Trevor Grant created MAHOUT-2006:
------------------------------------

             Summary: AsFactor has unexpected behavior when partitions not set
                 Key: MAHOUT-2006
                 URL: https://issues.apache.org/jira/browse/MAHOUT-2006
             Project: Mahout
          Issue Type: Bug
          Components: Algorithms
         Environment: Zeppelin Notebook, Spark 2.1, scala 2.10
            Reporter: Trevor Grant
             Fix For: 0.13.2




```
val drmA = drmParallelize(dense((0.0), (0.0), (1.0), (0.0), (2.0)), 
numPartitions = 2)

val factorizer = new AsFactor().fit(drmA)

val factoredA = factorizer.transform(drmA).collect
```

Yields:
```
drmA: org.apache.mahout.math.drm.CheckpointedDrm[Int] = 
org.apache.mahout.sparkbindings.drm.CheckpointedDrmSpark@75dcf2b2
factorizer: org.apache.mahout.math.algorithms.preprocessing.AsFactorModel = 
org.apache.mahout.math.algorithms.preprocessing.AsFactorModel@13b49f81
factoredA: org.apache.mahout.math.Matrix = 
{
 0 =>   {0:1.0}
 1 =>   {0:1.0}
 2 =>   {1:1.0}
 3 =>   {0:1.0}
 4 =>   {}
}
```

as expected, however

```
val drmA = drmParallelize(dense((0.0), (0.0), (1.0), (0.0), (2.0)))

val factorizer = new AsFactor().fit(drmA)

val factoredA = factorizer.transform(drmA).collect
```

Yields:
```
drmA: org.apache.mahout.math.drm.CheckpointedDrm[Int] = 
org.apache.mahout.sparkbindings.drm.CheckpointedDrmSpark@75dcf2b2
factorizer: org.apache.mahout.math.algorithms.preprocessing.AsFactorModel = 
org.apache.mahout.math.algorithms.preprocessing.AsFactorModel@13b49f81
factoredA: org.apache.mahout.math.Matrix = 
{
 0 =>   {}
 1 =>   {}
 2 =>   {}
 3 =>   {}
 4 =>   {}
}
```





--
This message was sent by Atlassian JIRA
(v6.4.14#64029)

Reply via email to