测试kmean时报错怎么解决

这篇文章主要介绍“测试kmean时报错怎么解决”，在日常操作中，相信很多人在测试kmean时报错怎么解决问题上存在疑惑，小编查阅了各式资料，整理出简单好用的操作方法，希望对大家解答”测试kmean时报错怎么解决”的疑惑有所帮助！接下来，请跟着小编一起来学习吧！

测试mahout in action 中kmean实例的时候,输入命令:

  bin/mahout kmeans -i reuters-vectors/tfidf-vectors/ \
-c reuters-initial-clusters \
-o reuters-kmeans-clusters \
-dm org.apache.mahout.common.distance.SquaredEuclideanDistanceMeasure \
-cd 1.0 -k 20 -x 20 -cl

报出先面的错误

mahout kmeans -i tfidf-vectors/ -c reuters-initial-clusters o reuters-kmeans-clusters dm org.apache.mahout.common.distance.SquaredEuclideanDistanceMeasure cd 1.0 -k 20 -x 20 -cl
Running on hadoop, using HADOOP_HOME=/usr/local/hadoop
HADOOP_CONF_DIR=/usr/local/hadoop/conf
14/01/23 12:43:34 ERROR common.AbstractJob: Unexpected o while processing Job-Specific Options:
usage: <command> [Generic Options] [Job-Specific Options]
Generic Options:
 -archives <paths>             comma separated archives to be unarchived
                               on the compute machines.
 -conf <configuration file>    specify an application configuration file
 -D <property=value>           use value for given property
 -files <paths>                comma separated files to be copied to the
                               map reduce cluster
 -fs <local|namenode:port>     specify a namenode
 -jt <local|jobtracker:port>   specify a job tracker
 -libjars <paths>              comma separated jar files to include in the
                               classpath.
Job-Specific Options:                                                           
  --input (-i) input                           Path to job input directory.     
  --output (-o) output                         The directory pathname for       
                                               output.                          
  --distanceMeasure (-dm) distanceMeasure      The classname of the             
                                               DistanceMeasure. Default is      
                                               SquaredEuclidean                 
  --clusters (-c) clusters                     The input centroids, as Vectors. 
                                               Must be a SequenceFile of        
                                               Writable, Cluster/Canopy.  If k  
                                               is also specified, then a random 
                                               set of vectors will be selected  
                                               and written out to this path     
                                               first                            
  --numClusters (-k) k                         The k in k-Means.  If specified, 
                                               then a random selection of k     
                                               Vectors will be chosen as the    
                                               Centroid and written to the      
                                               clusters input path.             
  --convergenceDelta (-cd) convergenceDelta    The convergence delta value.     
                                               Default is 0.5                   
  --maxIter (-x) maxIter                       The maximum number of            
                                               iterations.                      
  --overwrite (-ow)                            If present, overwrite the output 
                                               directory before running job     
  --clustering (-cl)                           If present, run clustering after 
                                               the iterations have taken place  
  --method (-xm) method                        The execution method to use:     
                                               sequential or mapreduce. Default 
                                               is mapreduce                     
  --help (-h)                                  Print out help                   
  --tempDir tempDir                            Intermediate output directory    
  --startPhase startPhase                      First phase to run               
  --endPhase endPhase                          Last phase to run                
14/01/23 12:43:35 INFO driver.MahoutDriver: Program took 149 ms

解决办法:输入命令参数o 前面没有加- ,mahout不能解析后面的参数,所以报出来的错误.

到此，关于“测试kmean时报错怎么解决”的学习就结束了，希望能够解决大家的疑惑。理论与实践的搭配能更好的帮助大家学习，快去试试吧！若想继续学习更多相关知识，请继续关注亿速云网站，小编会继续努力为大家带来更多实用的文章！