newInstance(), constructors etc of inherited from BaseTrainable pass …

…to BaseTrainable and reduce the extra code.
datumbox · Mar 28, 2015 · 7de9734 · 7de9734
1 parent 870225f
commit 7de9734
Show file tree

Hide file tree

Showing 9 changed files with 32 additions and 136 deletions.
diff --git a/TODO.txt b/TODO.txt
@@ -9,7 +9,6 @@ NEW ALGORITHMS
 CODE IMPROVEMENT
 ================
 
-- newInstance(), constructors etc of inherited from BaseTrainable could perhaps pass to BaseTrainable and reduce the extra code.
 - Make this a method DATA_SAFE_CALL_BY_REFERENCE and IS_BINARIZED?
 - should we pass validator class in the constructors of all MLmodels?
 - Test a call directly on predict() and modify code to return the correct exception

diff --git a/src/main/java/com/datumbox/applications/datamodeling/Modeler.java b/src/main/java/com/datumbox/applications/datamodeling/Modeler.java
@@ -19,6 +19,7 @@
 import com.datumbox.common.dataobjects.Dataset;
 import com.datumbox.common.persistentstorage.interfaces.DatabaseConfiguration;
 import com.datumbox.common.persistentstorage.interfaces.DatabaseConnector;
+import com.datumbox.framework.machinelearning.common.bases.BaseTrainable;
 import com.datumbox.framework.machinelearning.common.bases.featureselection.FeatureSelection;
 import com.datumbox.framework.machinelearning.common.bases.mlmodels.BaseMLmodel;
 import com.datumbox.framework.machinelearning.common.bases.wrappers.BaseWrapper;
@@ -75,7 +76,7 @@ public void _fit(Dataset trainingData) {
 
         boolean transformData = (dtClass!=null);
         if(transformData) {
-            dataTransformer = DataTransformer.newInstance(dtClass, dbName, dbConf);
+            dataTransformer = DataTransformer.<DataTransformer>newInstance(dtClass, dbName, dbConf);
             dataTransformer.fit_transform(trainingData, trainingParameters.getDataTransformerTrainingParameters());
         }
 
@@ -85,7 +86,7 @@ public void _fit(Dataset trainingData) {
 
         boolean selectFeatures = (fsClass!=null);
         if(selectFeatures) {
-            featureSelection = FeatureSelection.newInstance(fsClass, dbName, dbConf);
+            featureSelection = FeatureSelection.<FeatureSelection>newInstance(fsClass, dbName, dbConf);
             featureSelection.fit(trainingData, trainingParameters.getFeatureSelectionTrainingParameters()); 
 
             featureSelection.transform(trainingData);
@@ -95,7 +96,8 @@ public void _fit(Dataset trainingData) {
 
 
         //initialize mlmodel
-        mlmodel = BaseMLmodel.newInstance(trainingParameters.getMLmodelClass(), dbName, dbConf); 
+        Class mlClass = trainingParameters.getMLmodelClass();
+        mlmodel = BaseMLmodel.<BaseMLmodel>newInstance(mlClass, dbName, dbConf); 
         int k = trainingParameters.getkFolds();
 
         //call k-fold cross validation and get the average accuracy
@@ -142,7 +144,7 @@ private BaseMLmodel.ValidationMetrics evaluateData(Dataset data, boolean estimat
         boolean transformData = (dtClass!=null);
         if(transformData) {
             if(dataTransformer==null) {
-                dataTransformer = DataTransformer.newInstance(dtClass, dbName, dbConf);
+                dataTransformer = DataTransformer.<DataTransformer>newInstance(dtClass, dbName, dbConf);
             }        
             dataTransformer.transform(data);
         }
@@ -152,7 +154,7 @@ private BaseMLmodel.ValidationMetrics evaluateData(Dataset data, boolean estimat
         boolean selectFeatures = (fsClass!=null);
         if(selectFeatures) {
             if(featureSelection==null) {
-                featureSelection = FeatureSelection.newInstance(fsClass, dbName, dbConf);
+                featureSelection = FeatureSelection.<FeatureSelection>newInstance(fsClass, dbName, dbConf);
             }
 
             //remove unnecessary features
@@ -162,7 +164,8 @@ private BaseMLmodel.ValidationMetrics evaluateData(Dataset data, boolean estimat
 
         //initialize mlmodel
         if(mlmodel==null) {
-            mlmodel = BaseMLmodel.newInstance(trainingParameters.getMLmodelClass(), dbName, dbConf); 
+            Class mlClass = trainingParameters.getMLmodelClass();
+            mlmodel = BaseMLmodel.<BaseMLmodel>newInstance(mlClass, dbName, dbConf); 
         }
 
