Remove Guava from Framework, replace the permutations() method. Move …

…logback dependency to tests and use slf4j in core. Refactor loggers and printing messages.
datumbox · Apr 22, 2015 · 8e2a967 · 8e2a967
1 parent c30deee
commit 8e2a967
Show file tree

Hide file tree

Showing 18 changed files with 94 additions and 68 deletions.
diff --git a/NOTICE b/NOTICE
@@ -4,16 +4,6 @@ Copyright (C) 2013 Vasilis Vryniotis <bbriniotis@datumbox.com>
 
 The following libraries are included in packaged versions of this project:
 
-* JUnit
- * COPYRIGHT: Copyright 2002 JUnit
- * LICENSE: http://www.opensource.org/licenses/cpl.php (Common Public License Version 1.0)
- * HOMEPAGE: http://www.junit.org/
-
-* ClasspathSuite
- * COPYRIGHT: Copyright 2006 Johannes Link
- * LICENSE: http://www.apache.org/licenses/LICENSE-2.0.txt (Apache License, Version 2.0)
- * HOMEPAGE: https://github.com/takari/takari-cpsuite
-
 * Apache Commons Lang
  * COPYRIGHT: Copyright 2001 The Apache Software Foundation
  * LICENSE: http://www.apache.org/licenses/LICENSE-2.0.txt (Apache License, Version 2.0)
@@ -29,10 +19,10 @@ The following libraries are included in packaged versions of this project:
  * LICENSE: http://www.apache.org/licenses/LICENSE-2.0.txt (Apache License, Version 2.0)
  * HOMEPAGE: http://commons.apache.org/proper/commons-csv/
 
-* Guava
- * COPYRIGHT: Copyright 2007 Google Inc.
- * LICENSE: http://www.apache.org/licenses/LICENSE-2.0.txt (Apache License, Version 2.0)
- * HOMEPAGE: http://code.google.com/p/guava-libraries/
+* SLF4J API
+ * COPYRIGHT: Copyright 2004 QOS.ch
+ * LICENSE: http://www.slf4j.org/license.html (MIT License)
+ * HOMEPAGE: http://www.slf4j.org/
 
 * LIBSVM
  * COPYRIGHT: Copyright 2000 Chih-Chung Chang and Chih-Jen Lin
@@ -49,8 +39,30 @@ The following libraries are included in packaged versions of this project:
  * LICENSE: http://www.apache.org/licenses/LICENSE-2.0.txt (Apache License, Version 2.0)
  * HOMEPAGE: http://www.mapdb.org/
 
+
+Code from the following software is included in this project:
+
+* Guava
+ * COPYRIGHT: Copyright 2007 Google Inc.
+ * LICENSE: http://www.apache.org/licenses/LICENSE-2.0.txt (Apache License, Version 2.0)
+ * HOMEPAGE: http://code.google.com/p/guava-libraries/
+
+
+The following libraries are required for the tests of this project:
+
+* JUnit
+ * COPYRIGHT: Copyright 2002 JUnit
+ * LICENSE: http://www.opensource.org/licenses/cpl.php (Common Public License Version 1.0)
+ * HOMEPAGE: http://www.junit.org/
+
+* ClasspathSuite
+ * COPYRIGHT: Copyright 2006 Johannes Link
+ * LICENSE: http://www.apache.org/licenses/LICENSE-2.0.txt (Apache License, Version 2.0)
+ * HOMEPAGE: https://github.com/takari/takari-cpsuite
+
 * Logback
  * COPYRIGHT: Copyright 1999 QOS.ch
  * LICENSE: http://logback.qos.ch/license.html (Eclipse Public License v1.0 / GNU Lesser General Public License version 2.1)
  * HOMEPAGE: http://logback.qos.ch/
 
