You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@mahout.apache.org by sr...@apache.org on 2011/02/21 07:47:07 UTC
svn commit: r1072858 [1/3] - in /mahout/trunk: core/
core/src/main/java/org/apache/mahout/cf/taste/hadoop/item/
core/src/main/java/org/apache/mahout/cf/taste/hadoop/similarity/item/
core/src/main/java/org/apache/mahout/cf/taste/impl/common/jdbc/ core/s...
Author: srowen
Date: Mon Feb 21 06:47:02 2011
New Revision: 1072858
URL: http://svn.apache.org/viewvc?rev=1072858&view=rev
Log:
Another round of checkstyle / findbugs / PMD changes
Added:
mahout/trunk/utils/src/main/java/org/apache/mahout/utils/vectors/lucene/TermInfoClusterInOut.java
Modified:
mahout/trunk/core/ (props changed)
mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/hadoop/item/AggregateAndRecommendReducer.java
mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/hadoop/item/ItemFilterAsVectorAndPrefsReducer.java
mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/hadoop/similarity/item/MostSimilarItemPairsReducer.java
mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/hadoop/similarity/item/ToItemVectorsReducer.java
mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/common/jdbc/AbstractJDBCComponent.java
mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/eval/OrderBasedRecommenderEvaluator.java
mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/model/file/FileDataModel.java
mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/model/file/FileIDMigrator.java
mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/model/jdbc/AbstractBooleanPrefJDBCDataModel.java
mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/model/jdbc/PostgreSQLJDBCDataModel.java
mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/model/jdbc/ReloadFromJDBCDataModel.java
mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/recommender/FarthestNeighborClusterSimilarity.java
mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/recommender/TreeClusteringRecommender.java
mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/recommender/knn/ConjugateGradientOptimizer.java
mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/recommender/svd/ExpectationMaximizationSVDFactorizer.java
mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/recommender/svd/SVDRecommender.java
mahout/trunk/core/src/main/java/org/apache/mahout/classifier/AbstractVectorClassifier.java
mahout/trunk/core/src/main/java/org/apache/mahout/classifier/bayes/mapreduce/common/FeatureLabelComparator.java
mahout/trunk/core/src/main/java/org/apache/mahout/classifier/evaluation/Auc.java
mahout/trunk/core/src/main/java/org/apache/mahout/classifier/naivebayes/NaiveBayesModel.java
mahout/trunk/core/src/main/java/org/apache/mahout/classifier/naivebayes/StandardNaiveBayesClassifier.java
mahout/trunk/core/src/main/java/org/apache/mahout/classifier/naivebayes/trainer/NaiveBayesInstanceMapper.java
mahout/trunk/core/src/main/java/org/apache/mahout/classifier/naivebayes/trainer/NaiveBayesSumReducer.java
mahout/trunk/core/src/main/java/org/apache/mahout/classifier/naivebayes/trainer/NaiveBayesThetaComplementaryMapper.java
mahout/trunk/core/src/main/java/org/apache/mahout/classifier/naivebayes/trainer/NaiveBayesThetaMapper.java
mahout/trunk/core/src/main/java/org/apache/mahout/classifier/naivebayes/trainer/NaiveBayesWeightsMapper.java
mahout/trunk/core/src/main/java/org/apache/mahout/classifier/sequencelearning/hmm/HmmUtils.java
mahout/trunk/core/src/main/java/org/apache/mahout/classifier/sgd/AbstractOnlineLogisticRegression.java
mahout/trunk/core/src/main/java/org/apache/mahout/classifier/sgd/CsvRecordFactory.java
mahout/trunk/core/src/main/java/org/apache/mahout/classifier/sgd/ModelDissector.java
mahout/trunk/core/src/main/java/org/apache/mahout/classifier/sgd/PolymorphicWritable.java
mahout/trunk/core/src/main/java/org/apache/mahout/classifier/sgd/RankingGradient.java
mahout/trunk/core/src/main/java/org/apache/mahout/clustering/OnlineGaussianAccumulator.java
mahout/trunk/core/src/main/java/org/apache/mahout/clustering/RunningSumsGaussianAccumulator.java
mahout/trunk/core/src/main/java/org/apache/mahout/clustering/canopy/CanopyDriver.java
mahout/trunk/core/src/main/java/org/apache/mahout/clustering/dirichlet/DirichletDriver.java
mahout/trunk/core/src/main/java/org/apache/mahout/clustering/dirichlet/UncommonDistributions.java
mahout/trunk/core/src/main/java/org/apache/mahout/clustering/fuzzykmeans/FuzzyKMeansClusterMapper.java
mahout/trunk/core/src/main/java/org/apache/mahout/clustering/fuzzykmeans/FuzzyKMeansClusterer.java
mahout/trunk/core/src/main/java/org/apache/mahout/clustering/fuzzykmeans/FuzzyKMeansDriver.java
mahout/trunk/core/src/main/java/org/apache/mahout/clustering/kmeans/KMeansClusterer.java
mahout/trunk/core/src/main/java/org/apache/mahout/clustering/kmeans/KMeansDriver.java
mahout/trunk/core/src/main/java/org/apache/mahout/clustering/lda/LDADriver.java
mahout/trunk/core/src/main/java/org/apache/mahout/clustering/meanshift/MeanShiftCanopyDriver.java
mahout/trunk/core/src/main/java/org/apache/mahout/clustering/meanshift/MeanShiftCanopyReducer.java
mahout/trunk/core/src/main/java/org/apache/mahout/clustering/minhash/HashFactory.java
mahout/trunk/core/src/main/java/org/apache/mahout/clustering/minhash/MinHashMapper.java
mahout/trunk/core/src/main/java/org/apache/mahout/clustering/spectral/common/AffinityMatrixInputJob.java
mahout/trunk/core/src/main/java/org/apache/mahout/clustering/spectral/common/AffinityMatrixInputMapper.java
mahout/trunk/core/src/main/java/org/apache/mahout/clustering/spectral/common/IntDoublePairWritable.java
mahout/trunk/core/src/main/java/org/apache/mahout/clustering/spectral/common/VertexWritable.java
mahout/trunk/core/src/main/java/org/apache/mahout/clustering/spectral/eigencuts/EigencutsSensitivityMapper.java
mahout/trunk/core/src/main/java/org/apache/mahout/common/IntTuple.java
mahout/trunk/core/src/main/java/org/apache/mahout/common/iterator/StableFixedSizeSamplingIterator.java
mahout/trunk/core/src/main/java/org/apache/mahout/df/mapreduce/Builder.java
mahout/trunk/core/src/main/java/org/apache/mahout/df/mapreduce/Classifier.java
mahout/trunk/core/src/main/java/org/apache/mahout/df/mapreduce/partial/PartialBuilder.java
mahout/trunk/core/src/main/java/org/apache/mahout/df/mapreduce/partial/Step2Mapper.java
mahout/trunk/core/src/main/java/org/apache/mahout/math/MatrixWritable.java
mahout/trunk/core/src/main/java/org/apache/mahout/math/hadoop/TimesSquaredJob.java
mahout/trunk/core/src/main/java/org/apache/mahout/math/hadoop/decomposer/DistributedLanczosSolver.java
mahout/trunk/core/src/main/java/org/apache/mahout/math/hadoop/similarity/RowSimilarityJob.java
mahout/trunk/core/src/main/java/org/apache/mahout/vectorizer/collocations/llr/CollocReducer.java
mahout/trunk/core/src/main/java/org/apache/mahout/vectorizer/encoders/ConstantValueEncoder.java
mahout/trunk/core/src/main/java/org/apache/mahout/vectorizer/encoders/ContinuousValueEncoder.java
mahout/trunk/core/src/main/java/org/apache/mahout/vectorizer/encoders/FeatureVectorEncoder.java
mahout/trunk/core/src/main/java/org/apache/mahout/vectorizer/encoders/LuceneTextValueEncoder.java
mahout/trunk/core/src/main/java/org/apache/mahout/vectorizer/encoders/TextValueEncoder.java
mahout/trunk/core/src/main/java/org/apache/mahout/vectorizer/encoders/WordValueEncoder.java
mahout/trunk/core/src/test/java/org/apache/mahout/clustering/TestClusterInterface.java
mahout/trunk/core/src/test/java/org/apache/mahout/clustering/TestGaussianAccumulators.java
mahout/trunk/examples/ (props changed)
mahout/trunk/examples/src/main/java/org/apache/mahout/classifier/bayes/SplitBayesInput.java
mahout/trunk/examples/src/main/java/org/apache/mahout/classifier/sgd/SimpleCsvExamples.java
mahout/trunk/examples/src/main/java/org/apache/mahout/classifier/sgd/TrainLogistic.java
mahout/trunk/examples/src/main/java/org/apache/mahout/classifier/sgd/TrainNewsGroups.java
mahout/trunk/examples/src/main/java/org/apache/mahout/clustering/display/DisplayFuzzyKMeans.java
mahout/trunk/examples/src/main/java/org/apache/mahout/clustering/minhash/LastfmClusterEvaluator.java
mahout/trunk/examples/src/main/java/org/apache/mahout/clustering/syntheticcontrol/dirichlet/Job.java
mahout/trunk/examples/src/main/java/org/apache/mahout/clustering/syntheticcontrol/kmeans/Job.java
mahout/trunk/examples/src/main/java/org/apache/mahout/ga/watchmaker/cd/tool/CDInfosTool.java
mahout/trunk/math/src/main/java/org/apache/mahout/math/AbstractMatrix.java
mahout/trunk/math/src/main/java/org/apache/mahout/math/AbstractVector.java
mahout/trunk/math/src/main/java/org/apache/mahout/math/OrthonormalityVerifier.java
mahout/trunk/math/src/main/java/org/apache/mahout/math/SingularValueDecomposition.java
mahout/trunk/math/src/main/java/org/apache/mahout/math/VectorList.java
mahout/trunk/math/src/main/java/org/apache/mahout/math/decomposer/hebbian/HebbianSolver.java
mahout/trunk/math/src/main/java/org/apache/mahout/math/decomposer/lanczos/LanczosSolver.java
mahout/trunk/math/src/main/java/org/apache/mahout/math/jet/math/Polynomial.java
mahout/trunk/math/src/main/java/org/apache/mahout/math/jet/random/Gamma.java
mahout/trunk/math/src/main/java/org/apache/mahout/math/jet/stat/Gamma.java
mahout/trunk/math/src/main/java/org/apache/mahout/math/matrix/linalg/EigenvalueDecomposition.java
mahout/trunk/taste-web/ (props changed)
mahout/trunk/taste-web/src/main/java/org/apache/mahout/cf/taste/web/RecommenderWrapper.java
mahout/trunk/utils/ (props changed)
mahout/trunk/utils/src/main/java/org/apache/mahout/clustering/cdbw/CDbwEvaluator.java
mahout/trunk/utils/src/main/java/org/apache/mahout/clustering/evaluation/ClusterEvaluator.java
mahout/trunk/utils/src/main/java/org/apache/mahout/clustering/evaluation/RepresentativePointsDriver.java
mahout/trunk/utils/src/main/java/org/apache/mahout/text/SequenceFilesFromDirectory.java
mahout/trunk/utils/src/main/java/org/apache/mahout/utils/clustering/ClusterDumper.java
mahout/trunk/utils/src/main/java/org/apache/mahout/utils/vectors/TermEntry.java
mahout/trunk/utils/src/main/java/org/apache/mahout/utils/vectors/io/JWriterTermInfoWriter.java
mahout/trunk/utils/src/main/java/org/apache/mahout/utils/vectors/lucene/CachedTermInfo.java
mahout/trunk/utils/src/main/java/org/apache/mahout/utils/vectors/lucene/ClusterLabels.java
mahout/trunk/utils/src/main/java/org/apache/mahout/utils/vectors/lucene/TFDFMapper.java
