【发布时间】：2017-12-12 08:22:57
【问题描述】：
我有这个代码
import weka.core.Instances;
import weka.classifiers.bayes.NaiveBayes;
import weka.classifiers.trees.J48;
import weka.classifiers.Evaluation;
import weka.filters.Filter;
import weka.filters.supervised.instance.Resample;
import java.util.Random;
import java.io.BufferedReader;
import java.io.FileReader ;
import weka.filters.unsupervised.attribute.NumericToNominal;
public class LoadModelWeka
{
public static void main(String[] args) throws Exception {
// training
BufferedReader reader = null;
reader=new BufferedReader(new FileReader("D:\\aaaaaaaaaaaaaaaaaaaaaa\\Licenta\\partXML\\TestTask1_partXML.arff"));
Instances train =new Instances (reader);
Instances classBalancerInstances=homogenizeTrainingData(train);
// classBalancerInstances.setClassIndex(classBalancerInstances.numAttributes() - 1);
classBalancerInstances.setClassIndex(0);
reader.close();
NaiveBayes nb = new NaiveBayes();
nb.buildClassifier(classBalancerInstances);
Evaluation eval = new Evaluation(classBalancerInstances);
eval.crossValidateModel(nb, classBalancerInstances, 10 , new Random(1));
System.out.println(eval.toSummaryString("\n Results \n=====\n",true));
System.out.println(eval.fMeasure(1)+" "+eval.precision(1)+" "+eval.recall(1)+" ");
}
public static Instances homogenizeTrainingData(Instances data) throws Exception
{
final Resample filter = new Resample();
Instances filteredIns = null;
filter.setBiasToUniformClass(1.0);
Instances newData=convertToNominal(data);
try {
data.setClassIndex(data.numAttributes() - 1);
filter.setInputFormat(newData);
filter.setNoReplacement(false);
filter.setSampleSizePercent(100);
filteredIns = Filter.useFilter(newData, filter);
} catch (Exception e) {
System.out.println("Error when resampling input data!");
e.printStackTrace();
}
return filteredIns;
}
public static Instances convertToNominal(Instances originalTrain) throws Exception{
NumericToNominal convert= new NumericToNominal();
String[] options= new String[2];
options[0]="-R";
options[1]="1-9"; //range of variables to make numeric
convert.setOptions(options);
convert.setInputFormat(originalTrain);
Instances newData=Filter.useFilter(originalTrain, convert);
System.out.println("Before");
for(int i=0; i<2; i=i+1)
{
System.out.println("Nominal? "+originalTrain.attribute(i).isNominal());
}
System.out.println("After");
for(int i=0; i<2; i=i+1)
{
System.out.println("Nominal? "+newData.attribute(i).isNominal());
}
return newData;
}
}
这里我想用类平衡器过滤我的数据集，但是出现了一个错误，说属性必须是标称的。我试图转换为标称的，但是发生了这个错误：
weka.core.UnassignedClassException: weka.filters.supervised.instance.Resample: Class attribute not set!
at weka.core.Capabilities.test(Unknown Source)
at weka.core.Capabilities.test(Unknown Source)
at weka.core.Capabilities.testWithFail(Unknown Source)
at weka.filters.Filter.testInputFormat(Unknown Source)
at weka.filters.Filter.setInputFormat(Unknown Source)
at weka.filters.supervised.instance.Resample.setInputFormat(Unknown Source)
at licenta1.LoadModelWeka.homogenizeTrainingData(LoadModelWeka.java:44)
at licenta1.LoadModelWeka.main(LoadModelWeka.java:22)
Exception in thread "main" java.lang.NullPointerException
at licenta1.LoadModelWeka.main(LoadModelWeka.java:24)
我试图解决它以设置类索引，但我不太了解它是如何工作的，所以我不知道如何设置实例以使其正常工作。 我的文件是一个具有 9 个属性的 arff，并且分类是用属性号 9 进行的，所以我认为，我必须将所有列转换为 1 到 9，但我不确定我是否理解过滤器以及从数字到名义的转换有效。
可以帮帮我吗？
我的 .arff 文件如下所示：
