Re: TopItems.getTopUsers()

The PriorityQueue there is a min heap. It's used to keep finding the smallest among a collection of large values. So, no it can't be used directly to create a list of items from large to small as it has a "get smallest" method, not "get largest". The result of the method is a list of IDs, not a li

[jira] [Commented] (MAHOUT-830) Distribution should create .deb and .rpm packages

[ https://issues.apache.org/jira/browse/MAHOUT-830?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13146796#comment-13146796 ] Ted Dunning commented on MAHOUT-830: Half done and got diverted. A first checking is

Re: TopItems.getTopUsers()

On Tue, Nov 8, 2011 at 7:46 PM, Grant Ingersoll wrote: > Also, FWIW, Lucene's PQ implementation is faster than Java's, from what I > understand (which is why we wrote our own). That's why we *kept* our own (it predates the standard JDK implementation). The existence of adjustTop() (and our use o

TopItems.getTopUsers()

I've been reading code and am wondering about TopItems.getTopUsers Here's my pseudocode of it (lines 96-134 in TopItems) Get a Priority Queue (PQ) for all users estimate the similarity of user i with our current user (the one we are generating a rec for) put a SimilarUser object o

[jira] [Commented] (MAHOUT-830) Distribution should create .deb and .rpm packages

[ https://issues.apache.org/jira/browse/MAHOUT-830?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13146672#comment-13146672 ] Lance Norskog commented on MAHOUT-830: -- Never checked in. It's a fine idea.

Re: Recruiting new contributors and committers [was: Demoralized over JIRA state]

On 25 October 2011 03:39, Isabel Drost wrote: > On 24.10.2011 Grant Ingersoll wrote: >> Docs is one obvious one.  Also, just keep supplying patches. > > Speaking of supplying patches: Doing so seems non-trivial for quite some > developers. > > During the Apache Hadoop Hackathon Berlin earlier this

[jira] [Commented] (MAHOUT-627) Baum-Welch Algorithm on Map-Reduce for Parallel Hidden Markov Model Training.

[ https://issues.apache.org/jira/browse/MAHOUT-627?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13146479#comment-13146479 ] Suneel Marthi commented on MAHOUT-627: -- While reviewing the code in BaumWelchTrainer.

[jira] [Updated] (MAHOUT-843) Top Down Clustering

[ https://issues.apache.org/jira/browse/MAHOUT-843?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Paritosh Ranjan updated MAHOUT-843: --- Attachment: MAHOUT-843-patch-v1 Hi Jeff, I have added the patch which has the CLI and Junit t

Jenkins build is back to normal : Mahout-Quality #1161

See

[jira] [Commented] (MAHOUT-877) Enable the parallel ALS recommender to use implicit feedback data

[ https://issues.apache.org/jira/browse/MAHOUT-877?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13146309#comment-13146309 ] Hudson commented on MAHOUT-877: --- Integrated in Mahout-Quality #1161 (See [https://builds.ap

[jira] [Commented] (MAHOUT-865) Refactor Sequential Clustering algorithms

[ https://issues.apache.org/jira/browse/MAHOUT-865?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13146251#comment-13146251 ] Grant Ingersoll commented on MAHOUT-865: We likely need something similar to Abstr

[jira] [Commented] (MAHOUT-612) Simplify configuring and running Mahout MapReduce jobs from Java using Java bean configuration

[ https://issues.apache.org/jira/browse/MAHOUT-612?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13146246#comment-13146246 ] Grant Ingersoll commented on MAHOUT-612: It seems like we shouldn't have to wait f

[jira] [Commented] (MAHOUT-344) Minhash based clustering

[ https://issues.apache.org/jira/browse/MAHOUT-344?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13146215#comment-13146215 ] Ankur commented on MAHOUT-344: -- Grant, The idea behind keyGroups is to concatenate hashes fro

[jira] [Created] (MAHOUT-878) Provide better examples for the parallel ALS recommender code

Provide better examples for the parallel ALS recommender code - Key: MAHOUT-878 URL: https://issues.apache.org/jira/browse/MAHOUT-878 Project: Mahout Issue Type: Task Comp

[jira] [Resolved] (MAHOUT-877) Enable the parallel ALS recommender to use implicit feedback data

[ https://issues.apache.org/jira/browse/MAHOUT-877?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Sebastian Schelter resolved MAHOUT-877. --- Resolution: Fixed Fix Version/s: 0.6 > Enable the parallel ALS recommender

[jira] [Updated] (MAHOUT-877) Enable the parallel ALS recommender to use implicit feedback data

[ https://issues.apache.org/jira/browse/MAHOUT-877?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Sebastian Schelter updated MAHOUT-877: -- Attachment: MAHOUT-877.patch > Enable the parallel ALS recommender to use implicit

[jira] [Created] (MAHOUT-877) Enable the parallel ALS recommender to use implicit feedback data

Enable the parallel ALS recommender to use implicit feedback data - Key: MAHOUT-877 URL: https://issues.apache.org/jira/browse/MAHOUT-877 Project: Mahout Issue Type: New Feature