mahout/trunk/utils/src/test/java/org/apache/mahout/clustering/TestClusterDumper.java
Propchange: mahout/trunk/core/
------------------------------------------------------------------------------
--- svn:ignore (original)
+++ svn:ignore Mon Feb 21 06:47:02 2011
@@ -1,15 +1,10 @@
-build
-input
-output
-dist
-testdata
+.pmd
+.classpath
test
-target
*.iml
-.settings
-.classpath
+testdatatestdata
.project
-.checkstyle
-.pmd
.ruleset
-testdatatestdata
+.checkstyle
+dist
+.settings
Modified: mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/hadoop/item/AggregateAndRecommendReducer.java
URL: http://svn.apache.org/viewvc/mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/hadoop/item/AggregateAndRecommendReducer.java?rev=1072858&r1=1072857&r2=1072858&view=diff
==============================================================================
--- mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/hadoop/item/AggregateAndRecommendReducer.java (original)
+++ mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/hadoop/item/AggregateAndRecommendReducer.java Mon Feb 21 06:47:02 2011
@@ -184,9 +184,10 @@ public final class AggregateAndRecommend
* find the top entries in recommendationVector, map them to the real itemIDs and write back the result
*/
private void writeRecommendedItems(VarLongWritable userID, Vector recommendationVector, Context context)
- throws IOException, InterruptedException {
- Queue<RecommendedItem> topItems = new PriorityQueue<RecommendedItem>(recommendationsPerUser + 1,
- Collections.reverseOrder(ByValueRecommendedItemComparator.getInstance()));
+ throws IOException, InterruptedException {
+ Queue<RecommendedItem> topItems =
+ new PriorityQueue<RecommendedItem>(recommendationsPerUser + 1,
+ Collections.reverseOrder(ByValueRecommendedItemComparator.getInstance()));
Iterator<Vector.Element> recommendationVectorIterator = recommendationVector.iterateNonZero();
while (recommendationVectorIterator.hasNext()) {
Modified: mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/hadoop/item/ItemFilterAsVectorAndPrefsReducer.java
URL: http://svn.apache.org/viewvc/mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/hadoop/item/ItemFilterAsVectorAndPrefsReducer.java?rev=1072858&r1=1072857&r2=1072858&view=diff
==============================================================================
--- mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/hadoop/item/ItemFilterAsVectorAndPrefsReducer.java (original)
+++ mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/hadoop/item/ItemFilterAsVectorAndPrefsReducer.java Mon Feb 21 06:47:02 2011
@@ -37,7 +37,8 @@ public class ItemFilterAsVectorAndPrefsR
extends Reducer<VarLongWritable,VarLongWritable,VarIntWritable,VectorAndPrefsWritable> {
@Override
protected void reduce(VarLongWritable itemID, Iterable<VarLongWritable> values, Context ctx)
- throws IOException, InterruptedException {
+ throws IOException, InterruptedException {
+
int itemIDIndex = TasteHadoopUtils.idToIndex(itemID.get());
Vector vector = new RandomAccessSparseVector(Integer.MAX_VALUE, 1);
/* artificial NaN summand to exclude this item from the recommendations for all users specified in userIDs */
Modified: mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/hadoop/similarity/item/MostSimilarItemPairsReducer.java
URL: http://svn.apache.org/viewvc/mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/hadoop/similarity/item/MostSimilarItemPairsReducer.java?rev=1072858&r1=1072857&r2=1072858&view=diff
==============================================================================
--- mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/hadoop/similarity/item/MostSimilarItemPairsReducer.java (original)
+++ mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/hadoop/similarity/item/MostSimilarItemPairsReducer.java Mon Feb 21 06:47:02 2011
@@ -28,7 +28,7 @@ public class MostSimilarItemPairsReducer
@Override
protected void reduce(EntityEntityWritable itemIDPair, Iterable<DoubleWritable> values, Context ctx)
- throws IOException, InterruptedException {
+ throws IOException, InterruptedException {
ctx.write(itemIDPair, values.iterator().next());
}
}
Modified: mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/hadoop/similarity/item/ToItemVectorsReducer.java
URL: http://svn.apache.org/viewvc/mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/hadoop/similarity/item/ToItemVectorsReducer.java?rev=1072858&r1=1072857&r2=1072858&view=diff
==============================================================================
--- mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/hadoop/similarity/item/ToItemVectorsReducer.java (original)
+++ mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/hadoop/similarity/item/ToItemVectorsReducer.java Mon Feb 21 06:47:02 2011
@@ -31,7 +31,7 @@ public class ToItemVectorsReducer
@Override
protected void reduce(IntWritable rowIndex, Iterable<DistributedRowMatrix.MatrixEntryWritable> values, Context ctx)
- throws IOException, InterruptedException {
+ throws IOException, InterruptedException {
Vector vector = new RandomAccessSparseVector(Integer.MAX_VALUE, 100);
for (DistributedRowMatrix.MatrixEntryWritable entry : values) {
Modified: mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/common/jdbc/AbstractJDBCComponent.java
URL: http://svn.apache.org/viewvc/mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/common/jdbc/AbstractJDBCComponent.java?rev=1072858&r1=1072857&r2=1072858&view=diff
==============================================================================
--- mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/common/jdbc/AbstractJDBCComponent.java (original)
+++ mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/common/jdbc/AbstractJDBCComponent.java Mon Feb 21 06:47:02 2011
@@ -48,8 +48,7 @@ public abstract class AbstractJDBCCompon
}
protected static void checkNotNullAndLog(String argName, Object[] values) {
- Preconditions.checkArgument(values != null && values.length != 0,
- argName + " is null or zero-length" );
+ Preconditions.checkArgument(values != null && values.length != 0, argName + " is null or zero-length");
for (Object value : values) {
checkNotNullAndLog(argName, value);
}
Modified: mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/eval/OrderBasedRecommenderEvaluator.java
URL: http://svn.apache.org/viewvc/mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/eval/OrderBasedRecommenderEvaluator.java?rev=1072858&r1=1072857&r2=1072858&view=diff
==============================================================================
--- mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/eval/OrderBasedRecommenderEvaluator.java (original)
+++ mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/eval/OrderBasedRecommenderEvaluator.java Mon Feb 21 06:47:02 2011
@@ -247,14 +247,11 @@ public final class OrderBasedRecommender
}
for (int i = 1; i < samples - 1; i++) {
long itemID = itemsL[i];
- if ((itemsR[i] == itemID) ||
- (itemsR[i - 1] == itemID) ||
- (itemsR[i + 1] == itemID)) {
+ if (itemsR[i] == itemID || itemsR[i - 1] == itemID || itemsR[i + 1] == itemID) {
count++;
}
}
- if (itemsR[samples - 1].equals(itemsL[samples - 1]) ||
- itemsR[samples - 1].equals(itemsL[samples - 2])) {
+ if (itemsR[samples - 1].equals(itemsL[samples - 1]) || itemsR[samples - 1].equals(itemsL[samples - 2])) {
count++;
}
return count;
@@ -416,8 +413,7 @@ public final class OrderBasedRecommender
jump++;
}
}
- if ((j + jump < length) &&
- !(reference[j] == sortable[j] && reference[j + jump] == sortable[j + jump])) {
+ if ((j + jump < length) && !(reference[j] == sortable[j] && reference[j + jump] == sortable[j + jump])) {
long tmp = sortable[j];
sortable[j] = sortable[j + 1];
sortable[j + 1] = tmp;
Modified: mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/model/file/FileDataModel.java
URL: http://svn.apache.org/viewvc/mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/model/file/FileDataModel.java?rev=1072858&r1=1072857&r2=1072858&view=diff
==============================================================================
--- mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/model/file/FileDataModel.java (original)
+++ mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/model/file/FileDataModel.java Mon Feb 21 06:47:02 2011
@@ -23,7 +23,6 @@ import java.io.IOException;
import java.util.ArrayList;
import java.util.Collection;
import java.util.Iterator;
-import java.util.List;
import java.util.Map;
import java.util.TreeMap;
import java.util.concurrent.locks.ReentrantLock;
@@ -341,8 +340,8 @@ public class FileDataModel extends Abstr
* Reads one line from the input file and adds the data to a {@link FastByIDMap} data structure which maps user IDs
* to preferences. This assumes that each line of the input file corresponds to one preference. After
* reading a line and determining which user and item the preference pertains to, the method should look to
- * see if the data contains a mapping for the user ID already, and if not, add an empty {@link List} of
- * {@link Preference}s to the data.
+ * see if the data contains a mapping for the user ID already, and if not, add an empty data structure of preferences
+ * as appropriate to the data.
* </p>
*
* <p>
Modified: mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/model/file/FileIDMigrator.java
URL: http://svn.apache.org/viewvc/mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/model/file/FileIDMigrator.java?rev=1072858&r1=1072857&r2=1072858&view=diff
==============================================================================
--- mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/model/file/FileIDMigrator.java (original)
+++ mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/model/file/FileIDMigrator.java Mon Feb 21 06:47:02 2011
@@ -25,7 +25,6 @@ import java.util.concurrent.locks.Reentr
import org.apache.mahout.cf.taste.common.Refreshable;
import org.apache.mahout.cf.taste.impl.common.FastByIDMap;
import org.apache.mahout.cf.taste.impl.model.AbstractIDMigrator;
-import org.apache.mahout.cf.taste.model.IDMigrator;
import org.apache.mahout.common.FileLineIterable;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
@@ -34,7 +33,8 @@ import com.google.common.base.Preconditi
/**
* <p>
- * An {@link IDMigrator} backed by a file. This class typically expects a file where each line
+ * An {@link org.apache.mahout.cf.taste.model.IDMigrator} backed by a file.
