the default, causes
it not to be reordered. Note that if
\code{follow.on} is TRUE re-ordering likely makes no sense.
The result is ordered by the \code{order}
specification, each specified column in ascending order (except for
\code{value} which is in descending order if the optimization problem is a
maximization problem).}
\item{par.select}{a numeric, character or logical vector selecting
those \code{par} values to display. For example, \code{par=1:5} means
display only the first 5 parameters. Recycled so \code{par.select=FALSE}
selects no parameters.}
\item{\dots}{Further arguments to be passed to the function. Currently not
used.}
}
\details{
If \code{order} is specified then the result is reordered by the specified
columns, each in ascending order (except possibly for the
\code{value} column which is re-ordered in descending order for
maximization problems).}
\value{
\code{summary.optimx} returns \code{object} with the rows ordered according
to \code{order} and with those parameters selected by \code{par.select}.
}
\examples{
ans <- optimx(fn = function(x) sum(x*x), par = 1:2)
# order by method name.
summary(ans, order = rownames)
# order by objective value. Do not show parameter values.
summary(ans, order = value, par.select = FALSE)
# order by objective value and then number of function evaluations
# such that objectives that are the same to 3 decimals are
# considered the same. Show only first parameter.
summary(ans, order = list(round(value, 3), fevals), par.select = 1)
}
\keyword{nonlinear}
\keyword{optimize}
|
