Changeset 34798


Ignore:
Timestamp:
2021-02-02T22:33:22+13:00 (3 years ago)
Author:
davidb
Message:

Removal of commented out code

Location:
main/trunk/model-sites-dev/mars/src/java/org/greenstone/mars
Files:
4 edited

Legend:

Unmodified
Added
Removed

  • main/trunk/model-sites-dev/mars/src/java/org/greenstone/mars/WekaApplyArousalModel.java

    r34797 r34798  
    11package org.greenstone.mars;
    2 
    3 //import java.util.Random;
    42
    53import java.io.BufferedInputStream;
    … …  
    2826class WekaApplyArousalModel
    2927{
    30 
    31     //public final static String PREDICT_ATTRIBUTE_NAME = WekaUtil.AROUSAL_ATTRIBUTE_NAME;
    32    
    33 
    34    
    3528    public static void main(String[] args)
    3629    {
    37     /*
    38     if (args.length != 3) {
    39         System.err.println("Error: incorrect number of command-line arguments");
    40         System.err.println("Usage: trained-model.{model|ser} unclassified-data.{arff|csv} classified-data.{arff|csv}");
    41         System.exit(1);
    42     }
    43     */
    44 
    4530    WekaUtil.checkUsageApplyModel(args);
    4631   
    … …  
    4934    String classified_data_output_filename  = args[2];
    5035
    51 
    5236    Classifier classifier = WekaUtil.loadClassifierModel(classifier_input_filename);
    53 
    54     /*
    55         System.out.println("Loading Weka saved Classifier: " + classifier_input_filename);
    56 
    57     try {
    58         FileInputStream fis = new FileInputStream(classifier_input_filename);   
    59         BufferedInputStream bis= new BufferedInputStream(fis);
    60         Classifier classifier = (Classifier)SerializationHelper.read(bis);
    61 
    62         //
    63         // Load in unlabeled data
    64         //
    65        
    66         System.out.println("Loading unlabeled instances: " + unclassified_data_input_filename);
    67 
    68         DataSource data_source = new DataSource(unclassified_data_input_filename);
    69         Instances unlabeled_instances = data_source.getDataSet();
    70 
    71         // Work out if we're dealing with a ground-truth ARFF file or not
    72         // (i.e. already has the desired attribute)
    73 
    74         Instances groundtruth_instances = null;     
    75         Attribute predict_attribute = unlabeled_instances.attribute(PREDICT_ATTRIBUTE_NAME);
    76        
    77         if (predict_attribute == null) {
    78 
    79         unlabeled_instances = WekaUtil.applyFilter(unlabeled_instances,null); // no additional top-up to remove     
    80         int num_attributes = unlabeled_instances.numAttributes();
    81 
    82         Attribute arousal_attribute = new Attribute(WekaUtil.AROUSAL_ATTRIBUTE_NAME);
    83         unlabeled_instances.insertAttributeAt(arousal_attribute,num_attributes);
    84         num_attributes++;       
    85         }
    86         else {
    87         // Dealing with ground-truth data:
    88         //   => already has 'arousal' attribute
    89         //   => in fact has 'valence' attribute too, which we want to remove
    90        
    91         unlabeled_instances = WekaUtil.applyFilter(unlabeled_instances,"472"); // top-up with removal of 'valence'
    92        
    93         // reference share this as 'groundtruth_instances' to trigger error calculation and output
    94         groundtruth_instances = unlabeled_instances;
    95         }
    96 
    97         int num_attributes = unlabeled_instances.numAttributes();
    98        
    99         // Set class attribute
    100         unlabeled_instances.setClassIndex(num_attributes - 1);
    101        
    102         WekaUtil.checkDatasetInstancesCompatible(unlabeled_instances);
    103 
    104 
    105 
    106 
    107        
    108         // Create copy where the predictions are to be made
    109         Instances labeled_instances = new Instances(unlabeled_instances);
    110 
    111         //
    112         // Label instances
    113         //
    114        
    115         final int num_instances = unlabeled_instances.numInstances();       
    116         for (int i=0; i<num_instances; i++) {
    117         Instance unlabeled_instance = unlabeled_instances.instance(i);
    118 
    119         System.out.print("Making prediction for: " + i + "/" + num_instances);
    120         double classified_value = classifier.classifyInstance(unlabeled_instance);
    121         labeled_instances.instance(i).setClassValue(classified_value);
    122 
    123         String formatted_classified_value = String.format("% 06.3f", classified_value);
    124        
    125         System.out.print(" value = " + formatted_classified_value);
    126        
    127         if (groundtruth_instances != null) {
    128             Instance gt_instance = groundtruth_instances.instance(i);
    129             double gt_class_value = gt_instance.classValue();
    130             double error = Math.abs(classified_value - gt_class_value);
    131 
    132             String formatted_error = String.format("%.3f", error);
    133             System.out.print("  [error: " + formatted_error + "]");
    134         }
    135         System.out.println();
    136         }
    137 
    138         //
    139         // Save labeled data
    140         //
    141        
    142         System.out.println("Saving labeled instances: " + classified_data_output_filename);     
    143         FileWriter fw = new FileWriter(classified_data_output_filename);
    144         BufferedWriter bw = new BufferedWriter(fw);
    145        
    146         bw.write(labeled_instances.toString());
    147         bw.newLine();
    148         bw.flush();
    149         bw.close();
    150              
    151     }
    152     catch (Exception e) {
    153         e.printStackTrace();
    154     }
    155 
    156     */
    157 
    15837
    15938    Instances unlabeled_instances= WekaUtil.loadInstancesForClassification(unclassified_data_input_filename);