+ * This class typically expects a file where each line
* contains a single stringID to be stored in this migrator.
* </p>
*
Modified: mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/model/jdbc/AbstractBooleanPrefJDBCDataModel.java
URL: http://svn.apache.org/viewvc/mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/model/jdbc/AbstractBooleanPrefJDBCDataModel.java?rev=1072858&r1=1072857&r2=1072858&view=diff
==============================================================================
--- mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/model/jdbc/AbstractBooleanPrefJDBCDataModel.java (original)
+++ mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/model/jdbc/AbstractBooleanPrefJDBCDataModel.java Mon Feb 21 06:47:02 2011
@@ -96,7 +96,7 @@ public abstract class AbstractBooleanPre
@Override
public void setPreference(long userID, long itemID, float value) throws TasteException {
- Preconditions.checkArgument(!Float.isNaN(value), "Invalid value: " + value );
+ Preconditions.checkArgument(!Float.isNaN(value), "Invalid value: " + value);
log.debug("Setting preference for user {}, item {}", userID, itemID);
Connection conn = null;
Modified: mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/model/jdbc/PostgreSQLJDBCDataModel.java
URL: http://svn.apache.org/viewvc/mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/model/jdbc/PostgreSQLJDBCDataModel.java?rev=1072858&r1=1072857&r2=1072858&view=diff
==============================================================================
--- mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/model/jdbc/PostgreSQLJDBCDataModel.java (original)
+++ mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/model/jdbc/PostgreSQLJDBCDataModel.java Mon Feb 21 06:47:02 2011
@@ -202,7 +202,7 @@ public class PostgreSQLJDBCDataModel ext
log.debug("Executing SQL update: {}", setPreferenceSQL);
try {
- stmt1.executeUpdate();
+ stmt1.executeUpdate();
} catch (SQLException sqle) {
if (!POSTGRESQL_DUPLICATE_KEY_STATE.equals(sqle.getSQLState())) {
throw sqle;
Modified: mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/model/jdbc/ReloadFromJDBCDataModel.java
URL: http://svn.apache.org/viewvc/mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/model/jdbc/ReloadFromJDBCDataModel.java?rev=1072858&r1=1072857&r2=1072858&view=diff
==============================================================================
--- mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/model/jdbc/ReloadFromJDBCDataModel.java (original)
+++ mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/model/jdbc/ReloadFromJDBCDataModel.java Mon Feb 21 06:47:02 2011
@@ -73,9 +73,10 @@ public final class ReloadFromJDBCDataMod
try {
// Load new in-memory representation,
log.info("Loading new JDBC delegate data...");
- DataModel newDelegateInMemory = delegate.hasPreferenceValues() ?
- new GenericDataModel(delegate.exportWithPrefs()) :
- new GenericBooleanPrefDataModel(delegate.exportWithIDsOnly());
+ DataModel newDelegateInMemory =
+ delegate.hasPreferenceValues()
+ ? new GenericDataModel(delegate.exportWithPrefs())
+ : new GenericBooleanPrefDataModel(delegate.exportWithIDsOnly());
// and then swap to it.
log.info("New data loaded.");
delegateInMemory = newDelegateInMemory;
Modified: mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/recommender/FarthestNeighborClusterSimilarity.java
URL: http://svn.apache.org/viewvc/mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/recommender/FarthestNeighborClusterSimilarity.java?rev=1072858&r1=1072857&r2=1072858&view=diff
==============================================================================
--- mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/recommender/FarthestNeighborClusterSimilarity.java (original)
+++ mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/recommender/FarthestNeighborClusterSimilarity.java Mon Feb 21 06:47:02 2011
@@ -61,7 +61,8 @@ public final class FarthestNeighborClust
*/
public FarthestNeighborClusterSimilarity(UserSimilarity similarity, double samplingRate) {
Preconditions.checkArgument(similarity != null, "similarity is null");
- Preconditions.checkArgument(!Double.isNaN(samplingRate) && samplingRate > 0.0 && samplingRate <= 1.0, "samplingRate is invalid: %.4f", samplingRate);
+ Preconditions.checkArgument(!Double.isNaN(samplingRate) && samplingRate > 0.0 && samplingRate <= 1.0,
+ "samplingRate is invalid: %.4f", samplingRate);
this.similarity = similarity;
this.samplingRate = samplingRate;
}
Modified: mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/recommender/TreeClusteringRecommender.java
URL: http://svn.apache.org/viewvc/mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/recommender/TreeClusteringRecommender.java?rev=1072858&r1=1072857&r2=1072858&view=diff
==============================================================================
--- mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/recommender/TreeClusteringRecommender.java (original)
+++ mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/recommender/TreeClusteringRecommender.java Mon Feb 21 06:47:02 2011
@@ -169,7 +169,7 @@ public final class TreeClusteringRecomme
double clusteringThreshold,
double samplingRate) throws TasteException {
super(dataModel);
- Preconditions.checkArgument( clusterSimilarity != null, "clusterSimilarity is null" );
+ Preconditions.checkArgument(clusterSimilarity != null, "clusterSimilarity is null");
Preconditions.checkArgument(!Double.isNaN(clusteringThreshold), "clusteringThreshold must not be NaN");
Preconditions.checkArgument(samplingRate > 0.0 && samplingRate <= 1.0, "samplingRate is invalid: %f", samplingRate);
this.clusterSimilarity = clusterSimilarity;
Modified: mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/recommender/knn/ConjugateGradientOptimizer.java
URL: http://svn.apache.org/viewvc/mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/recommender/knn/ConjugateGradientOptimizer.java?rev=1072858&r1=1072857&r2=1072858&view=diff
==============================================================================
--- mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/recommender/knn/ConjugateGradientOptimizer.java (original)
+++ mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/recommender/knn/ConjugateGradientOptimizer.java Mon Feb 21 06:47:02 2011
@@ -72,9 +72,9 @@ public final class ConjugateGradientOpti
// w = -r;
for (int i = 0; i < k; i++) {
double v = 0.0;
- double[] Ai = matrix[i];
+ double[] ai = matrix[i];
for (int j = 0; j < k; j++) {
- v += Ai[j] * x[j];
+ v += ai[j] * x[j];
}
double ri = b[i] - v;
r[i] = ri;
@@ -86,9 +86,9 @@ public final class ConjugateGradientOpti
// z = A*w;
for (int i = 0; i < k; i++) {
double v = 0.0;
- double[] Ai = matrix[i];
+ double[] ai = matrix[i];
for (int j = 0; j < k; j++) {
- v += Ai[j] * w[j];
+ v += ai[j] * w[j];
}
z[i] = v;
}
@@ -120,14 +120,14 @@ public final class ConjugateGradientOpti
}
// B = (r'*z)/(w'*z);
- double Bnum = 0.0;
- double Bden = 0.0;
+ double bnum = 0.0;
+ double bden = 0.0;
for (int i = 0; i < k; i++) {
double zi = z[i];
- Bnum += r[i] * zi;
- Bden += w[i] * zi;
+ bnum += r[i] * zi;
+ bden += w[i] * zi;
}
- double B = Bnum / Bden;
+ double B = bnum / bden;
// w = -r + B*w;
for (int i = 0; i < k; i++) {
Modified: mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/recommender/svd/ExpectationMaximizationSVDFactorizer.java
URL: http://svn.apache.org/viewvc/mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/recommender/svd/ExpectationMaximizationSVDFactorizer.java?rev=1072858&r1=1072857&r2=1072858&view=diff
==============================================================================
--- mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/recommender/svd/ExpectationMaximizationSVDFactorizer.java (original)
+++ mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/recommender/svd/ExpectationMaximizationSVDFactorizer.java Mon Feb 21 06:47:02 2011
@@ -60,14 +60,19 @@ public class ExpectationMaximizationSVDF
private static final Logger log = LoggerFactory.getLogger(ExpectationMaximizationSVDFactorizer.class);
- public ExpectationMaximizationSVDFactorizer(DataModel dataModel, int numFeatures, int numIterations)
- throws TasteException {
- /* use the default parameters from the old SVDRecommender implementation */
+ public ExpectationMaximizationSVDFactorizer(DataModel dataModel,
+ int numFeatures,
+ int numIterations) throws TasteException {
+ // use the default parameters from the old SVDRecommender implementation
this(dataModel, numFeatures, 0.005, 0.02, 0.005, numIterations);
}
- public ExpectationMaximizationSVDFactorizer(DataModel dataModel, int numFeatures, double learningRate,
- double preventOverfitting, double randomNoise, int numIterations) throws TasteException {
+ public ExpectationMaximizationSVDFactorizer(DataModel dataModel,
+ int numFeatures,
+ double learningRate,
+ double preventOverfitting,
+ double randomNoise,
+ int numIterations) throws TasteException {
super(dataModel);
random = RandomUtils.getRandom();
this.dataModel = dataModel;
@@ -151,10 +156,10 @@ public class ExpectationMaximizationSVDF
double err = value - getDotProduct(userIndex, itemIndex);
double[] leftVector = leftVectors[userIndex];
double[] rightVector = rightVectors[itemIndex];
- leftVector[currentFeature] += learningRate *
- (err * rightVector[currentFeature] - preventOverfitting * leftVector[currentFeature]);
- rightVector[currentFeature] += learningRate *
- (err * leftVector[currentFeature] - preventOverfitting * rightVector[currentFeature]);
+ leftVector[currentFeature] +=
+ learningRate * (err * rightVector[currentFeature] - preventOverfitting * leftVector[currentFeature]);
+ rightVector[currentFeature] +=
+ learningRate * (err * leftVector[currentFeature] - preventOverfitting * rightVector[currentFeature]);
}
}
Modified: mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/recommender/svd/SVDRecommender.java
URL: http://svn.apache.org/viewvc/mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/recommender/svd/SVDRecommender.java?rev=1072858&r1=1072857&r2=1072858&view=diff
==============================================================================
--- mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/recommender/svd/SVDRecommender.java (original)
+++ mahout/trunk/core/src/main/java/org/apache/mahout/cf/taste/impl/recommender/svd/SVDRecommender.java Mon Feb 21 06:47:02 2011
@@ -51,12 +51,12 @@ public final class SVDRecommender extend
}
public SVDRecommender(DataModel dataModel, Factorizer factorizer, CandidateItemsStrategy candidateItemsStrategy)
- throws TasteException {
+ throws TasteException {
super(dataModel, candidateItemsStrategy);
factorization = factorizer.factorize();
refreshHelper = new RefreshHelper(new Callable<Object>() {
@Override
- public Object call() throws TasteException {
+ public Object call() {
// TODO: train again
return null;
}
Modified: mahout/trunk/core/src/main/java/org/apache/mahout/classifier/AbstractVectorClassifier.java
URL: http://svn.apache.org/viewvc/mahout/trunk/core/src/main/java/org/apache/mahout/classifier/AbstractVectorClassifier.java?rev=1072858&r1=1072857&r2=1072858&view=diff
==============================================================================
--- mahout/trunk/core/src/main/java/org/apache/mahout/classifier/AbstractVectorClassifier.java (original)
+++ mahout/trunk/core/src/main/java/org/apache/mahout/classifier/AbstractVectorClassifier.java Mon Feb 21 06:47:02 2011
@@ -56,8 +56,8 @@ public abstract class AbstractVectorClas
* @return A vector of scores. If transformed by the link function, these will become probabilities.
