[
https://issues.apache.org/jira/browse/MAHOUT-1853?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15431098#comment-15431098
]
ASF GitHub Bot commented on MAHOUT-1853:
----------------------------------------
Github user dlyubimov commented on a diff in the pull request:
https://github.com/apache/mahout/pull/251#discussion_r75708181
--- Diff:
spark/src/test/scala/org/apache/mahout/cf/SimilarityAnalysisSuite.scala ---
@@ -91,13 +93,13 @@ class SimilarityAnalysisSuite extends FunSuite with
MahoutSuite with Distributed
//cross similarity
val matrixCrossCooc = drmCooc(1).checkpoint().collect
- val diff2Matrix = matrixCrossCooc.minus(matrixLLRCoocBtAControl)
+ val diff2Matrix = matrixCrossCooc.minus(matrixLLRCoocAtBControl)
n = (new MatrixOps(m = diff2Matrix)).norm
n should be < 1E-10
}
- test("cooccurrence [A'A], [B'A] double data using LLR") {
+ test("Cross-occurrence [A'A], [B'A] double data using LLR") {
val a = dense(
(100000.0D, 1.0D, 0.0D, 0.0D, 0.0D),
( 0.0D, 0.0D, 10.0D, 1.0D, 0.0D),
--- End diff --
same note. either 0.0 or 0d but not 0.0d
> Improvements to CCO (Correlated Cross-Occurrence)
> -------------------------------------------------
>
> Key: MAHOUT-1853
> URL: https://issues.apache.org/jira/browse/MAHOUT-1853
> Project: Mahout
> Issue Type: New Feature
> Affects Versions: 0.12.0
> Reporter: Andrew Palumbo
> Assignee: Pat Ferrel
> Fix For: 0.13.0
>
>
> Improvements to CCO (Correlated Cross-Occurrence) to include auto-threshold
> calculation for LLR downsampling, and possible multiple fixed thresholds for
> A’A, A’B etc. This is to account for the vast difference in dimensionality
> between indicator types.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