+
diff --git a/pom.xml b/pom.xml
@@ -69,6 +69,7 @@
     </issueManagement>
 
     <properties>
+        <!-- Build Plugins -->
         <java-version>1.8</java-version>
         <maven-compiler-plugin-version>3.3</maven-compiler-plugin-version>
         <maven-javadoc-plugin-version>2.10.3</maven-javadoc-plugin-version>
@@ -79,17 +80,21 @@
         <license-maven-plugin-version>2.10</license-maven-plugin-version>
         <gpg-plugin-version>1.6</gpg-plugin-version>
 
-        <junit-version>4.12</junit-version>
-        <cpsuite-version>1.2.7</cpsuite-version>
+        <!-- Code Dependencies -->
         <commons-lang-version>3.4</commons-lang-version>
         <commons-math-version>3.5</commons-math-version>
         <commons-csv-version>1.1</commons-csv-version>
-        <guava-version>18.0</guava-version>
+        <slf4j-api-version>1.7.12</slf4j-api-version>
         <libsvm-version>3.18.1</libsvm-version>
         <lpsolve-version>5.5.2.0</lpsolve-version>
         <mapdb-version>1.0.7</mapdb-version>
+
+        <!-- Test Dependencies -->
+        <junit-version>4.12</junit-version>
+        <cpsuite-version>1.2.7</cpsuite-version>
         <logback-classic-version>1.1.3</logback-classic-version>
 
+        <!-- Configuration -->
         <gpg.keyname>7083A486</gpg.keyname>
         <buildNumber>${maven.build.timestamp}</buildNumber>
         <maven.build.timestamp.format>yyyyMMdd</maven.build.timestamp.format>
@@ -247,18 +252,6 @@
     </build>
 
     <dependencies>
-        <dependency>
-            <groupId>junit</groupId>
-            <artifactId>junit</artifactId>
-            <version>${junit-version}</version>
-            <scope>test</scope>
-        </dependency>
-        <dependency>
-            <groupId>io.takari.junit</groupId>
-            <artifactId>takari-cpsuite</artifactId>
-            <version>${cpsuite-version}</version>
-            <scope>test</scope>
-        </dependency>
         <dependency>
             <groupId>org.apache.commons</groupId>
             <artifactId>commons-lang3</artifactId>
@@ -275,9 +268,9 @@
             <version>${commons-csv-version}</version>
         </dependency>
         <dependency>
-            <groupId>com.google.guava</groupId>
-            <artifactId>guava</artifactId>
-            <version>${guava-version}</version>
+            <groupId>org.slf4j</groupId>
+            <artifactId>slf4j-api</artifactId>
+            <version>${slf4j-api-version}</version>
         </dependency>
         <dependency>
             <groupId>com.facebook.thirdparty</groupId>
@@ -294,10 +287,24 @@
             <artifactId>mapdb</artifactId>
             <version>${mapdb-version}</version>
         </dependency>
+
+        <dependency>
+            <groupId>junit</groupId>
+            <artifactId>junit</artifactId>
+            <version>${junit-version}</version>
+            <scope>test</scope>
+        </dependency>
+        <dependency>
+            <groupId>io.takari.junit</groupId>
+            <artifactId>takari-cpsuite</artifactId>
+            <version>${cpsuite-version}</version>
+            <scope>test</scope>
+        </dependency>
         <dependency>
             <groupId>ch.qos.logback</groupId>
             <artifactId>logback-classic</artifactId>
             <version>${logback-classic-version}</version>
+            <scope>test</scope>
         </dependency>
     </dependencies>
 

diff --git a/src/main/java/com/datumbox/common/dataobjects/Dataset.java b/src/main/java/com/datumbox/common/dataobjects/Dataset.java
@@ -57,7 +57,7 @@ public static Dataset parseTextFiles(Map<Object, URI> textFilesMap, TextExtracto
                 Object theClass = entry.getKey();
                 URI datasetURI = entry.getValue();
 