*/
public Vector classifyNoLink(Vector features) {
- throw new UnsupportedOperationException("Classifier " + this.getClass().getName() +
- " doesn't support classification without a link");
+ throw new UnsupportedOperationException(
+ this.getClass().getName() + " doesn't support classification without a link");
}
/**
@@ -101,7 +101,7 @@ public abstract class AbstractVectorClas
*/
public Vector classifyFull(Vector r, Vector instance) {
r.viewPart(1, numCategories() - 1).assign(classify(instance));
- r.setQuick(0, 1 - r.zSum());
+ r.setQuick(0, 1.0 - r.zSum());
return r;
}
@@ -168,16 +168,16 @@ public abstract class AbstractVectorClas
if (numCategories() == 2) {
double p = classifyScalar(data);
if (actual > 0) {
- return Math.max(-100, Math.log(p));
+ return Math.max(-100.0, Math.log(p));
} else {
- return Math.max(-100, Math.log(1 - p));
+ return Math.max(-100.0, Math.log(1.0 - p));
}
} else {
Vector p = classify(data);
if (actual > 0) {
- return Math.max(-100, Math.log(p.get(actual - 1)));
+ return Math.max(-100.0, Math.log(p.get(actual - 1)));
} else {
- return Math.max(-100, Math.log(1 - p.zSum()));
+ return Math.max(-100.0, Math.log(1.0 - p.zSum()));
}
}
}
Modified: mahout/trunk/core/src/main/java/org/apache/mahout/classifier/bayes/mapreduce/common/FeatureLabelComparator.java
URL: http://svn.apache.org/viewvc/mahout/trunk/core/src/main/java/org/apache/mahout/classifier/bayes/mapreduce/common/FeatureLabelComparator.java?rev=1072858&r1=1072857&r2=1072858&view=diff
==============================================================================
--- mahout/trunk/core/src/main/java/org/apache/mahout/classifier/bayes/mapreduce/common/FeatureLabelComparator.java (original)
+++ mahout/trunk/core/src/main/java/org/apache/mahout/classifier/bayes/mapreduce/common/FeatureLabelComparator.java Mon Feb 21 06:47:02 2011
@@ -37,7 +37,8 @@ public class FeatureLabelComparator exte
public int compare(WritableComparable a, WritableComparable b) {
StringTuple ta = (StringTuple) a;
StringTuple tb = (StringTuple) b;
- Preconditions.checkArgument(ta.length() >= 2 && ta.length() <= 3 && tb.length() >= 2 && tb.length() <= 3, "StringTuple length out of bounds");
+ Preconditions.checkArgument(ta.length() >= 2 && ta.length() <= 3 && tb.length() >= 2 && tb.length() <= 3,
+ "StringTuple length out of bounds");
// token
String tmpa = ta.length() == 2 ? ta.stringAt(1) : ta.stringAt(2);
String tmpb = tb.length() == 2 ? tb.stringAt(1) : tb.stringAt(2);
@@ -57,17 +58,15 @@ public class FeatureLabelComparator exte
return -1;
} else if (tb.stringAt(0).equals(BayesConstants.FEATURE_COUNT)) {
return 1;
- } else {
- return cmp;
}
+ return cmp;
}
// label or empty.
tmpa = ta.length() == 2 ? "" : ta.stringAt(1);
tmpb = tb.length() == 2 ? "" : tb.stringAt(1);
- cmp = tmpa.compareTo(tmpb);
- return cmp;
+ return tmpa.compareTo(tmpb);
}
}
Modified: mahout/trunk/core/src/main/java/org/apache/mahout/classifier/evaluation/Auc.java
URL: http://svn.apache.org/viewvc/mahout/trunk/core/src/main/java/org/apache/mahout/classifier/evaluation/Auc.java?rev=1072858&r1=1072857&r2=1072858&view=diff
==============================================================================
--- mahout/trunk/core/src/main/java/org/apache/mahout/classifier/evaluation/Auc.java (original)
+++ mahout/trunk/core/src/main/java/org/apache/mahout/classifier/evaluation/Auc.java Mon Feb 21 06:47:02 2011
@@ -70,9 +70,9 @@ public class Auc {
* @param score The score for this example.
*/
public void add(int trueValue, double score) {
+ Preconditions.checkArgument(trueValue == 0 || trueValue == 1, "True value must be 0 or 1");
hasScore = true;
- Preconditions.checkArgument(trueValue == 0 || trueValue == 1, "True value must be 0 or 1" );
-
+
int predictedClass = (score > threshold) ? 1 : 0;
confusion.set(trueValue, predictedClass, confusion.get(trueValue, predictedClass) + 1);
Modified: mahout/trunk/core/src/main/java/org/apache/mahout/classifier/naivebayes/NaiveBayesModel.java
URL: http://svn.apache.org/viewvc/mahout/trunk/core/src/main/java/org/apache/mahout/classifier/naivebayes/NaiveBayesModel.java?rev=1072858&r1=1072857&r2=1072858&view=diff
==============================================================================
--- mahout/trunk/core/src/main/java/org/apache/mahout/classifier/naivebayes/NaiveBayesModel.java (original)
+++ mahout/trunk/core/src/main/java/org/apache/mahout/classifier/naivebayes/NaiveBayesModel.java Mon Feb 21 06:47:02 2011
@@ -41,17 +41,14 @@ import com.google.gson.JsonDeserializati
import com.google.gson.JsonDeserializer;
import com.google.gson.JsonElement;
import com.google.gson.JsonObject;
-import com.google.gson.JsonParseException;
import com.google.gson.JsonPrimitive;
import com.google.gson.JsonSerializationContext;
import com.google.gson.JsonSerializer;
/**
- *
* NaiveBayesModel holds the weight Matrix, the feature and label sums and the weight normalizer vectors.
- *
*/
-public class NaiveBayesModel implements JsonDeserializer<NaiveBayesModel>, JsonSerializer<NaiveBayesModel>, Cloneable {
+public class NaiveBayesModel implements JsonDeserializer<NaiveBayesModel>, JsonSerializer<NaiveBayesModel> {
private Vector labelSum;
private Vector perlabelThetaNormalizer;
@@ -246,7 +243,7 @@ public class NaiveBayesModel implements
@Override
public NaiveBayesModel deserialize(JsonElement json,
Type type,
- JsonDeserializationContext context) throws JsonParseException {
+ JsonDeserializationContext context) {
// register the builders for matrix / vector
GsonBuilder builder = new GsonBuilder();
builder.registerTypeAdapter(Matrix.class, new JsonMatrixAdapter());
@@ -290,13 +287,13 @@ public class NaiveBayesModel implements
"Error: The number of labels has to be greater than 0 or defined!");
}
- if (model.getPerlabelThetaNormalizer() == null ||
- model.getPerlabelThetaNormalizer().getNumNondefaultElements() <= 0) {
+ if (model.getPerlabelThetaNormalizer() == null
+ || model.getPerlabelThetaNormalizer().getNumNondefaultElements() <= 0) {
throw new IllegalArgumentException(
"Error: The number of theta normalizers has to be greater than 0 or defined!");
}
- if (model.getFeatureSum() == null ||model.getFeatureSum().getNumNondefaultElements() <= 0) {
+ if (model.getFeatureSum() == null || model.getFeatureSum().getNumNondefaultElements() <= 0) {
throw new IllegalArgumentException(
"Error: The number of features has to be greater than 0 or defined!");
}
Modified: mahout/trunk/core/src/main/java/org/apache/mahout/classifier/naivebayes/StandardNaiveBayesClassifier.java
URL: http://svn.apache.org/viewvc/mahout/trunk/core/src/main/java/org/apache/mahout/classifier/naivebayes/StandardNaiveBayesClassifier.java?rev=1072858&r1=1072857&r2=1072858&view=diff
==============================================================================
--- mahout/trunk/core/src/main/java/org/apache/mahout/classifier/naivebayes/StandardNaiveBayesClassifier.java (original)
+++ mahout/trunk/core/src/main/java/org/apache/mahout/classifier/naivebayes/StandardNaiveBayesClassifier.java Mon Feb 21 06:47:02 2011
@@ -36,7 +36,7 @@ public class StandardNaiveBayesClassifie
double sumLabelWeight = model.getLabelSum().get(label);
double numerator = result + model.getAlphaI();
double denominator = sumLabelWeight + vocabCount;
- double weight = - Math.log(numerator / denominator);
+ double weight = -Math.log(numerator / denominator);
result = weight / model.getPerlabelThetaNormalizer().get(label);
return result;
}
Modified: mahout/trunk/core/src/main/java/org/apache/mahout/classifier/naivebayes/trainer/NaiveBayesInstanceMapper.java
URL: http://svn.apache.org/viewvc/mahout/trunk/core/src/main/java/org/apache/mahout/classifier/naivebayes/trainer/NaiveBayesInstanceMapper.java?rev=1072858&r1=1072857&r2=1072858&view=diff
==============================================================================
--- mahout/trunk/core/src/main/java/org/apache/mahout/classifier/naivebayes/trainer/NaiveBayesInstanceMapper.java (original)
+++ mahout/trunk/core/src/main/java/org/apache/mahout/classifier/naivebayes/trainer/NaiveBayesInstanceMapper.java Mon Feb 21 06:47:02 2011
@@ -38,7 +38,7 @@ public class NaiveBayesInstanceMapper ex
@Override
protected void map(Text key, VectorWritable value, Context context)
- throws IOException, InterruptedException {
+ throws IOException, InterruptedException {
if (!labelMap.containsKey(key.toString())) {
context.getCounter("NaiveBayes", "Skipped instance: not in label list").increment(1);
return;
Modified: mahout/trunk/core/src/main/java/org/apache/mahout/classifier/naivebayes/trainer/NaiveBayesSumReducer.java
URL: http://svn.apache.org/viewvc/mahout/trunk/core/src/main/java/org/apache/mahout/classifier/naivebayes/trainer/NaiveBayesSumReducer.java?rev=1072858&r1=1072857&r2=1072858&view=diff
==============================================================================
--- mahout/trunk/core/src/main/java/org/apache/mahout/classifier/naivebayes/trainer/NaiveBayesSumReducer.java (original)
+++ mahout/trunk/core/src/main/java/org/apache/mahout/classifier/naivebayes/trainer/NaiveBayesSumReducer.java Mon Feb 21 06:47:02 2011
@@ -27,11 +27,12 @@ import org.apache.mahout.math.VectorWrit
/**
* Can also be used as a local Combiner. This accumulates all the features and the weights and sums them up.
