The mapper and reducer can have unit tests if they are sufficiently interesting.
More commonly, the MiniDFSCluster is used to provide an emulated hadoop environment. On 2/6/08 12:25 PM, "Jeff Eastman" <[EMAIL PROTECTED]> wrote: > This list is just full of interesting information :). > > On topic, I'd like to reserve some namespace, specifically > org.apache.mahout.clustering.canopy which I am using for an initial prototype > of, duh, canopy clustering. I'm going to start with a little unit test to get > the basic algorithm sorted out, then M/R it. Should I open a Jira for this? > > Looking at the test subtree, the same package would also appear there. Is that > ok? > > Finally, thinking ahead about tests, how does one typically write unit tests > for this stuff? For the Mapper and Reducer? > > Jeff > > -----Original Message----- > From: Ted Dunning [mailto:[EMAIL PROTECTED] > Sent: Wednesday, February 06, 2008 10:41 AM > To: [email protected] > Subject: Re: Package Structure > > > That's OK. It was a diaresis anyway, not an umlaut. > > > On 2/6/08 10:33 AM, "Jeff Eastman" <[EMAIL PROTECTED]> wrote: > >> Yes, it was the spelling corrector, no umlaut intended. >> Jeff >> >> -----Original Message----- >> From: Ted Dunning [mailto:[EMAIL PROTECTED] >> Sent: Wednesday, February 06, 2008 9:13 AM >> To: [email protected] >> Subject: Re: Package Structure >> >> >> This is Microsoft's spelling corrector in action. Presumably his >> programming editor won't do this. >> >> >> On 2/6/08 12:44 AM, "Thilo Goetz" <[EMAIL PROTECTED]> wrote: >> >>> Jeff Eastman wrote: >>>> naïve_bayes. >>> >>> You probably didn't mean this literally, but just in case: >>> I'd vote for ascii-only characters in all things Java (and >>> source code in general); anything other than ascii is just >>> asking for trouble. I write the language of umlauts, so I >>> know what I'm talking about ;-) >>> >>> --Thilo >>> >> >
