Checksum error on K-means

2012-05-10 Thread Michael Kazekin
Hello! I have about 1.3M vectors from lucene.vector utility that I later try to clusterize in 550 clusters. Everything seems to be fine, clusterization starts, but in an hour I get: 12/05/10 18:26:50 INFO fs.FSInputChecker: Found checksum error: b[196,

Re: Checksum error on K-means

2012-05-10 Thread Paritosh Ranjan
I just googled out this exception and this looks likes a hdfs issue. Can you try formatting your hdfs and then rerun the K-Means clustering? On 10-05-2012 21:50, Michael Kazekin wrote: Hello! I have about 1.3M vectors from lucene.vector utility that I later try to clusterize in 550 clusters.

[jira] [Updated] (MAHOUT-1011) RecommenderJob is ignoring the command line threshold parameter

2012-05-10 Thread Bhaskar Devireddy (JIRA)
[ https://issues.apache.org/jira/browse/MAHOUT-1011?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Bhaskar Devireddy updated MAHOUT-1011: -- Attachment: thresold.patch RecommenderJob is ignoring the command line threshold

[jira] [Created] (MAHOUT-1011) RecommenderJob is ignoring the command line threshold parameter

2012-05-10 Thread Bhaskar Devireddy (JIRA)
Bhaskar Devireddy created MAHOUT-1011: - Summary: RecommenderJob is ignoring the command line threshold parameter Key: MAHOUT-1011 URL: https://issues.apache.org/jira/browse/MAHOUT-1011 Project:

[jira] [Resolved] (MAHOUT-1011) RecommenderJob is ignoring the command line threshold parameter

2012-05-10 Thread Sean Owen (JIRA)
[ https://issues.apache.org/jira/browse/MAHOUT-1011?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Sean Owen resolved MAHOUT-1011. --- Resolution: Fixed Yes, sounds good. RecommenderJob is ignoring the command line

[jira] [Resolved] (MAHOUT-1008) Remove link analysis package

2012-05-10 Thread Sebastian Schelter (JIRA)
[ https://issues.apache.org/jira/browse/MAHOUT-1008?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Sebastian Schelter resolved MAHOUT-1008. Resolution: Fixed Remove link analysis package

[jira] [Commented] (MAHOUT-1011) RecommenderJob is ignoring the command line threshold parameter

2012-05-10 Thread Hudson (JIRA)
[ https://issues.apache.org/jira/browse/MAHOUT-1011?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13272878#comment-13272878 ] Hudson commented on MAHOUT-1011: Integrated in Mahout-Quality #1471 (See

Jenkins build is still unstable: Mahout-Quality #1471

2012-05-10 Thread Apache Jenkins Server
See https://builds.apache.org/job/Mahout-Quality/changes

[jira] [Commented] (MAHOUT-1008) Remove link analysis package

2012-05-10 Thread Hudson (JIRA)
[ https://issues.apache.org/jira/browse/MAHOUT-1008?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13272990#comment-13272990 ] Hudson commented on MAHOUT-1008: Integrated in Mahout-Quality #1472 (See

Jenkins build is still unstable: Mahout-Quality #1472

2012-05-10 Thread Apache Jenkins Server
See https://builds.apache.org/job/Mahout-Quality/changes

Re: Making Mahout Leaner

2012-05-10 Thread Robin Anil
Any directions on what pattern I should follow for the redesign. -- Robin Anil On Wed, May 9, 2012 at 9:49 AM, Robin Anil robin.a...@gmail.com wrote: I believe most of this new NB discussion has been over chat. So here is the state of the NB universe from my view 1) Original NB and CNB