*/
-public class NaiveBayesSumReducer extends Reducer<WritableComparable<?>, VectorWritable, WritableComparable<?>, VectorWritable> {
+public class NaiveBayesSumReducer
+ extends Reducer<WritableComparable<?>, VectorWritable, WritableComparable<?>, VectorWritable> {
@Override
protected void reduce(WritableComparable<?> key, Iterable<VectorWritable> values, Context context)
- throws IOException, InterruptedException {
+ throws IOException, InterruptedException {
Vector vector = null;
for (VectorWritable v : values) {
if (vector == null) {
Modified: mahout/trunk/core/src/main/java/org/apache/mahout/classifier/naivebayes/trainer/NaiveBayesThetaComplementaryMapper.java
URL: http://svn.apache.org/viewvc/mahout/trunk/core/src/main/java/org/apache/mahout/classifier/naivebayes/trainer/NaiveBayesThetaComplementaryMapper.java?rev=1072858&r1=1072857&r2=1072858&view=diff
==============================================================================
--- mahout/trunk/core/src/main/java/org/apache/mahout/classifier/naivebayes/trainer/NaiveBayesThetaComplementaryMapper.java (original)
+++ mahout/trunk/core/src/main/java/org/apache/mahout/classifier/naivebayes/trainer/NaiveBayesThetaComplementaryMapper.java Mon Feb 21 06:47:02 2011
@@ -44,7 +44,7 @@ public class NaiveBayesThetaComplementar
private Vector perLabelThetaNormalizer;
private double alphaI = 1.0;
private double vocabCount;
- private double totalSum = 0;
+ private double totalSum = 0.0;
@Override
protected void map(IntWritable key, VectorWritable value, Context context)
Modified: mahout/trunk/core/src/main/java/org/apache/mahout/classifier/naivebayes/trainer/NaiveBayesThetaMapper.java
URL: http://svn.apache.org/viewvc/mahout/trunk/core/src/main/java/org/apache/mahout/classifier/naivebayes/trainer/NaiveBayesThetaMapper.java?rev=1072858&r1=1072857&r2=1072858&view=diff
==============================================================================
--- mahout/trunk/core/src/main/java/org/apache/mahout/classifier/naivebayes/trainer/NaiveBayesThetaMapper.java (original)
+++ mahout/trunk/core/src/main/java/org/apache/mahout/classifier/naivebayes/trainer/NaiveBayesThetaMapper.java Mon Feb 21 06:47:02 2011
@@ -44,8 +44,7 @@ public class NaiveBayesThetaMapper exten
private double vocabCount;
@Override
- protected void map(IntWritable key, VectorWritable value, Context context)
- throws IOException, InterruptedException {
+ protected void map(IntWritable key, VectorWritable value, Context context) throws IOException, InterruptedException {
Vector vector = value.get();
int label = key.get();
double weight = Math.log((vector.zSum() + alphaI) / (labelSum.get(label) + vocabCount));
Modified: mahout/trunk/core/src/main/java/org/apache/mahout/classifier/naivebayes/trainer/NaiveBayesWeightsMapper.java
URL: http://svn.apache.org/viewvc/mahout/trunk/core/src/main/java/org/apache/mahout/classifier/naivebayes/trainer/NaiveBayesWeightsMapper.java?rev=1072858&r1=1072857&r2=1072858&view=diff
==============================================================================
--- mahout/trunk/core/src/main/java/org/apache/mahout/classifier/naivebayes/trainer/NaiveBayesWeightsMapper.java (original)
+++ mahout/trunk/core/src/main/java/org/apache/mahout/classifier/naivebayes/trainer/NaiveBayesWeightsMapper.java Mon Feb 21 06:47:02 2011
@@ -42,8 +42,7 @@ public class NaiveBayesWeightsMapper ext
private Vector labelSum;
@Override
- protected void map(IntWritable key, VectorWritable value, Context context)
- throws IOException, InterruptedException {
+ protected void map(IntWritable key, VectorWritable value, Context context) throws IOException, InterruptedException {
Vector vector = value.get();
if (featureSum == null) {
featureSum = new RandomAccessSparseVector(vector.size(), vector.getNumNondefaultElements());
Modified: mahout/trunk/core/src/main/java/org/apache/mahout/classifier/sequencelearning/hmm/HmmUtils.java
URL: http://svn.apache.org/viewvc/mahout/trunk/core/src/main/java/org/apache/mahout/classifier/sequencelearning/hmm/HmmUtils.java?rev=1072858&r1=1072857&r2=1072858&view=diff
==============================================================================
--- mahout/trunk/core/src/main/java/org/apache/mahout/classifier/sequencelearning/hmm/HmmUtils.java (original)
+++ mahout/trunk/core/src/main/java/org/apache/mahout/classifier/sequencelearning/hmm/HmmUtils.java Mon Feb 21 06:47:02 2011
@@ -183,9 +183,9 @@ public final class HmmUtils {
for (int i = 0; i < model.getEmissionMatrix().numRows(); i++) {
sum = 0;
for (int j = 0; j < model.getEmissionMatrix().numCols(); j++) {
- Preconditions.checkArgument(model.getEmissionMatrix().get(i, j) >= 0,
- "Error: The output state probability from hidden state " + i +
- " to output state " + j + " is negative");
+ Preconditions.checkArgument(
+ model.getEmissionMatrix().get(i, j) >= 0,
+ "The output state probability from hidden state " + i + " to output state " + j + " is negative");
sum += model.getEmissionMatrix().get(i, j);
}
Preconditions.checkArgument(Maths.approxEquals(sum, 1, 0.00001),
Modified: mahout/trunk/core/src/main/java/org/apache/mahout/classifier/sgd/AbstractOnlineLogisticRegression.java
URL: http://svn.apache.org/viewvc/mahout/trunk/core/src/main/java/org/apache/mahout/classifier/sgd/AbstractOnlineLogisticRegression.java?rev=1072858&r1=1072857&r2=1072858&view=diff
==============================================================================
--- mahout/trunk/core/src/main/java/org/apache/mahout/classifier/sgd/AbstractOnlineLogisticRegression.java (original)
+++ mahout/trunk/core/src/main/java/org/apache/mahout/classifier/sgd/AbstractOnlineLogisticRegression.java Mon Feb 21 06:47:02 2011
@@ -45,7 +45,7 @@ public abstract class AbstractOnlineLogi
// number of categories we are classifying. This should the number of rows of beta plus one.
protected int numCategories;
- protected int step = 0;
+ protected int step;
// information about how long since coefficient rows were updated. This allows lazy regularization.
protected Vector updateSteps;
@@ -59,7 +59,7 @@ public abstract class AbstractOnlineLogi
protected PriorFunction prior;
// can we ignore any further regularization when doing classification?
- private boolean sealed = false;
+ private boolean sealed;
// by default we don't do any fancy training
private Gradient gradient = new DefaultGradient();
@@ -100,13 +100,13 @@ public abstract class AbstractOnlineLogi
* @param r The value to transform.
* @return The logit of r.
*/
- public double link(double r){
- if (r < 0) {
+ public double link(double r) {
+ if (r < 0.0) {
double s = Math.exp(r);
- return s / (1 + s);
+ return s / (1.0 + s);
} else {
double s = Math.exp(-r);
- return 1 / (1 + s);
+ return 1.0 / (1.0 + s);
}
}
Modified: mahout/trunk/core/src/main/java/org/apache/mahout/classifier/sgd/CsvRecordFactory.java
URL: http://svn.apache.org/viewvc/mahout/trunk/core/src/main/java/org/apache/mahout/classifier/sgd/CsvRecordFactory.java?rev=1072858&r1=1072857&r2=1072858&view=diff
==============================================================================
--- mahout/trunk/core/src/main/java/org/apache/mahout/classifier/sgd/CsvRecordFactory.java (original)
+++ mahout/trunk/core/src/main/java/org/apache/mahout/classifier/sgd/CsvRecordFactory.java Mon Feb 21 06:47:02 2011
@@ -72,7 +72,7 @@ public class CsvRecordFactory implements
// commas inside. Also, escaped quotes will not be unescaped. Good enough for now.