@attribute Word {Induction,of,during,monocyte,differentiation,by,HIV,type,1.0,infection,Electrophoretic,mobility,experiments,were,usedare,the,mostmayat,me,to,detect,located,in,long,terminal,repeat,PMA,treatment,acts,probablyin,inducing,expression,nuclei,In,nuclear,extracts,from,monocytes,or,macrophages,induction,occurred,only,if,cells,previously,infected,with,When,U937,no,factor,was,detected,whereas,high,level,progeny,virions,produced,suggesting,that,this,not,required,These,results,indicate,monocytic,cell,lineage,could,mimic,some,stimuli,allowing,Positive,and,negative,regulation,immunoglobulinis,most,suggesthere,gene,a,novel,enhancer,element,A,new,specific,Tandem,copies,fragment,transcription,B,but,Jurkat,T,HeLa,Footprinting,analysis,revealed,Hex,Gel,addition,ubiquitous,band,found,also,agreement,gel,may,Furthermorein,vivo,activity,striking,contrast,mouse,sequence,very,its,own,Interestingly,suppress,promoter,Moreover,simian,virus,40.0,blocked,Thus,identified,study,is,probably,target,site,for,both,positive,factors,The,production,human,immunodeficiency,purified}
@attribute w-2 {'null ','Induction ','of ','during ','monocyte ','differentiation ','by ','HIV ','type ',1.0,'infection ','Electrophoretic ','mobility ','experiments ','were ','usedare ','the ','mostmayat ','me ','to ','detect ','located ','in ','long ','terminal ','repeat ','PMA ','treatment ','acts ','probablyin ','inducing ','expression ','nuclei ','In ','nuclear ','extracts ','from ','monocytes ','or ','macrophages ','induction ','occurred ','only ','if ','cells ','previously ','infected ','with ','When ','U937 ','no ','factor ','was ','detected ','whereas ','high ','level ','progeny ','virions ','was ','produced ','suggesting ','that ','this ','not ','required ','These ','results ','indicate ','monocytic ','cell ','lineage ','could ','mimic ','some ','stimuli ','allowing ','Positive ','and ','negative ','regulation ','immunoglobulinis ','most ','suggesthere ','gene ','a ','novel ','enhancer ','element ','A ','new ','specific ','Tandem ','copies ','fragment ','transcription ','B ','but ','Jurkat ','T ','HeLa ','Footprinting ','analysis ','revealed ','from ','Hex ','Gel ','addition ','ubiquitous ','band ','found ','also ','T ','agreement ','gel ','in ','may ','Furthermorein ','vivo ','activity ','striking ','contrast ','mouse ','sequence ','very ','its ','own ','Interestingly ','the ','suppress ','promoter ','Moreover ','simian ','virus ',40.0,'blocked ','Thus ','identified ','study ','is ','probably ','target ','site ','for ','both ','positive ','factors ','The ','production ','human ','immunodeficiency ','purified ','human ','monocytes '}
@attribute w-1 {' null ',' Induction ',' of ',' during ',' monocyte ',' differentiation ',' by ',' HIV ',' type ',1.0,' infection ',' Electrophoretic ',' mobility ',' experiments ',' were ',' usedare ',' the ',' mostmayat ',' me ',' to ',' detect ',' located ',' in ',' long ',' terminal ',' repeat ',' PMA ',' treatment ',' acts ',' probablyin ',' inducing ',' expression ',' nuclei ',' In ',' nuclear ',' extracts ',' from ',' monocytes ',' or ',' macrophages ',' induction ',' occurred ',' only ',' if ',' cells ',' previously ',' infected ',' with ',' When ',' U937 ',' no ',' factor ',' was ',' detected ',' whereas ',' high ',' level ',' progeny ',' virions ',' produced ',' suggesting ',' that ',' this ',' not ',' required ',' These ',' results ',' indicate ',' monocytic ',' cell ',' lineage ',' could ',' mimic ',' some ',' stimuli ',' allowing ',' Positive ',' and ',' negative ',' regulation ',' immunoglobulinis ',' most ',' suggesthere ',' gene ',' a ',' novel ',' enhancer ',' element ',' A ',' new ',' specific ',' Tandem ',' copies ',' fragment ',' transcription ',' B ',' but ',' Jurkat ',' T ',' cells ',' HeLa ',' Footprinting ',' analysis ',' revealed ',' Hex ',' Gel ',' addition ',' ubiquitous ',' band ',' found ',' also ',' agreement ',' gel ',' Jurkat ',' may ',' Furthermorein ',' vivo ',' activity ',' striking ',' contrast ',' mouse ',' sequence ',' very ',' its ',' own ',' Interestingly ',' fragment ',' suppress ',' promoter ',' Moreover ',' simian ',' virus ',40.0,' blocked ',' Thus ',' identified ',' study ',' is ',' probably ',' target ',' site ',' for ',' both ',' positive ',' factors ',' The ',' production ',' human ',' immunodeficiency ',' purified ',' and '}
