admin 管理员组文章数量: 1184232
2024年4月16日发(作者：android镜像文件下载)
Package‘AdaSampling’
October12,2022
TypePackage
TitleAdaptiveSamplingforPositiveUnlabeledandLabelNoise
Learning
Version1.3
AuthorPengyiYang
MaintainerPengyiYang<*****************>
DescriptionImplementstheadaptivesamplingprocedure,aframeworkforbothpositiveunla-
beledlearningandlearningwithclassla-
,P.,Ormerod,J.,Liu,W.,Ma,C.,Zomaya,A.,Yang,J.(2018)
LicenseGPL-3
EncodingUTF-8
DependsR(>=3.4.0)
LazyDatatrue
Importscaret(>=6.0-78),class(>=7.3-14),e1071(>=1.6-8),
stats,MASS
BugReports/PengyiYang/AdaSampling/issues
Suggestsknitr,rmarkdown
VignetteBuilderknitr
URL/PengyiYang/AdaSampling/
RoxygenNote6.1.1
NeedsCompilationno
RepositoryCRAN
Date/Publication2019-05-2108:20:04UTC
Rtopicsdocumented:
adaSvmBenchmark
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
1
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
2
4
5
6
7
2
Index
adaSample
8
adaSampleImplementationofAdaSamplingforpositiveunlabelledandlabel
noiselearning.
Description
adaSample()appliestheAdaSamplingproceduretoreducenoiseinthetrainingset,andsubse-
quentlytrainsaclassifihrow(observation)inthetestset,it
returnstheprobabilitiesofitbeingapositive("P)ornegative("N")instance,asatwocolumndata
frame.
Usage
adaSample(Ps,Ns,,,classifier="svm",s=1,
C=1,sampleFactor=1,weights=NULL)
Arguments
Ps
Ns
classifier
names(eachinstanceinthedatahastobenamed)ofpositiveexamples
names(eachinstanceinthedatahastobenamed)ofnegativeexamples
trainingdatamatrix,withoutclasslabels.
testdatamatrix,withoutclasslabels.
classifitoptionsaresupportvec-
tormachine,"svm",k-nearestneighbour,"knn",logisticregression"logit",
lineardiscriminantanalysis"lda",andfeatureweightedknn,"wKNN".
setstheseed.
setshowmanytimestoruntheclassifier,C>1inducesanensemblelearning
model.
providesacontrolonthesamplesizeforresampling.
featureweights,requiredwhenusingweightedknn.
s
C
sampleFactor
weights
Details
adaSample()isanadaptivesampling-basednoisereductionmethodtodealwithnoisyclassla-
belleddata,whichactsasawrapperfortraditionalclassifiers,suchassupportvectormachines,
k-nearestneighbours,logisticregression,andlineardiscriminantanalysis.
Thisprocessisusedtobuildupanoise-minimizedtrainingsetthatisderivedbyiterativelyresam-
plingthetrainingset,(train)basedonprobabilitiesderivedafteritsclassification.
Thissampledtrainingsetisthenusedtotrainaclassifier,whichisthenexecutedonthetestset.
adaSample()returnsaseriesofpredictionsforeachrowofthetestset.
Notethatthisfunctiondoesnotevaluatethequalityofthemodelandthusdoesnotcompareits
sspleaseseeadaSvmBenchmark().
版权声明：本文标题：AdaSampling包的文档说明书 内容由网友自发贡献，该文观点仅代表作者本人， 转载请联系作者并注明出处：http://www.roclinux.cn/p/1713237866a625373.html， 本站仅提供信息存储空间服务，不拥有所有权，不承担相关法律责任。如发现本站有涉嫌抄袭侵权/违法违规的内容，一经查实，本站将立刻删除。
发表评论