         //call predict of the mlmodel for the new dataset

diff --git a/src/main/java/com/datumbox/applications/nlp/TextClassifier.java b/src/main/java/com/datumbox/applications/nlp/TextClassifier.java
@@ -21,7 +21,7 @@
 import com.datumbox.common.dataobjects.Record;
 import com.datumbox.common.persistentstorage.interfaces.DatabaseConfiguration;
 import com.datumbox.common.persistentstorage.interfaces.DatabaseConnector;
-import com.datumbox.configuration.GeneralConfiguration;
+import com.datumbox.framework.machinelearning.common.bases.BaseTrainable;
 import com.datumbox.framework.machinelearning.common.bases.featureselection.CategoricalFeatureSelection;
 import com.datumbox.framework.machinelearning.common.bases.featureselection.FeatureSelection;
 import com.datumbox.framework.machinelearning.common.bases.mlmodels.BaseMLmodel;
@@ -133,15 +133,15 @@ protected void _fit(Dataset trainingDataset) {
 
         boolean transformData = (dtClass!=null);
         if(transformData) {
-            dataTransformer = DataTransformer.newInstance(dtClass, dbName, dbConf);
+            dataTransformer = DataTransformer.<DataTransformer>newInstance(dtClass, dbName, dbConf);
             dataTransformer.fit_transform(trainingDataset, trainingParameters.getDataTransformerTrainingParameters());
         }
 
         Class fsClass = trainingParameters.getFeatureSelectionClass();
 
         boolean selectFeatures = (fsClass!=null);
         if(selectFeatures) {
-            featureSelection = FeatureSelection.newInstance(fsClass, dbName, dbConf);
+            featureSelection = FeatureSelection.<FeatureSelection>newInstance(fsClass, dbName, dbConf);
             FeatureSelection.TrainingParameters featureSelectionParameters = trainingParameters.getFeatureSelectionTrainingParameters();
             if(CategoricalFeatureSelection.TrainingParameters.class.isAssignableFrom(featureSelectionParameters.getClass())) {
                 ((CategoricalFeatureSelection.TrainingParameters)featureSelectionParameters).setIgnoringNumericalFeatures(false); //this should be turned off in feature selection
@@ -251,7 +251,7 @@ public BaseMLmodel.ValidationMetrics validate(Map<Object, URI> dataset) {
         boolean transformData = (dtClass!=null);
         if(transformData) {
             if(dataTransformer==null) {
-                dataTransformer = DataTransformer.newInstance(dtClass, dbName, dbConf);
+                dataTransformer = DataTransformer.<DataTransformer>newInstance(dtClass, dbName, dbConf);
             }        
 
             dataTransformer.transform(testDataset);
@@ -262,7 +262,7 @@ public BaseMLmodel.ValidationMetrics validate(Map<Object, URI> dataset) {
         boolean selectFeatures = (fsClass!=null);
         if(selectFeatures) {
             if(featureSelection==null) {
-                featureSelection = FeatureSelection.newInstance(fsClass, dbName, dbConf);
+                featureSelection = FeatureSelection.<FeatureSelection>newInstance(fsClass, dbName, dbConf);
             }
 
             //remove unnecessary features
@@ -324,7 +324,7 @@ private Dataset getPredictions(List<String> text) {
         boolean transformData = (dtClass!=null);
         if(transformData) {
             if(dataTransformer==null) {
-                dataTransformer = DataTransformer.newInstance(dtClass, dbName, dbConf);
+                dataTransformer = DataTransformer.<DataTransformer>newInstance(dtClass, dbName, dbConf);
             }        
             dataTransformer.transform(newData);
         }
@@ -334,7 +334,7 @@ private Dataset getPredictions(List<String> text) {
         boolean selectFeatures = (fsClass!=null);
         if(selectFeatures) {
             if(featureSelection==null) {
-                featureSelection = FeatureSelection.newInstance(fsClass, dbName, dbConf);
+                featureSelection = FeatureSelection.<FeatureSelection>newInstance(fsClass, dbName, dbConf);
             }
 
             //remove unnecessary features
@@ -344,7 +344,8 @@ private Dataset getPredictions(List<String> text) {
 
         //initialize mlmodel
         if(mlmodel==null) {
-            mlmodel = BaseMLmodel.newInstance(trainingParameters.getMLmodelClass(), dbName, dbConf); 
+            Class mlClass = trainingParameters.getMLmodelClass();
+            mlmodel = BaseMLmodel.<BaseMLmodel>newInstance(mlClass, dbName, dbConf); 
         }
 