-                logger.info("Dataset Parsing " + theClass + " class");
+                logger.info("Dataset Parsing {} class", theClass);
 
                 try (final BufferedReader br = new BufferedReader(new InputStreamReader(new FileInputStream(new File(datasetURI)), "UTF8"))) {
                     for (String line; (line = br.readLine()) != null;) {
@@ -79,7 +79,7 @@ public static Dataset parseCSVFile(Reader reader, String yVariable, Map<String,
             logger.info("Parsing CSV file");
 
             if (!headerDataTypes.containsKey(yVariable)) {
-                logger.warn("WARNING: The file is missing the response variable column " + yVariable + ".");
+                logger.warn("WARNING: The file is missing the response variable column {}.", yVariable);
             }
 
             TypeInference.DataType yDataType = headerDataTypes.get(yVariable);
@@ -99,7 +99,7 @@ public static Dataset parseCSVFile(Reader reader, String yVariable, Map<String,
                 for (CSVRecord row : parser) {
 
                     if (!row.isConsistent()) {
-                        logger.warn("WARNING: Skipping row " + row.getRecordNumber() + " because its size does not match the header size.");
+                        logger.warn("WARNING: Skipping row {} because its size does not match the header size.", row.getRecordNumber());
                         continue;
                     }
 

diff --git a/src/main/java/com/datumbox/framework/machinelearning/classification/MaximumEntropy.java b/src/main/java/com/datumbox/framework/machinelearning/classification/MaximumEntropy.java
@@ -215,7 +215,7 @@ private void IIS(Dataset trainingData, Map<List<Object>, Double> EpFj_observed,
         DatabaseConnector dbc = knowledgeBase.getDbc();
         for(int iteration=0;iteration<totalIterations;++iteration) {
 
-            logger.debug("Iteration "+iteration);
+            logger.debug("Iteration {}", iteration);
 
             Map<List<Object>, Double> tmp_EpFj_model = dbc.getBigMap("tmp_EpFj_model", true);
             Collection<List<Object>> infiniteLambdaWeights = new ArrayList<>();

diff --git a/src/main/java/com/datumbox/framework/machinelearning/classification/OrdinalRegression.java b/src/main/java/com/datumbox/framework/machinelearning/classification/OrdinalRegression.java
@@ -218,7 +218,7 @@ protected void _fit(Dataset trainingData) {
         DatabaseConnector dbc = knowledgeBase.getDbc();
         for(int iteration=0;iteration<totalIterations;++iteration) {
 
-            logger.debug("Iteration "+iteration);
+            logger.debug("Iteration {}", iteration);
 
             Map<Object, Double> tmp_newThitas = dbc.getBigMap("tmp_newThitas", true);
 

diff --git a/src/main/java/com/datumbox/framework/machinelearning/classification/SoftMaxRegression.java b/src/main/java/com/datumbox/framework/machinelearning/classification/SoftMaxRegression.java
@@ -184,7 +184,7 @@ protected void _fit(Dataset trainingData) {
         DatabaseConnector dbc = knowledgeBase.getDbc();
         for(int iteration=0;iteration<totalIterations;++iteration) {
 
-            logger.debug("Iteration "+iteration);
+            logger.debug("Iteration {}", iteration);
 
             Map<List<Object>, Double> tmp_newThitas = dbc.getBigMap("tmp_newThitas", true);
 

diff --git a/src/main/java/com/datumbox/framework/machinelearning/clustering/Kmeans.java b/src/main/java/com/datumbox/framework/machinelearning/clustering/Kmeans.java
@@ -587,7 +587,7 @@ private void calculateClusters(Dataset trainingData) {
 
         AssociativeArray clusterDistances = new AssociativeArray();
         for(int iteration=0;iteration<maxIterations;++iteration) {
-            logger.debug("Iteration "+iteration);
+            logger.debug("Iteration {}", iteration);
 
