I uncommented line 39 and am getting the same errors (index error with kmeans, 
0 LL with LDA) as before. I am running on real clusters (CDH3 & MapR). Trying 
to run locally, I get this curious output. I don't have much time today to 
pursue it (in meetings all day) but will do my best:

[dev@devbox mahout]$ ./examples/bin/build-reuters.sh 
Please select a number to choose the corresponding clustering algorithm
1. kmeans clustering
2. lda clustering
Enter your choice : 1
ok. You chose 1 and we'll use kmeans Clustering
Downloading Reuters-21578
  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                                 Dload  Upload   Total   Spent    Left  Speed
100 7959k  100 7959k    0     0  1145k      0  0:00:06  0:00:06 --:--:-- 1135k
Extracting...
no HADOOP_HOME set, running locally
May 20, 2011 12:35:51 PM org.slf4j.impl.JCLLoggerAdapter warn
WARNING: No org.apache.lucene.benchmark.utils.ExtractReuters.props found on 
classpath, will use command-line arguments only
Deleting all files in ./examples/bin/work/reuters-out
May 20, 2011 12:35:56 PM org.slf4j.impl.JCLLoggerAdapter info
INFO: Program took 4690 ms
no HADOOP_HOME set, running locally
May 20, 2011 12:35:57 PM org.slf4j.impl.JCLLoggerAdapter info
INFO: Command line arguments: {--charset=UTF-8, --chunkSize=5, 
--endPhase=2147483647, 
--fileFilterClass=org.apache.mahout.text.PrefixAdditionFilter, 
--input=./examples/bin/work/reuters-out/, --keyPrefix=, 
--output=./examples/bin/work/reuters-out-seqdir, --startPhase=0, --tempDir=temp}
Exception in thread "main" java.lang.IllegalStateException: 
java.io.FileNotFoundException: 
/home/dev/workspace/mahout/examples/bin/work/reuters-out/reut2-018.sgm-835.txt 
(Too many open files)
        at 
org.apache.mahout.text.SequenceFilesFromDirectoryFilter.accept(SequenceFilesFromDirectoryFilter.java:79)
        at org.apache.hadoop.fs.FileSystem.listStatus(FileSystem.java:724)
        at org.apache.hadoop.fs.FileSystem.listStatus(FileSystem.java:746)
        at 
org.apache.mahout.text.SequenceFilesFromDirectory.run(SequenceFilesFromDirectory.java:76)
        at 
org.apache.mahout.text.SequenceFilesFromDirectory.run(SequenceFilesFromDirectory.java:106)
        at org.apache.hadoop.util.ToolRunner.run(ToolRunner.java:65)
        at org.apache.hadoop.util.ToolRunner.run(ToolRunner.java:79)
        at 
org.apache.mahout.text.SequenceFilesFromDirectory.main(SequenceFilesFromDirectory.java:81)
        at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
        at 
sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:39)
        at 
sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:25)
        at java.lang.reflect.Method.invoke(Method.java:597)
        at 
org.apache.hadoop.util.ProgramDriver$ProgramDescription.invoke(ProgramDriver.java:68)
        at org.apache.hadoop.util.ProgramDriver.driver(ProgramDriver.java:139)
        at org.apache.mahout.driver.MahoutDriver.main(MahoutDriver.java:187)
C

-----Original Message-----
From: Grant Ingersoll [mailto:[email protected]] 
Sent: Friday, May 20, 2011 11:17 AM
To: [email protected]
Subject: Re: Is LDA Broken?

yeah, sorry.  I commented out line 39: cd examples/bin

On May 20, 2011, at 1:58 PM, Jeff Eastman wrote:

> It does seem these two symptoms are of the same problem. I applied the patch; 
> however, and now neither option runs. It appears the cd is off but I can't 
> see where.
> 
> [dev@devbox mahout-distribution-0.5]$ time ./examples/bin/build-reuters.sh 
> Please select a number to choose the corresponding clustering algorithm
> 1. kmeans clustering
> 2. lda clustering
> Enter your choice : 1
> ok. You chose 1 and we'll use kmeans Clustering
> ./examples/bin/build-reuters.sh: line 54: ./bin/mahout: No such file or 
> directory
> ./examples/bin/build-reuters.sh: line 64: ./bin/mahout: No such file or 
> directory
> 
> 
> -----Original Message-----
> From: Grant Ingersoll [mailto:[email protected]] 
> Sent: Friday, May 20, 2011 10:50 AM
> To: [email protected]
> Subject: Re: Is LDA Broken?
> 
> Likely so, see MAHOUT-694.  
> 
> 
> On May 20, 2011, at 1:39 PM, Sean Owen wrote:
> 
>> Oh sorry these are the same issue? Great!
>> On May 20, 2011 5:44 PM, "Jake Mannix" <[email protected]> wrote:
>>> Looks like Grant got a fix posted? Has anyone else tried it?
>>> 
>>> -jake
>>> 
>>> On Fri, May 20, 2011 at 9:32 AM, Sean Owen <[email protected]> wrote:
>>> 
>>>> I think we definitely need to figure out whether it's a bug or some other
>>>> confusion. If it's a doesn't-work-at-all bug yes probably the kind of
>> thing
>>>> that needs a fix ASAP in which case write up all you know and everyone
>> will
>>>> pile in to look at it.
>>>> 
>>>> On Fri, May 20, 2011 at 5:29 PM, Jeff Eastman <[email protected]> wrote:
>>>> 
>>>>> Is this an issue that should be fixed before we release? It seems to be
>>>>> broken to me.
>>>>> 
>>>>> -----Original Message-----
>>>>> From: Jeff Eastman [mailto:[email protected]]
>>>>> Sent: Thursday, May 19, 2011 5:05 PM
>>>>> To: [email protected]
>>>>> Subject: Is LDA Broken?
>>>>> 
>>>>> I'm running build-reuters option 2 and the LDA runs to maxIterations
>> (20)
>>>>> without ever producing a non-zero Log Likelihood. This is not the
>>>> behavior
>>>>> that I recall from earlier runs and seems quite unlikely to be correct.
>>>>> 
>>>>> 
>>>> 
> 
> 

--------------------------------------------
Grant Ingersoll
Join the LUCENE REVOLUTION
Lucene & Solr User Conference
May 25-26, San Francisco
www.lucenerevolution.org

Reply via email to