         //call predict of the mlmodel for the new dataset

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
@@ -23,6 +23,7 @@
 import com.datumbox.framework.machinelearning.common.bases.dataobjects.BaseModelParameters;
 import com.datumbox.framework.machinelearning.common.bases.dataobjects.BaseTrainingParameters;
 import com.datumbox.framework.machinelearning.common.dataobjects.KnowledgeBase;
+import java.lang.reflect.InvocationTargetException;
 
 /**
  *
@@ -37,6 +38,19 @@ public abstract class BaseTrainable<MP extends BaseModelParameters, TP extends B
 
     protected KB knowledgeBase;
     protected String dbName;
+
+
+    public static <BT extends BaseTrainable> BT newInstance(Class<BT> aClass, String dbName, DatabaseConfiguration dbConfig) {
+        BT algorithm = null;
+        try {
+            algorithm = aClass.getConstructor(String.class, DatabaseConfiguration.class).newInstance(dbName, dbConfig);;
+        } 
+        catch (InstantiationException | IllegalAccessException | IllegalArgumentException | InvocationTargetException | NoSuchMethodException | SecurityException ex) {
+            throw new RuntimeException(ex);
+        }
+
+        return algorithm;
+    }
 
     protected BaseTrainable(String dbName, DatabaseConfiguration dbConf) {
         String methodName = this.getClass().getSimpleName();

diff --git a/...m/datumbox/framework/machinelearning/common/bases/datatransformation/DataTransformer.java b/...m/datumbox/framework/machinelearning/common/bases/datatransformation/DataTransformer.java
@@ -24,7 +24,6 @@
 import com.datumbox.framework.machinelearning.common.bases.dataobjects.BaseModelParameters;
 import com.datumbox.framework.machinelearning.common.bases.dataobjects.BaseTrainingParameters;
 import com.datumbox.framework.machinelearning.common.dataobjects.KnowledgeBase;
-import java.lang.reflect.InvocationTargetException;
 
 /**
  *
@@ -50,28 +49,6 @@ public static abstract class TrainingParameters extends BaseTrainingParameters {
     }
 
 
-    /**
-     * Generates a new instance of a MLmodel by providing the dbName and 
- the Class of the algorithm.
-     * 
-     * @param <D>
-     * @param dbName
-     * @param aClass
-     * @param dbConfig
-     * @return 
-     */
-    public static <D extends DataTransformer> D newInstance(Class<D> aClass, String dbName, DatabaseConfiguration dbConfig) {
-        D algorithm = null;
-        try {
-            algorithm = (D) aClass.getConstructor(String.class, DatabaseConfiguration.class).newInstance(dbName, dbConfig);
-        } 
-        catch (InstantiationException | IllegalAccessException | IllegalArgumentException | InvocationTargetException | NoSuchMethodException | SecurityException ex) {
-            throw new RuntimeException(ex);
-        }
-
-        return algorithm;
-    }
-
 
     /*
         IMPORTANT METHODS FOR THE FUNCTIONALITY

diff --git a/...om/datumbox/framework/machinelearning/common/bases/featureselection/FeatureSelection.java b/...om/datumbox/framework/machinelearning/common/bases/featureselection/FeatureSelection.java
@@ -24,7 +24,6 @@
 import com.datumbox.framework.machinelearning.common.bases.dataobjects.BaseModelParameters;
 import com.datumbox.framework.machinelearning.common.bases.dataobjects.BaseTrainingParameters;
 import com.datumbox.framework.machinelearning.common.dataobjects.KnowledgeBase;
-import java.lang.reflect.InvocationTargetException;
 
 /**
  *
@@ -48,30 +47,6 @@ public static abstract class TrainingParameters extends BaseTrainingParameters {
         //here goes public fields that are used as initial training parameters        
     }
 
-
-    /**
-     * Generates a new instance of a MLmodel by providing the dbName and 
- the Class of the algorithm.
-     * 
-     * @param <F>
-     * @param dbName
-     * @param aClass
-     * @param dbConfig
-     * @return 
-     */
-    public static <F extends FeatureSelection> F newInstance(Class<F> aClass, String dbName, DatabaseConfiguration dbConfig) {
-        F algorithm = null;
-        try {
-            algorithm = (F) aClass.getConstructor(String.class, DatabaseConfiguration.class).newInstance(dbName, dbConfig);
-        } 
-        catch (InstantiationException | IllegalAccessException | IllegalArgumentException | InvocationTargetException | NoSuchMethodException | SecurityException ex) {
-            throw new RuntimeException(ex);
-        }
-
-        return algorithm;
-    }
-
-
     /*
         IMPORTANT METHODS FOR THE FUNCTIONALITY
     */