private final Splitter onComma = Splitter.on(",").trimResults(CharMatcher.is('"'));
- private static final Map<String, Class<? extends FeatureVectorEncoder>> typeDictionary =
+ private static final Map<String, Class<? extends FeatureVectorEncoder>> TYPE_DICTIONARY =
ImmutableMap.<String, Class<? extends FeatureVectorEncoder>>builder()
.put("continuous", ContinuousValueEncoder.class)
.put("numeric", ContinuousValueEncoder.class)
@@ -116,9 +116,10 @@ public class CsvRecordFactory implements
*/
@Override
public void defineTargetCategories(List<String> values) {
- Preconditions.checkArgument(values.size() <= maxTargetValue,
- "Must have less than or equal to " + maxTargetValue +
- " categories for target variable, but found " + values.size());
+ Preconditions.checkArgument(
+ values.size() <= maxTargetValue,
+ "Must have less than or equal to " + maxTargetValue + " categories for target variable, but found "
+ + values.size());
if (maxTargetValue == Integer.MAX_VALUE) {
maxTargetValue = values.size();
}
@@ -189,11 +190,11 @@ public class CsvRecordFactory implements
c = ConstantValueEncoder.class;
} else {
name = variableNames.get(predictor);
- c = typeDictionary.get(typeMap.get(name));
+ c = TYPE_DICTIONARY.get(typeMap.get(name));
}
try {
Preconditions.checkArgument(c != null, "Invalid type of variable %s, wanted one of %s",
- typeMap.get(name), typeDictionary.keySet());
+ typeMap.get(name), TYPE_DICTIONARY.keySet());
Constructor<? extends FeatureVectorEncoder> constructor = c.getConstructor(String.class);
Preconditions.checkArgument(constructor != null, "Can't find correct constructor for %s", typeMap.get(name));
FeatureVectorEncoder encoder = constructor.newInstance(name);
Modified: mahout/trunk/core/src/main/java/org/apache/mahout/classifier/sgd/ModelDissector.java
URL: http://svn.apache.org/viewvc/mahout/trunk/core/src/main/java/org/apache/mahout/classifier/sgd/ModelDissector.java?rev=1072858&r1=1072857&r2=1072858&view=diff
==============================================================================
--- mahout/trunk/core/src/main/java/org/apache/mahout/classifier/sgd/ModelDissector.java (original)
+++ mahout/trunk/core/src/main/java/org/apache/mahout/classifier/sgd/ModelDissector.java Mon Feb 21 06:47:02 2011
@@ -117,7 +117,7 @@ public class ModelDissector {
return r;
}
- private static class Category implements Comparable<Category> {
+ private static final class Category implements Comparable<Category> {
private final int index;
private final double weight;
Modified: mahout/trunk/core/src/main/java/org/apache/mahout/classifier/sgd/PolymorphicWritable.java
URL: http://svn.apache.org/viewvc/mahout/trunk/core/src/main/java/org/apache/mahout/classifier/sgd/PolymorphicWritable.java?rev=1072858&r1=1072857&r2=1072858&view=diff
==============================================================================
--- mahout/trunk/core/src/main/java/org/apache/mahout/classifier/sgd/PolymorphicWritable.java (original)
+++ mahout/trunk/core/src/main/java/org/apache/mahout/classifier/sgd/PolymorphicWritable.java Mon Feb 21 06:47:02 2011
@@ -26,7 +26,11 @@ import java.io.IOException;
/**
* Utilities that write a class name and then serialize using writables.
*/
-public class PolymorphicWritable<T> {
+public final class PolymorphicWritable<T> {
+
+ private PolymorphicWritable() {
+ }
+
public static <T extends Writable> void write(DataOutput dataOutput, T value) throws IOException {
dataOutput.writeUTF(value.getClass().getName());
value.write(dataOutput);
Modified: mahout/trunk/core/src/main/java/org/apache/mahout/classifier/sgd/RankingGradient.java
URL: http://svn.apache.org/viewvc/mahout/trunk/core/src/main/java/org/apache/mahout/classifier/sgd/RankingGradient.java?rev=1072858&r1=1072857&r2=1072858&view=diff
==============================================================================
--- mahout/trunk/core/src/main/java/org/apache/mahout/classifier/sgd/RankingGradient.java (original)
+++ mahout/trunk/core/src/main/java/org/apache/mahout/classifier/sgd/RankingGradient.java Mon Feb 21 06:47:02 2011
@@ -35,7 +35,8 @@ import java.util.List;
* goals.
*/
public class RankingGradient implements Gradient {
- private static final Gradient basic = new DefaultGradient();
+
+ private static final Gradient BASIC = new DefaultGradient();
private int window = 10;
@@ -55,7 +56,7 @@ public class RankingGradient implements
Vector r = null;
for (Vector other : otherSide) {
- Vector g = basic.apply(groupKey, actual, instance.minus(other), classifier);
+ Vector g = BASIC.apply(groupKey, actual, instance.minus(other), classifier);
if (r == null) {
r = g;
@@ -79,6 +80,6 @@ public class RankingGradient implements
}
public Gradient getBaseGradient() {
- return basic;
+ return BASIC;
}
}
Modified: mahout/trunk/core/src/main/java/org/apache/mahout/clustering/OnlineGaussianAccumulator.java
URL: http://svn.apache.org/viewvc/mahout/trunk/core/src/main/java/org/apache/mahout/clustering/OnlineGaussianAccumulator.java?rev=1072858&r1=1072857&r2=1072858&view=diff
==============================================================================
--- mahout/trunk/core/src/main/java/org/apache/mahout/clustering/OnlineGaussianAccumulator.java (original)
+++ mahout/trunk/core/src/main/java/org/apache/mahout/clustering/OnlineGaussianAccumulator.java Mon Feb 21 06:47:02 2011
@@ -63,26 +63,25 @@ public class OnlineGaussianAccumulator i
* Variance = S / (sumweight-1) # if sample is the population, omit -1
* return Variance
*/
-
@Override
public void observe(Vector x, double weight) {
double temp = weight + sumWeight;
- Vector Q;
+ Vector q;
if (mean == null) {
mean = x.like();
- Q = x.clone();
+ q = x.clone();
} else {
- Q = x.minus(mean);
+ q = x.minus(mean);
}
- Vector R = Q.times(weight).divide(temp);
+ Vector r = q.times(weight).divide(temp);
if (s == null) {
- s = Q.times(sumWeight).times(R);
+ s = q.times(sumWeight).times(r);
} else {
- s = s.plus(Q.times(sumWeight).times(R));
+ s = s.plus(q.times(sumWeight).times(r));
}
- mean = mean.plus(R);
+ mean = mean.plus(r);
sumWeight = temp;
- variance = s.divide(sumWeight - 1);// # if sample is the population, omit -1
+ variance = s.divide(sumWeight - 1); // # if sample is the population, omit -1
}
@Override
Modified: mahout/trunk/core/src/main/java/org/apache/mahout/clustering/RunningSumsGaussianAccumulator.java
URL: http://svn.apache.org/viewvc/mahout/trunk/core/src/main/java/org/apache/mahout/clustering/RunningSumsGaussianAccumulator.java?rev=1072858&r1=1072857&r2=1072858&view=diff
==============================================================================
--- mahout/trunk/core/src/main/java/org/apache/mahout/clustering/RunningSumsGaussianAccumulator.java (original)
+++ mahout/trunk/core/src/main/java/org/apache/mahout/clustering/RunningSumsGaussianAccumulator.java Mon Feb 21 06:47:02 2011
@@ -26,7 +26,7 @@ import org.apache.mahout.math.function.S
*/
public class RunningSumsGaussianAccumulator implements GaussianAccumulator {
- private double s0 = 0.0;
+ private double s0;
private Vector s1;
private Vector s2;
private Vector mean;
Modified: mahout/trunk/core/src/main/java/org/apache/mahout/clustering/canopy/CanopyDriver.java
URL: http://svn.apache.org/viewvc/mahout/trunk/core/src/main/java/org/apache/mahout/clustering/canopy/CanopyDriver.java?rev=1072858&r1=1072857&r2=1072858&view=diff
==============================================================================
--- mahout/trunk/core/src/main/java/org/apache/mahout/clustering/canopy/CanopyDriver.java (original)
+++ mahout/trunk/core/src/main/java/org/apache/mahout/clustering/canopy/CanopyDriver.java Mon Feb 21 06:47:02 2011
@@ -84,7 +84,8 @@ public class CanopyDriver extends Abstra
double t1 = Double.parseDouble(getOption(DefaultOptionCreator.T1_OPTION));
double t2 = Double.parseDouble(getOption(DefaultOptionCreator.T2_OPTION));
boolean runClustering = hasOption(DefaultOptionCreator.CLUSTERING_OPTION);
- boolean runSequential = (getOption(DefaultOptionCreator.METHOD_OPTION).equalsIgnoreCase(DefaultOptionCreator.SEQUENTIAL_METHOD));
+ boolean runSequential = getOption(DefaultOptionCreator.METHOD_OPTION).equalsIgnoreCase(
+ DefaultOptionCreator.SEQUENTIAL_METHOD);
ClassLoader ccl = Thread.currentThread().getContextClassLoader();
DistanceMeasure measure = ccl.loadClass(measureClass).asSubclass(DistanceMeasure.class).newInstance();
@@ -158,9 +159,10 @@ public class CanopyDriver extends Abstra
DistanceMeasure measure,
double t1,
double t2,
- boolean runSequential) throws InstantiationException, IllegalAccessException, IOException,
- InterruptedException, ClassNotFoundException {
- log.info("Build Clusters Input: {} Out: {} " + "Measure: {} t1: {} t2: {}", new Object[] { input, output, measure, t1, t2 });
+ boolean runSequential)
+ throws InstantiationException, IllegalAccessException, IOException, InterruptedException, ClassNotFoundException {
+ log.info("Build Clusters Input: {} Out: {} " + "Measure: {} t1: {} t2: {}",
+ new Object[] { input, output, measure, t1, t2 });
if (runSequential) {
return buildClustersSeq(input, output, measure, t1, t2);
} else {
@@ -248,7 +250,7 @@ public class CanopyDriver extends Abstra
FileInputFormat.addInputPath(job, input);
Path canopyOutputDir = new Path(output, Cluster.CLUSTERS_DIR + '0');
FileOutputFormat.setOutputPath(job, canopyOutputDir);
- if (job.waitForCompletion(true) == false) {
+ if (!job.waitForCompletion(true)) {
throw new InterruptedException("Canopy Job failed processing " + input.toString());
}
return canopyOutputDir;
@@ -349,7 +351,7 @@ public class CanopyDriver extends Abstra
FileOutputFormat.setOutputPath(job, outPath);
HadoopUtil.overwriteOutput(outPath);
- if (job.waitForCompletion(true) == false) {
+ if (!job.waitForCompletion(true)) {
throw new InterruptedException("Canopy Clustering failed processing " + canopies.toString());
