How to get the Rhadopp output directory in the java?

[PF L]

Hi ,  I have a question now.  I want to use java code to invoke Rhadoop job and return the output to the user,  but I found that every time the Rhadoop job just set the output directory with random number. So  how can I set the output directory before the Rhadoop job run, or get the random output directory  in the java code?  Much thanks.

figo

 

[Antonio Piccolboni]

help(mapreduce)

output argument

A

[PF L]

Hi Antonio Piccolboni，

                  Much thanks for your reply.  I still have no idea how to set this argument.  For example,  I want to run the rhadoop demo in https://github.com/RevolutionAnalytics/rmr2/blob/master/pkg/tests/kmeans.R  ,  I used to run this demo by command:  source(kmeans.R) . So where I can set the output argument? or my running methond is wrong? 

在 2013年11月23日星期六UTC+8上午12时23分56秒，Antonio Piccolboni写道：