diff --git a/src/main/java/com/datumbox/framework/machinelearning/common/bases/mlmodels/BaseMLmodel.java b/src/main/java/com/datumbox/framework/machinelearning/common/bases/mlmodels/BaseMLmodel.java
@@ -26,7 +26,6 @@
 import com.datumbox.framework.machinelearning.common.bases.dataobjects.BaseTrainingParameters;
 import com.datumbox.framework.machinelearning.common.bases.dataobjects.BaseValidationMetrics;
 import com.datumbox.framework.machinelearning.common.dataobjects.MLmodelKnowledgeBase;
-import java.lang.reflect.InvocationTargetException;
 
 /**
  * Abstract Class for a Machine Learning algorithm.
@@ -70,36 +69,6 @@ public static abstract class ValidationMetrics extends BaseValidationMetrics {
 
 
 
-
-
-    /**
-     * Generates a new instance of a BaseMLmodel by providing the dbName and 
- the Class of the algorithm.
-     * 
-     * @param <M>
-     * @param dbName
-     * @param aClass
-     * @param dbConfig
-     * @return 
-     */
-    public static <M extends BaseMLmodel> M newInstance(Class<M> aClass, String dbName, DatabaseConfiguration dbConfig) {
-        M algorithm = null;
-        try {
-            algorithm = (M) aClass.getConstructor(String.class, DatabaseConfiguration.class).newInstance(dbName, dbConfig);
-        } 
-        catch (InstantiationException | IllegalAccessException | IllegalArgumentException | InvocationTargetException | NoSuchMethodException | SecurityException ex) {
-            throw new RuntimeException(ex);
-        }
-
-        return algorithm;
-    }
-
-
-
-
-
-
-
     /*
         IMPORTANT METHODS FOR THE FUNCTIONALITY
     */

diff --git a/.../java/com/datumbox/framework/machinelearning/common/bases/mlmodels/BaseMLrecommender.java b/.../java/com/datumbox/framework/machinelearning/common/bases/mlmodels/BaseMLrecommender.java
@@ -57,35 +57,6 @@ public static abstract class TrainingParameters extends BaseTrainingParameters {
 
 
 
-
-
-    /**
-     * Generates a new instance of a BaseMLrecommender by providing the dbName and 
- the Class of the algorithm.
-     * 
-     * @param <M>
-     * @param dbName
-     * @param aClass
-     * @return 
-     */
-    public static <M extends BaseMLrecommender> M newInstance(Class<M> aClass, String dbName, DatabaseConfiguration dbConfig) {
-        M algorithm = null;
-        try {
-            algorithm = (M) aClass.getConstructor(String.class, DatabaseConfiguration.class).newInstance(dbName, dbConfig);;
-        } 
-        catch (InstantiationException | IllegalAccessException | IllegalArgumentException | InvocationTargetException | NoSuchMethodException | SecurityException ex) {
-            throw new RuntimeException(ex);
-        }
-
-        return algorithm;
-    }
-
-
-
-
-
-
-
     /*
         IMPORTANT METHODS FOR THE FUNCTIONALITY
     */

diff --git a/src/main/java/com/datumbox/framework/machinelearning/common/bases/wrappers/BaseWrapper.java b/src/main/java/com/datumbox/framework/machinelearning/common/bases/wrappers/BaseWrapper.java
@@ -25,7 +25,6 @@
 import com.datumbox.framework.machinelearning.common.dataobjects.KnowledgeBase;
 import com.datumbox.framework.machinelearning.common.bases.datatransformation.DataTransformer;
 import com.datumbox.framework.machinelearning.common.bases.featureselection.FeatureSelection;
-import java.lang.reflect.InvocationTargetException;
 
 /**
  * The BaseWrapper is a trainable object that uses composition instead of inheritance
@@ -128,18 +127,6 @@ public void setMLmodelTrainingParameters(ML.TrainingParameters mlmodelTrainingPa
     }
 
 
-    public static <W extends BaseWrapper> W newInstance(Class<W> aClass, String dbName, DatabaseConfiguration dbConfig) {
-        W algorithm = null;
-        try {
-            algorithm = (W) aClass.getConstructor(String.class, DatabaseConfiguration.class).newInstance(dbName, dbConfig);;
-        } 
-        catch (InstantiationException | IllegalAccessException | IllegalArgumentException | InvocationTargetException | NoSuchMethodException | SecurityException ex) {
-            throw new RuntimeException(ex);
-        }
-
-        return algorithm;
-    }
-
 
     /*
         IMPORTANT METHODS FOR THE FUNCTIONALITY