             //reset cluster points
             for(Cluster c : clusterList.values()) {

diff --git a/src/main/java/com/datumbox/framework/machinelearning/common/bases/BaseTrainable.java b/src/main/java/com/datumbox/framework/machinelearning/common/bases/BaseTrainable.java
@@ -38,7 +38,7 @@
  */
 public abstract class BaseTrainable<MP extends BaseModelParameters, TP extends BaseTrainingParameters, KB extends KnowledgeBase<MP, TP>> implements Trainable<MP, TP> {
 
-    protected final Logger logger;
+    protected final Logger logger = LoggerFactory.getLogger(getClass());
 
     protected KB knowledgeBase;
     protected String dbName;
@@ -64,8 +64,6 @@ protected BaseTrainable(String dbName, DatabaseConfiguration dbConf) {
         }
 
         this.dbName = dbName;
-
-        logger = LoggerFactory.getLogger(this.getClass());
     }
 
     protected BaseTrainable(String dbName, DatabaseConfiguration dbConf, Class<MP> mpClass, Class<TP> tpClass) {

diff --git a/...a/com/datumbox/framework/machinelearning/common/bases/basemodels/BaseBoostingBagging.java b/...a/com/datumbox/framework/machinelearning/common/bases/basemodels/BaseBoostingBagging.java
@@ -206,7 +206,7 @@ protected void _fit(Dataset trainingData) {
         int t=0;
         int retryCounter = 0;
         while(t<totalWeakClassifiers) {
-            logger.debug("Training Weak learner "+t);
+            logger.debug("Training Weak learner {}", t);
             //We sample a list of Ids based on their weights
             FlatDataList sampledIDs = SRS.weightedSampling(observationWeights, n, true).toFlatDataList();
 

diff --git a/src/main/java/com/datumbox/framework/machinelearning/common/bases/basemodels/BaseDPMM.java b/src/main/java/com/datumbox/framework/machinelearning/common/bases/basemodels/BaseDPMM.java
@@ -280,7 +280,7 @@ private int collapsedGibbsSampling(Dataset dataset) {
         int iteration=0;
         while(iteration<maxIterations && noChangeMade==false) {
 
-            logger.debug("Iteration "+iteration);
+            logger.debug("Iteration {}", iteration);
 
             noChangeMade=true;
             for(Integer rId : dataset) {

diff --git a/...x/framework/machinelearning/common/bases/featureselection/ScoreBasedFeatureSelection.java b/...x/framework/machinelearning/common/bases/featureselection/ScoreBasedFeatureSelection.java
@@ -17,7 +17,7 @@
 
 import com.datumbox.common.persistentstorage.interfaces.DatabaseConfiguration;
 import com.datumbox.common.persistentstorage.interfaces.DatabaseConnector;
-import com.google.common.collect.Ordering;
+import com.datumbox.common.utilities.SelectKth;
 import java.util.Iterator;
 import java.util.Map;
 import org.slf4j.Logger;
@@ -56,8 +56,7 @@ public static void selectHighScoreFeatures(Map<Object, Double> featureScores, In
         logger.debug("selectHighScoreFeatures()");
 
         logger.debug("Estimating the minPermittedScore");
-        Double minPermittedScore=Ordering.<Double>natural().greatestOf(featureScores.values().iterator(), maxFeatures).get(maxFeatures-1);
-        //Double minPermittedScore = SelectKth.largest(featureScores.values().iterator(), maxFeatures);
+        Double minPermittedScore = SelectKth.largest(featureScores.values().iterator(), maxFeatures);
 
         //remove any entry with score less than the minimum permitted one
         logger.debug("Removing features with scores less than threshold");

diff --git a/.../java/com/datumbox/framework/machinelearning/common/bases/validation/ModelValidation.java b/.../java/com/datumbox/framework/machinelearning/common/bases/validation/ModelValidation.java
@@ -36,12 +36,12 @@
  */
 public abstract class ModelValidation<MP extends BaseMLmodel.ModelParameters, TP extends BaseMLmodel.TrainingParameters, VM extends BaseMLmodel.ValidationMetrics> {
 