}
}
Modified: mahout/trunk/core/src/main/java/org/apache/mahout/clustering/dirichlet/DirichletDriver.java
URL: http://svn.apache.org/viewvc/mahout/trunk/core/src/main/java/org/apache/mahout/clustering/dirichlet/DirichletDriver.java?rev=1072858&r1=1072857&r2=1072858&view=diff
==============================================================================
--- mahout/trunk/core/src/main/java/org/apache/mahout/clustering/dirichlet/DirichletDriver.java (original)
+++ mahout/trunk/core/src/main/java/org/apache/mahout/clustering/dirichlet/DirichletDriver.java Mon Feb 21 06:47:02 2011
@@ -56,25 +56,13 @@ import org.slf4j.LoggerFactory;
public class DirichletDriver extends AbstractJob {
public static final String STATE_IN_KEY = "org.apache.mahout.clustering.dirichlet.stateIn";
-
public static final String MODEL_DISTRIBUTION_KEY = "org.apache.mahout.clustering.dirichlet.modelFactory";
-
- public static final String MODEL_PROTOTYPE_KEY = "org.apache.mahout.clustering.dirichlet.modelPrototype";
-
- public static final String PROTOTYPE_SIZE_KEY = "org.apache.mahout.clustering.dirichlet.prototypeSize";
-
public static final String NUM_CLUSTERS_KEY = "org.apache.mahout.clustering.dirichlet.numClusters";
-
public static final String ALPHA_0_KEY = "org.apache.mahout.clustering.dirichlet.alpha_0";
-
public static final String EMIT_MOST_LIKELY_KEY = "org.apache.mahout.clustering.dirichlet.emitMostLikely";
-
public static final String THRESHOLD_KEY = "org.apache.mahout.clustering.dirichlet.threshold";
-
public static final String MODEL_PROTOTYPE_CLASS_OPTION = "modelPrototype";
-
public static final String MODEL_DISTRIBUTION_CLASS_OPTION = "modelDist";
-
public static final String ALPHA_OPTION = "alpha";
private static final Logger log = LoggerFactory.getLogger(DirichletDriver.class);
@@ -371,7 +359,7 @@ public class DirichletDriver extends Abs
FileInputFormat.addInputPath(job, input);
FileOutputFormat.setOutputPath(job, stateOut);
- if (job.waitForCompletion(true) == false) {
+ if (!job.waitForCompletion(true)) {
throw new InterruptedException("Dirichlet Iteration failed processing " + stateIn.toString());
}
}
@@ -511,7 +499,7 @@ public class DirichletDriver extends Abs
}
private static void clusterDataSeq(Path input, Path stateIn, Path output, boolean emitMostLikely, double threshold)
- throws IOException, InstantiationException, IllegalAccessException {
+ throws IOException, InstantiationException, IllegalAccessException {
Configuration conf = new Configuration();
List<DirichletCluster> clusters = DirichletClusterMapper.loadClusters(conf, stateIn);
DirichletClusterer clusterer = new DirichletClusterer(emitMostLikely, threshold);
@@ -562,7 +550,7 @@ public class DirichletDriver extends Abs
FileInputFormat.addInputPath(job, input);
FileOutputFormat.setOutputPath(job, output);
- if (job.waitForCompletion(true) == false) {
+ if (!job.waitForCompletion(true)) {
throw new InterruptedException("Dirichlet Clustering failed processing " + stateIn.toString());
}
}
Modified: mahout/trunk/core/src/main/java/org/apache/mahout/clustering/dirichlet/UncommonDistributions.java
URL: http://svn.apache.org/viewvc/mahout/trunk/core/src/main/java/org/apache/mahout/clustering/dirichlet/UncommonDistributions.java?rev=1072858&r1=1072857&r2=1072858&view=diff
==============================================================================
--- mahout/trunk/core/src/main/java/org/apache/mahout/clustering/dirichlet/UncommonDistributions.java (original)
+++ mahout/trunk/core/src/main/java/org/apache/mahout/clustering/dirichlet/UncommonDistributions.java Mon Feb 21 06:47:02 2011
@@ -100,7 +100,7 @@ public final class UncommonDistributions
/**
* Returns a vector of random samples from a beta distribution with the given shapes
*
- * @param K
+ * @param k
* the number of samples to return
* @param shape1
* a double representing shape1
@@ -108,12 +108,12 @@ public final class UncommonDistributions
* a double representing shape2
* @return a Vector of samples
*/
- public static Vector rBeta(int K, double shape1, double shape2) {
+ public static Vector rBeta(int k, double shape1, double shape2) {
// List<Double> params = new ArrayList<Double>(2);
// params.add(shape1);
// params.add(Math.max(0, shape2));
- Vector result = new DenseVector(K);
- for (int i = 0; i < K; i++) {
+ Vector result = new DenseVector(k);
+ for (int i = 0; i < k; i++) {
result.set(i, rBeta(shape1, shape2));
}
return result;
@@ -174,11 +174,11 @@ public final class UncommonDistributions
double nextDouble = UncommonDistributions.RANDOM.nextDouble();
double p = nextDouble * total;
for (int i = 0; i < probabilities.size(); i++) {
- double p_i = probabilities.get(i);
- if (p < p_i) {
+ double pi = probabilities.get(i);
+ if (p < pi) {
return i;
} else {
- p -= p_i;
+ p -= pi;
}
}
// can't happen except for round-off error so we don't care what we return here
Modified: mahout/trunk/core/src/main/java/org/apache/mahout/clustering/fuzzykmeans/FuzzyKMeansClusterMapper.java
URL: http://svn.apache.org/viewvc/mahout/trunk/core/src/main/java/org/apache/mahout/clustering/fuzzykmeans/FuzzyKMeansClusterMapper.java?rev=1072858&r1=1072857&r2=1072858&view=diff
==============================================================================
--- mahout/trunk/core/src/main/java/org/apache/mahout/clustering/fuzzykmeans/FuzzyKMeansClusterMapper.java (original)
+++ mahout/trunk/core/src/main/java/org/apache/mahout/clustering/fuzzykmeans/FuzzyKMeansClusterMapper.java Mon Feb 21 06:47:02 2011
@@ -30,21 +30,23 @@ import org.apache.hadoop.mapreduce.Mappe
import org.apache.mahout.clustering.WeightedVectorWritable;
import org.apache.mahout.math.VectorWritable;
-public class FuzzyKMeansClusterMapper extends Mapper<WritableComparable<?>, VectorWritable, IntWritable, WeightedVectorWritable> {
+public class FuzzyKMeansClusterMapper
+ extends Mapper<WritableComparable<?>, VectorWritable, IntWritable, WeightedVectorWritable> {
private final List<SoftCluster> clusters = new ArrayList<SoftCluster>();
private FuzzyKMeansClusterer clusterer;
@Override
- protected void map(WritableComparable<?> key, VectorWritable point, Context context) throws IOException, InterruptedException {
+ protected void map(WritableComparable<?> key, VectorWritable point, Context context)
+ throws IOException, InterruptedException {
clusterer.emitPointToClusters(point, clusters, context);
}
@Override
protected void setup(Context context) throws IOException, InterruptedException {
- super.setup(context);
- Configuration conf = context.getConfiguration();
+ super.setup(context);
+ Configuration conf = context.getConfiguration();
clusterer = new FuzzyKMeansClusterer(conf);
String clusterPath = conf.get(FuzzyKMeansConfigKeys.CLUSTER_PATH_KEY);
Modified: mahout/trunk/core/src/main/java/org/apache/mahout/clustering/fuzzykmeans/FuzzyKMeansClusterer.java
URL: http://svn.apache.org/viewvc/mahout/trunk/core/src/main/java/org/apache/mahout/clustering/fuzzykmeans/FuzzyKMeansClusterer.java?rev=1072858&r1=1072857&r2=1072858&view=diff
==============================================================================
--- mahout/trunk/core/src/main/java/org/apache/mahout/clustering/fuzzykmeans/FuzzyKMeansClusterer.java (original)
+++ mahout/trunk/core/src/main/java/org/apache/mahout/clustering/fuzzykmeans/FuzzyKMeansClusterer.java Mon Feb 21 06:47:02 2011
@@ -40,13 +40,9 @@ public class FuzzyKMeansClusterer {
private static final double MINIMAL_VALUE = 0.0000000001;
private DistanceMeasure measure;
-
private double convergenceDelta;
-
private double m = 2.0; // default value
-
private boolean emitMostLikely = true;
-
private double threshold;
/**
@@ -63,7 +59,6 @@ public class FuzzyKMeansClusterer {
}
public FuzzyKMeansClusterer() {
- // TODO Auto-generated constructor stub
}
/**
@@ -163,7 +158,7 @@ public class FuzzyKMeansClusterer {
public void emitPointProbToCluster(Vector point,
List<SoftCluster> clusters,
Mapper<?,?,Text,ClusterObservations>.Context context)
- throws IOException, InterruptedException {
+ throws IOException, InterruptedException {
List<Double> clusterDistanceList = new ArrayList<Double>();
for (SoftCluster cluster : clusters) {
@@ -216,7 +211,7 @@ public class FuzzyKMeansClusterer {
public void emitPointToClusters(VectorWritable point,
List<SoftCluster> clusters,
Mapper<?,?,IntWritable,WeightedVectorWritable>.Context context)
- throws IOException, InterruptedException {
+ throws IOException, InterruptedException {
// calculate point distances for all clusters
List<Double> clusterDistanceList = new ArrayList<Double>();
for (SoftCluster cluster : clusters) {
@@ -247,9 +242,9 @@ public class FuzzyKMeansClusterer {
List<SoftCluster> clusters,
Vector pi,
Mapper<?,?,IntWritable,WeightedVectorWritable>.Context context)
- throws IOException, InterruptedException {
+ throws IOException, InterruptedException {
int clusterId = -1;
- double clusterPdf = 0;
+ double clusterPdf = 0.0;
for (int i = 0; i < clusters.size(); i++) {
// System.out.println("cluster-" + clusters.get(i).getId() + "@ " + ClusterBase.formatVector(center, null));
double pdf = pi.get(i);
@@ -269,7 +264,7 @@ public class FuzzyKMeansClusterer {
Collection<SoftCluster> clusters,
Vector pi,
Mapper<?,?,IntWritable,WeightedVectorWritable>.Context context)
- throws IOException, InterruptedException {
+ throws IOException, InterruptedException {
for (int i = 0; i < clusters.size(); i++) {
double pdf = pi.get(i);
if (pdf > threshold) {
@@ -317,7 +312,7 @@ public class FuzzyKMeansClusterer {
}
private void emitAllClusters(Vector point, Collection<SoftCluster> clusters, Vector pi, Writer writer)
- throws IOException {
+ throws IOException {
for (int i = 0; i < clusters.size(); i++) {
double pdf = pi.get(i);
if (pdf > threshold) {
@@ -328,9 +323,9 @@ public class FuzzyKMeansClusterer {
}
