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
61
62
63
64
65
66
67
68
69
|
% Generated by roxygen2: do not edit by hand
% Please edit documentation in R/MulticlassWrapper.R
\name{makeMulticlassWrapper}
\alias{makeMulticlassWrapper}
\title{Fuse learner with multiclass method.}
\usage{
makeMulticlassWrapper(learner, mcw.method = "onevsrest")
}
\arguments{
\item{learner}{(\link{Learner} | \code{character(1)})\cr
The learner.
If you pass a string the learner will be created via \link{makeLearner}.}
\item{mcw.method}{(\code{character(1)} | \code{function}) \cr
\dQuote{onevsone} or \dQuote{onevsrest}.
You can also pass a function, with signature \verb{function(task)} and which
returns a ECOC codematrix with entries +1,-1,0.
Columns define new binary problems, rows correspond to classes (rows must be named).
0 means class is not included in binary problem.
Default is \dQuote{onevsrest}.}
}
\value{
\link{Learner}.
}
\description{
Fuses a base learner with a multi-class method.
Creates a learner object, which can be used like any other learner object.
This way learners which can only handle binary classification will be able to
handle multi-class problems, too.
We use a multiclass-to-binary reduction principle, where multiple binary
problems are created from the multiclass task. How these binary problems
are generated is defined by an error-correcting-output-code (ECOC) code book.
This also allows the simple and well-known one-vs-one and one-vs-rest
approaches. Decoding is currently done via Hamming decoding, see
e.g. here \url{https://jmlr.org/papers/volume11/escalera10a/escalera10a.pdf}.
Currently, the approach always operates on the discrete predicted labels
of the binary base models (instead of their probabilities) and the created
wrapper cannot predict posterior probabilities.
}
\seealso{
Other wrapper:
\code{\link{makeBaggingWrapper}()},
\code{\link{makeClassificationViaRegressionWrapper}()},
\code{\link{makeConstantClassWrapper}()},
\code{\link{makeCostSensClassifWrapper}()},
\code{\link{makeCostSensRegrWrapper}()},
\code{\link{makeDownsampleWrapper}()},
\code{\link{makeDummyFeaturesWrapper}()},
\code{\link{makeExtractFDAFeatsWrapper}()},
\code{\link{makeFeatSelWrapper}()},
\code{\link{makeFilterWrapper}()},
\code{\link{makeImputeWrapper}()},
\code{\link{makeMultilabelBinaryRelevanceWrapper}()},
\code{\link{makeMultilabelClassifierChainsWrapper}()},
\code{\link{makeMultilabelDBRWrapper}()},
\code{\link{makeMultilabelNestedStackingWrapper}()},
\code{\link{makeMultilabelStackingWrapper}()},
\code{\link{makeOverBaggingWrapper}()},
\code{\link{makePreprocWrapper}()},
\code{\link{makePreprocWrapperCaret}()},
\code{\link{makeRemoveConstantFeaturesWrapper}()},
\code{\link{makeSMOTEWrapper}()},
\code{\link{makeTuneWrapper}()},
\code{\link{makeUndersampleWrapper}()},
\code{\link{makeWeightedClassesWrapper}()}
}
\concept{wrapper}
|