-    protected final Logger logger;
+    protected final Logger logger = LoggerFactory.getLogger(getClass());
 
     public static final String DB_INDICATOR="Kfold";
 
     public ModelValidation() {
-        logger = LoggerFactory.getLogger(this.getClass());
+
     }
 
     public VM kFoldCrossValidation(Dataset dataset, int k, String dbName, DatabaseConfiguration dbConf, Class<? extends BaseMLmodel> aClass, TP trainingParameters) {
@@ -69,7 +69,7 @@ public VM kFoldCrossValidation(Dataset dataset, int k, String dbName, DatabaseCo
         List<VM> validationMetricsList = new LinkedList<>();
         for(int fold=0;fold<k;++fold) {
 
-            logger.info("Kfold "+fold);
+            logger.info("Kfold {}", fold);
 
             //as fold window we consider the part of the ids that are used for validation
             FlatDataList foldTrainingIds = new FlatDataList(new ArrayList<>(n-foldSize));

diff --git a/src/main/java/com/datumbox/framework/machinelearning/featureselection/scorebased/TFIDF.java b/src/main/java/com/datumbox/framework/machinelearning/featureselection/scorebased/TFIDF.java
@@ -23,7 +23,6 @@
 import com.datumbox.common.dataobjects.TypeInference;
 
 import com.datumbox.framework.machinelearning.common.bases.featureselection.ScoreBasedFeatureSelection;
-import java.util.Iterator;
 import java.util.Map;
 
 

diff --git a/src/main/java/com/datumbox/framework/machinelearning/regression/NLMS.java b/src/main/java/com/datumbox/framework/machinelearning/regression/NLMS.java
@@ -109,7 +109,7 @@ protected void _fit(Dataset trainingData) {
         DatabaseConnector dbc = knowledgeBase.getDbc();
         for(int iteration=0;iteration<totalIterations;++iteration) {
 
-            logger.debug("Iteration "+iteration);
+            logger.debug("Iteration {}", iteration);
 
             Map<Object, Double> tmp_newThitas = dbc.getBigMap("tmp_newThitas", true);
 

diff --git a/.../java/com/datumbox/framework/machinelearning/topicmodeling/LatentDirichletAllocation.java b/.../java/com/datumbox/framework/machinelearning/topicmodeling/LatentDirichletAllocation.java
@@ -306,7 +306,7 @@ protected void _fit(Dataset trainingData) {
         int iteration=0;
         while(iteration<maxIterations) {
 
-            logger.debug("Iteration "+iteration);
+            logger.debug("Iteration {}", iteration);
 
             int changedCounter = 0;
             //collapsed gibbs sampler
@@ -387,7 +387,7 @@ protected void _fit(Dataset trainingData) {
             }
             ++iteration;
 
-            logger.debug("Reassigned Records "+ changedCounter);
+            logger.debug("Reassigned Records {}", changedCounter);
 
             if(changedCounter==0) {
                 break;
@@ -533,7 +533,7 @@ private ValidationMetrics predictAndValidate(Dataset newData) {
         double perplexity = Double.MAX_VALUE;
         for(int iteration=0;iteration<maxIterations;++iteration) {
 
-            logger.debug("Iteration "+iteration);
+            logger.debug("Iteration {}", iteration);
 
 
             //collapsed gibbs sampler
@@ -632,7 +632,7 @@ private ValidationMetrics predictAndValidate(Dataset newData) {
 
             perplexity=Math.exp(-perplexity/totalDatasetWords);
 
-            logger.debug("Reassigned Records "+ changedCounter +" - Perplexity: "+perplexity);
+            logger.debug("Reassigned Records {} - Perplexity: {}", changedCounter, perplexity);
 
             if(changedCounter==0) {
                 break;