@attribute w {' Induction ',' of ',' during ',' monocyte ',' differentiation ',' by ',' HIV ',' type ',1.0,' infection ',' Electrophoretic ',' mobility ',' experiments ',' were ',' usedare ',' the ',' mostmayat ',' me ',' to ',' detect ',' located ',' in ',' long ',' terminal ',' repeat ',' PMA ',' treatment ',' acts ',' probablyin ',' inducing ',' expression ',' nuclei ',' In ',' nuclear ',' extracts ',' from ',' monocytes ',' or ',' macrophages ',' induction ',' occurred ',' only ',' if ',' cells ',' previously ',' infected ',' with ',' When ',' U937 ',' no ',' factor ',' was ',' detected ',' whereas ',' high ',' level ',' progeny ',' virions ',' produced ',' suggesting ',' that ',' this ',' not ',' required ',' These ',' results ',' indicate ',' monocytic ',' cell ',' lineage ',' could ',' mimic ',' some ',' stimuli ',' allowing ',' Positive ',' and ',' negative ',' regulation ',' immunoglobulinis ',' most ',' suggesthere ',' gene ',' a ',' novel ',' enhancer ',' element ',' A ',' new ',' specific ',' Tandem ',' copies ',' fragment ',' transcription ',' B ',' but ',' Jurkat ',' T ',' HeLa ',' Footprinting ',' analysis ',' revealed ',' Hex ',' Gel ',' addition ',' ubiquitous ',' band ',' found ',' also ',' agreement ',' gel ',' may ',' Furthermorein ',' vivo ',' activity ',' striking ',' contrast ',' mouse ',' sequence ',' very ',' its ',' own ',' Interestingly ',' suppress ',' promoter ',' Moreover ',' simian ',' virus ',40.0,' blocked ',' Thus ',' identified ',' study ',' is ',' probably ',' target ',' site ',' for ',' both ',' positive ',' factors ',' The ',' production ',' human ',' immunodeficiency ',' purified ',' macrophages '}
@attribute w+1 {' of ',' during ',' monocyte ',' differentiation ',' by ',' HIV ',' type ',1.0,' infection ',' Electrophoretic ',' mobility ',' experiments ',' were ',' usedare ',' the ',' mostmayat ',' me ',' to ',' detect ',' located ',' in ',' long ',' terminal ',' repeat ',' PMA ',' treatment ',' acts ',' probablyin ',' inducing ',' expression ',' nuclei ',' In ',' nuclear ',' extracts ',' from ',' monocytes ',' or ',' macrophages ',' induction ',' of ',' occurred ',' only ',' if ',' cells ',' previously ',' infected ',' with ',' When ',' U937 ',' no ',' factor ',' was ',' detected ',' whereas ',' high ',' level ',' progeny ',' virions ',' produced ',' suggesting ',' that ',' this ',' not ',' required ',' These ',' results ',' indicate ',' that ',' monocytic ',' cell ',' lineage ',' could ',' mimic ',' some ',' stimuli ',' allowing ',' Positive ',' and ',' negative ',' regulation ',' immunoglobulinis ',' most ',' suggesthere ',' gene ',' a ',' novel ',' enhancer ',' element ',' A ',' new ',' specific ',' Tandem ',' copies ',' fragment ',' transcription ',' B ',' but ',' Jurkat ',' T ',' HeLa ',' Footprinting ',' analysis ',' revealed ',' Hex ',' Gel ',' addition ',' ubiquitous ',' band ',' found ',' also ',' HeLa ',' agreement ',' gel ',' may ',' Furthermorein ',' vivo ',' activity ',' striking ',' contrast ',' mouse ',' sequence ',' very ',' its ',' own ',' Interestingly ',' suppress ',' promoter ',' Moreover ',' simian ',' virus ',40.0,' blocked ',' Thus ',' identified ',' study ',' is ',' probably ',' a ',' target ',' site ',' for ',' both ',' positive ',' factors ',' The ',' production ',' human ',' immunodeficiency ',' purified ',' macrophages ',' null '}