private static void emitMostLikelyCluster(Vector point, List<SoftCluster> clusters, Vector pi, Writer writer)
- throws IOException {
+ throws IOException {
int clusterId = -1;
- double clusterPdf = 0;
+ double clusterPdf = 0.0;
for (int i = 0; i < clusters.size(); i++) {
// System.out.println("cluster-" + clusters.get(i).getId() + "@ " + ClusterBase.formatVector(center, null));
double pdf = pi.get(i);
Modified: mahout/trunk/core/src/main/java/org/apache/mahout/clustering/fuzzykmeans/FuzzyKMeansDriver.java
URL: http://svn.apache.org/viewvc/mahout/trunk/core/src/main/java/org/apache/mahout/clustering/fuzzykmeans/FuzzyKMeansDriver.java?rev=1072858&r1=1072857&r2=1072858&view=diff
==============================================================================
--- mahout/trunk/core/src/main/java/org/apache/mahout/clustering/fuzzykmeans/FuzzyKMeansDriver.java (original)
+++ mahout/trunk/core/src/main/java/org/apache/mahout/clustering/fuzzykmeans/FuzzyKMeansDriver.java Mon Feb 21 06:47:02 2011
@@ -71,7 +71,8 @@ public class FuzzyKMeansDriver extends A
addOption(DefaultOptionCreator.distanceMeasureOption().create());
addOption(DefaultOptionCreator.clustersInOption()
.withDescription("The input centroids, as Vectors. Must be a SequenceFile of Writable, Cluster/Canopy. "
- + "If k is also specified, then a random set of vectors will be selected" + " and written out to this path first")
+ + "If k is also specified, then a random set of vectors will be selected"
+ + " and written out to this path first")
.create());
addOption(DefaultOptionCreator.numClustersOption()
.withDescription("The k in k-Means. If specified, then a random selection of k Vectors will be chosen"
@@ -113,7 +114,8 @@ public class FuzzyKMeansDriver extends A
.get(DefaultOptionCreator.NUM_CLUSTERS_OPTION)), measure);
}
boolean runClustering = hasOption(DefaultOptionCreator.CLUSTERING_OPTION);
- boolean runSequential = getOption(DefaultOptionCreator.METHOD_OPTION).equalsIgnoreCase(DefaultOptionCreator.SEQUENTIAL_METHOD);
+ boolean runSequential = getOption(DefaultOptionCreator.METHOD_OPTION).equalsIgnoreCase(
+ DefaultOptionCreator.SEQUENTIAL_METHOD);
run(getConf(),
input,
clusters,
@@ -191,7 +193,6 @@ public class FuzzyKMeansDriver extends A
/**
* Run the iteration using supplied arguments
- * @param conf TODO
* @param input
* the directory pathname for input points
* @param clustersIn
@@ -242,7 +243,7 @@ public class FuzzyKMeansDriver extends A
FileInputFormat.addInputPath(job, input);
FileOutputFormat.setOutputPath(job, clustersOut);
- if (job.waitForCompletion(true) == false) {
+ if (!job.waitForCompletion(true)) {
throw new InterruptedException("Fuzzy K-Means Iteration failed processing " + clustersIn.toString());
}
FileSystem fs = FileSystem.get(clustersOut.toUri(), conf);
@@ -284,9 +285,10 @@ public class FuzzyKMeansDriver extends A
boolean runClustering,
boolean emitMostLikely,
double threshold,
- boolean runSequential) throws IOException, ClassNotFoundException, InterruptedException,
- InstantiationException, IllegalAccessException {
- Path clustersOut = buildClusters(conf, input, clustersIn, output, measure, convergenceDelta, maxIterations, m, runSequential);
+ boolean runSequential)
+ throws IOException, ClassNotFoundException, InterruptedException, InstantiationException, IllegalAccessException {
+ Path clustersOut =
+ buildClusters(conf, input, clustersIn, output, measure, convergenceDelta, maxIterations, m, runSequential);
if (runClustering) {
log.info("Clustering ");
clusterData(input,
@@ -333,7 +335,8 @@ public class FuzzyKMeansDriver extends A
double convergenceDelta,
int maxIterations,
float m,
- boolean runSequential) throws IOException, InstantiationException, IllegalAccessException, InterruptedException, ClassNotFoundException {
+ boolean runSequential)
+ throws IOException, InstantiationException, IllegalAccessException, InterruptedException, ClassNotFoundException {
if (runSequential) {
return buildClustersSeq(input, clustersIn, output, measure, convergenceDelta, maxIterations, m);
} else {
@@ -451,8 +454,8 @@ public class FuzzyKMeansDriver extends A
float m,
boolean emitMostLikely,
double threshold,
- boolean runSequential) throws IOException, ClassNotFoundException, InterruptedException,
- InstantiationException, IllegalAccessException {
+ boolean runSequential)
+ throws IOException, ClassNotFoundException, InterruptedException, InstantiationException, IllegalAccessException {
if (runSequential) {
clusterDataSeq(input, clustersIn, output, measure, convergenceDelta, m);
} else {
@@ -531,7 +534,7 @@ public class FuzzyKMeansDriver extends A
job.setNumReduceTasks(0);
job.setJarByClass(FuzzyKMeansDriver.class);
- if (job.waitForCompletion(true) == false) {
+ if (!job.waitForCompletion(true)) {
throw new InterruptedException("Fuzzy K-Means Clustering failed processing " + clustersIn.toString());
}
}
@@ -557,7 +560,8 @@ public class FuzzyKMeansDriver extends A
}
};
- FileStatus[] matches = fs.listStatus(FileUtil.stat2Paths(fs.globStatus(clusterPath, clusterFileFilter)), clusterFileFilter);
+ FileStatus[] matches =
+ fs.listStatus(FileUtil.stat2Paths(fs.globStatus(clusterPath, clusterFileFilter)), clusterFileFilter);
for (FileStatus match : matches) {
result.add(fs.makeQualified(match.getPath()));
Modified: mahout/trunk/core/src/main/java/org/apache/mahout/clustering/kmeans/KMeansClusterer.java
URL: http://svn.apache.org/viewvc/mahout/trunk/core/src/main/java/org/apache/mahout/clustering/kmeans/KMeansClusterer.java?rev=1072858&r1=1072857&r2=1072858&view=diff
==============================================================================
--- mahout/trunk/core/src/main/java/org/apache/mahout/clustering/kmeans/KMeansClusterer.java (original)
+++ mahout/trunk/core/src/main/java/org/apache/mahout/clustering/kmeans/KMeansClusterer.java Mon Feb 21 06:47:02 2011
@@ -66,7 +66,7 @@ public class KMeansClusterer {
public void emitPointToNearestCluster(Vector point,
Iterable<Cluster> clusters,
Mapper<?,?,Text,ClusterObservations>.Context context)
- throws IOException, InterruptedException {
+ throws IOException, InterruptedException {
Cluster nearestCluster = null;
double nearestDistance = Double.MAX_VALUE;
for (Cluster cluster : clusters) {
@@ -118,7 +118,7 @@ public class KMeansClusterer {
public void outputPointWithClusterInfo(Vector vector,
Iterable<Cluster> clusters,
Mapper<?,?,IntWritable,WeightedVectorWritable>.Context context)
- throws IOException, InterruptedException {
+ throws IOException, InterruptedException {
AbstractCluster nearestCluster = null;
double nearestDistance = Double.MAX_VALUE;
for (AbstractCluster cluster : clusters) {
Modified: mahout/trunk/core/src/main/java/org/apache/mahout/clustering/kmeans/KMeansDriver.java
URL: http://svn.apache.org/viewvc/mahout/trunk/core/src/main/java/org/apache/mahout/clustering/kmeans/KMeansDriver.java?rev=1072858&r1=1072857&r2=1072858&view=diff
==============================================================================
--- mahout/trunk/core/src/main/java/org/apache/mahout/clustering/kmeans/KMeansDriver.java (original)
+++ mahout/trunk/core/src/main/java/org/apache/mahout/clustering/kmeans/KMeansDriver.java Mon Feb 21 06:47:02 2011
@@ -356,7 +356,7 @@ public class KMeansDriver extends Abstra
job.setJarByClass(KMeansDriver.class);
HadoopUtil.overwriteOutput(clustersOut);
- if (job.waitForCompletion(true) == false) {
+ if (!job.waitForCompletion(true)) {
throw new InterruptedException("K-Means Iteration failed processing " + clustersIn.toString());
}
FileSystem fs = FileSystem.get(clustersOut.toUri(), conf);
@@ -498,7 +498,7 @@ public class KMeansDriver extends Abstra
job.setNumReduceTasks(0);
job.setJarByClass(KMeansDriver.class);
- if (job.waitForCompletion(true) == false) {
+ if (!job.waitForCompletion(true)) {
throw new InterruptedException("K-Means Clustering failed processing " + clustersIn.toString());
}
}
Modified: mahout/trunk/core/src/main/java/org/apache/mahout/clustering/lda/LDADriver.java
URL: http://svn.apache.org/viewvc/mahout/trunk/core/src/main/java/org/apache/mahout/clustering/lda/LDADriver.java?rev=1072858&r1=1072857&r2=1072858&view=diff
==============================================================================
--- mahout/trunk/core/src/main/java/org/apache/mahout/clustering/lda/LDADriver.java (original)
+++ mahout/trunk/core/src/main/java/org/apache/mahout/clustering/lda/LDADriver.java Mon Feb 21 06:47:02 2011
@@ -51,23 +51,14 @@ import com.google.common.base.Preconditi
public final class LDADriver extends AbstractJob {
private static final String TOPIC_SMOOTHING_OPTION = "topicSmoothing";
-
private static final String NUM_WORDS_OPTION = "numWords";
-
private static final String NUM_TOPICS_OPTION = "numTopics";
-
static final String STATE_IN_KEY = "org.apache.mahout.clustering.lda.stateIn";
-
static final String NUM_TOPICS_KEY = "org.apache.mahout.clustering.lda.numTopics";
-
static final String NUM_WORDS_KEY = "org.apache.mahout.clustering.lda.numWords";
-
static final String TOPIC_SMOOTHING_KEY = "org.apache.mahout.clustering.lda.topicSmoothing";
-
static final int LOG_LIKELIHOOD_KEY = -2;
-
static final int TOPIC_SUM_KEY = -1;
-
static final double OVERALL_CONVERGENCE = 1.0E-5;
private static final Logger log = LoggerFactory.getLogger(LDADriver.class);
@@ -239,7 +230,6 @@ public final class LDADriver extends Abs
/**
* Run the job using supplied arguments
- * @param conf TODO
* @param input
* the directory pathname for input points
* @param stateIn
@@ -275,7 +265,7 @@ public final class LDADriver extends Abs
job.setInputFormatClass(SequenceFileInputFormat.class);
job.setJarByClass(LDADriver.class);
- if (job.waitForCompletion(true) == false) {
+ if (!job.waitForCompletion(true)) {
throw new InterruptedException("LDA Iteration failed processing " + stateIn.toString());
}
return findLL(stateOut, conf);