Home | About | Sematext search-lucene.com search-hadoop.com
clear query|facets|time Search criteria: position.   Results from 91 to 100 from 424 (0.234s).
Loading phrases to help you
refine your search...
Re: Question regarding correlation value produced by UncenteredCosineSimilarity - Mahout - [mail # user]
...I think the key point here is that these vectors should be logically thought of as sparse vectors (not sure how they are represented in Mahout).  If a value in the vector at some position i...
... is empty, it is essentially not part of the calculation.  And only positions, i, that have a value in both vectors can be used as part of the calculation for the denominator.  On Wed, Aug 22, 2012...
   Author: John Conwell, 2012-08-22, 21:48
Matrix Factorization with Implicit Feedback - Mahout - [mail # user]
... usage points are positive (i.e, the user watched the movie).  I changed the MovieLens Example:  $MAHOUT parallelALS --input /tmp/mahout-work-cloudera/input.txt --output ${WORK...
   Author: Royi Ronen, 2012-12-10, 15:59
FastIDSet.java - Mahout - [source code]
...ForAdd(long)    */   private int find(long key) {     int theHashCode = (int) key & 0x7FFFFFFF; // make sure it's positive     long[] keys = this.keys;     int hashSize = keys.length;     int jump = 1 + the...
...HashCode = (int) key & 0x7FFFFFFF; // make sure it's positive     long[] keys = this.keys;     int hashSize = keys.length;     int jump = 1 + theHashCode % (hashSize - 2);     int index = the...
[+ show more]
  
FastMap.java - Mahout - [source code]
...(Object key) {     int theHashCode = key.hashCode() & 0x7FFFFFFF; // make sure it's positive     K[] keys = this.keys;     int hashSize = keys.length;     int jump = 1 + theHashCode % (hashSize - 2...
...;       currentKey = keys[index];     }     return index;   }   private int findForAdd(Object key) {     int theHashCode = key.hashCode() & 0x7FFFFFFF; // make sure it's positive     K[] keys = this...
[+ show more]
  
Click probability prediction using Mahout. From model output to probability - Mahout - [mail # user]
... 000) and positive (target variable = 1, total examples = 2 000) sets.  Q1: Is it correct? Or should I downsample negative set from 250 000 to 2 000? (thanks to Mahout In Action examples)  Q2...
   Author: Abramov Pavel, 2012-12-27, 13:09
StableFixedSizeSamplingIterator.java - Mahout - [source code]
...(sofar, v));       } else {         int position = random.nextInt(sofar);         if (position < buf.size()) {           buf.set(position, new Pair(sofar, v...
  
MultiNormal - Mahout - [javadoc] - [see source code]
....  The Cholesky decomposition can be used to compute A if \Sigma is positive definite. Slightly more expensive is to use the SVD U S V' = \Sigma and then set A = U \sqrt{S}. Useful special cases occur...
http://search-lucene.com/jd/mahout/math/org/apache/mahout/m.../MultiNormal.html   
ConjugateGradientSolver # solve(org.apache.mahout.math.VectorIterable a, org.apache.mahout.math.Vector b, Preconditioner preconditioner, int maxIterations, double maxError) - Mahout - [javadoc] - [see source code]
... than zero, or if maxIterations is not positive. ...
http://search-lucene.com/jd/ma...hout.math.Vector, Preconditioner, int, double)   
Instance.java - Mahout - [source code]
... {      /** attributes, except LABEL and IGNORED */   private final Vector attrs;      public Instance(Vector attrs) {     this.attrs = attrs;   }      /**    * Return the attribute at the specified position...
...    *    * @param index    *          position of the attribute to retrieve    * @return value of the attribute    */   public double get(int index) {     return attrs.getQuick(index);   }      /**    * Set...
  
NegativeBinomial.java - Mahout - [source code]
... getting a fixed number of positive    * trials (r) where each positive trial has probability (p) of being successful.    *    * @param r               the required number of positive trials...
...);   }   /**    * Returns a sample from this distribution.  The value returned will    * be the number of negative samples required before achieving r    * positive samples.  Each successive sample is taken...
[+ show more]
  
Sort:
project
Lucene (7213)
Solr (1782)
ElasticSearch (571)
Mahout (419)
Lucene.Net (165)
ManifoldCF (133)
Nutch (92)
Tika (83)
Lucy (45)
PyLucene (11)
Droids (10)
OpenRelevance (6)
type
mail # user (204)
source code (70)
mail # dev (67)
javadoc (53)
issue (17)
wiki (8)
Sematext # blog (5)
date
last 7 days (0)
last 30 days (4)
last 90 days (12)
last 6 months (26)
last 9 months (301)
author
Ted Dunning (68)
Sean Owen (32)
Lance Norskog (18)
Jake Mannix (12)
Jeff Eastman (10)
Grant Ingersoll (9)
Robin Anil (6)
Benson Margulies (4)
Paulo Villegas (4)
Salman Mahmood (4)
Dan Brickley (3)
Dmitriy Lyubimov (3)
Drew Farris (3)
Paritosh Ranjan (3)
Sebastian Schelter (3)