@attribute w+2 {' during ',' monocyte ',' differentiation ',' by ',' HIV ',' type ',1.0,' infection ',' Electrophoretic ',' mobility ',' experiments ',' were ',' usedare ',' the ',' mostmayat ',' me ',' to ',' detect ',' located ',' in ',' long ',' terminal ',' repeat ',' PMA ',' treatment ',' acts ',' probablyin ',' inducing ',' expression ',' nuclei ',' In ',' nuclear ',' extracts ',' from ',' monocytes ',' or ',' macrophages ',' induction ',' of ',' occurred ',' only ',' if ',' cells ',' previously ',' infected ',' with ',' When ',' U937 ',' no ',' induction ',' factor ',' was ',' detected ',' whereas ',' high ',' level ',' progeny ',' virions ',' produced ',' suggesting ',' that ',' this ',' not ',' required ',' These ',' results ',' indicate ',' monocytic ',' cell ',' lineage ',' could ',' mimic ',' some ',' stimuli ',' allowing ',' Positive ',' and ',' negative ',' regulation ',' immunoglobulinis ',' most ',' suggesthere ',' gene ',' a ',' novel ',' enhancer ',' element ',' A ',' new ',' specific ',' Tandem ',' copies ',' this ',' fragment ',' transcription ',' B ',' but ',' Jurkat ',' T ',' HeLa ',' cells ',' Footprinting ',' analysis ',' revealed ',' Hex ',' Gel ',' addition ',' ubiquitous ',' band ',' found ',' also ',' agreement ',' gel ',' may ',' Furthermorein ',' vivo ',' activity ',' striking ',' contrast ',' mouse ',' sequence ',' very ',' its ',' own ',' Interestingly ',' suppress ',' promoter ',' Moreover ',' simian ',' virus ',40.0,' blocked ',' Thus ',' identified ',' study ',' is ',' probably ',' target ',' site ',' for ',' both ',' positive ',' factors ',' The ',' production ',' human ',' immunodeficiency ',' purified ',' null ',' null '}
@attribute Lemma {induction,of,during,monocyte,differentiation,by,hiv,type,1.0,infection,electrophoretic,mobility,experiment,be,usedare,the,mostmayat,I,to,detect,located,in,long,terminal,repeat,pma,treatment,act,probablyin,induce,expression,nucleus,nuclear,extract,from,or,macrophage,occur,only,if,cell,previously,infected,with,when,u937,no,factor,whereas,high,level,progeny,virion,produce,suggest,that,this,not,require,these,result,indicate,monocytic,lineage,could,mimic,some,stimulus,allow,positive,and,negative,regulation,immunoglobulinis,most,suggesthere,gene,a,novel,enhancer,element,new,specific,Tandem,copy,fragment,transcription,b,but,jurkat,t,hela,footprinting,analysis,reveal,hex,gel,addition,ubiquitous,band,find,also,agreement,may,furthermorein,vivo,activity,striking,contrast,mouse,sequence,very,its,own,interestingly,suppress,promoter,moreover,simian,virus,40.0,block,thus,identify,study,probably,target,site,for,both,production,human,immunodeficiency,purify}
@attribute POS {NNP,IN,NN,CD,RB,NNS,VBD,DT,PRP,TO,VB,VBN,UH,WRB,JJ,VBG,MD,JJS,SYM,VBZ}
@attribute IsCue {' F',' T'}
@data
Induction,'null ',' null ',' Induction ',' of ',' during ',induction,NNP,' F'
of,'null ',' Induction ',' of ',' during ',' monocyte ',of,IN,' F'
during,'Induction ',' of ',' during ',' monocyte ',' differentiation ',during,IN,' F'
monocyte,'of ',' during ',' monocyte ',' differentiation ',' by ',monocyte,NN,' F'
differentiation,'during ',' monocyte ',' differentiation ',' by ',' HIV ',differentiation,NN,' F'
by,'monocyte ',' differentiation ',' by ',' HIV ',' type ',by,IN,' F'
HIV,'differentiation ',' by ',' HIV ',' type ',1.0,hiv,NNP,' F'
type,'by ',' HIV ',' type ',1.0,' infection ',type,NN,' F'
1.0,'HIV ',' type ',1.0,' infection ',' Electrophoretic ',1.0,CD,' F'
infection,'type ',1.0,' infection ',' Electrophoretic ',' mobility ',infection,NN,' F'
分类必须在 IsCue 属性的函数中。如果当前单词（我是指行中的第一个单词）是肯定的，则 IsCue 为真，否则为假。
【问题讨论】：
标签： java classification weka