Author: srowen
Date: Fri Apr 6 12:41:15 2012
New Revision: 1310302
URL: http://svn.apache.org/viewvc?rev=1310302&view=rev
Log:
MAHOUT-973 fix treatment of value as percentage
Modified:
mahout/trunk/core/src/main/java/org/apache/mahout/vectorizer/tfidf/TFIDFPartialVectorReducer.java
Modified:
mahout/trunk/core/src/main/java/org/apache/mahout/vectorizer/tfidf/TFIDFPartialVectorReducer.java
URL:
http://svn.apache.org/viewvc/mahout/trunk/core/src/main/java/org/apache/mahout/vectorizer/tfidf/TFIDFPartialVectorReducer.java?rev=1310302&r1=1310301&r2=1310302&view=diff
==============================================================================
---
mahout/trunk/core/src/main/java/org/apache/mahout/vectorizer/tfidf/TFIDFPartialVectorReducer.java
(original)
+++
mahout/trunk/core/src/main/java/org/apache/mahout/vectorizer/tfidf/TFIDFPartialVectorReducer.java
Fri Apr 6 12:41:15 2012
@@ -78,7 +78,7 @@ public class TFIDFPartialVectorReducer e
continue;
}
long df = dictionary.get(e.index());
- if (maxDf > -1 && df > maxDf) {
+ if (maxDf > -1 && (100.0 * df) / vectorCount > maxDf) {
continue;
}
if (df < minDf) {