  • main/trunk/model-sites-dev/mars/src/java/org/greenstone/mars/WekaApplyValenceModel.java

    r34797 r34798  
    11package org.greenstone.mars;
    2 
    3 //import java.util.Random;
    42
    53import java.io.BufferedInputStream;

  • main/trunk/model-sites-dev/mars/src/java/org/greenstone/mars/WekaTrainArousalModel.java

    r34797 r34798  
    11package org.greenstone.mars;
    22   
    3 //import weka.core.converters.ConverterUtils.DataSource;
    43import weka.core.Instances;
    54 
    6 //import weka.filters.Filter;
    7 //import weka.filters.unsupervised.attribute.Remove;
    8 
    95import weka.classifiers.Classifier;
    106import weka.classifiers.Evaluation;
    … …  
    3531   
    3632    public static void main(String[] args)
    37     {
    38     /*
    39     if (args.length != 2) {
    40         System.err.println("Error: incorrect number of command-line arguments");
    41         System.err.println("Usage: input_training_data.arff output-model.{model|ser}");
    42         System.exit(1);
    43         }*/
    44    
     33    {   
    4534    WekaUtil.checkUsageTraining(args);
    4635   
    … …  
    6251    catch (Exception e) {
    6352        e.printStackTrace();
    64     }
    65    
    66 
    67     /*
    68         System.out.println("Training on ARFF file: " + input_arff_filename);
    69 
    70     try {
    71         DataSource data_source = new DataSource(input_arff_filename);
    72         Instances data_instances = data_source.getDataSet();
    73 
    74         // *********
    75         Instances filtered_data_instances = WekaUtil.applyFilter(data_instances,"472"); // top-up with removal of 'valence'
    76 
    77            
    78         // Filtering above has removed R472, so class to predict is numAttributes()-1
    79         filtered_data_instances.setClassIndex(filtered_data_instances.numAttributes()-1);
    80        
    81 
    82         // Build scheme/classifier
    83         REPTree classifier = new REPTree(); // scheme
    84         String[] reptree_options = weka.core.Utils.splitOptions("-M 2 -V 0.001 -N 3 -S 1 -L -1 -I 0.0");
    85         classifier.setOptions(reptree_options);
    86 
    87         classifier.buildClassifier(filtered_data_instances);
    88         System.out.println(classifier);
    89 
    90         / *
    91         // Evaluate     
    92         Evaluation eval = new Evaluation(filtered_data_instances);
    93         Random rand = new Random(1);
    94         int folds = 10;
    95         eval.crossValidateModel(classifier, filtered_data_instances, folds, rand);
    96         System.out.println(eval.toSummaryString());
    97         * /
    98        
    99         Evaluation eval = WekaUtil.evaluateClassifier(classifier,filtered_data_instances);
    100         System.out.println(eval.toSummaryString());
    101 
    102         System.out.println("Saving REPTree classifier model as: " + output_model_filename);
    103         weka.core.SerializationHelper.write(output_model_filename, classifier);
    104 
    105          
    106     }
    107     catch (Exception e) {
    108         e.printStackTrace();
    109     }
    110     */
     53    }   
    11154    }
    11255}

  • main/trunk/model-sites-dev/mars/src/java/org/greenstone/mars/WekaTrainValenceModel.java

    r34797 r34798  
    11package org.greenstone.mars;
    22   
    3 //import weka.core.converters.ConverterUtils.DataSource;
    43import weka.core.Instances;
    54 
    6 //import weka.filters.Filter;
    7 //import weka.filters.unsupervised.attribute.Remove;
    8 
    95import weka.classifiers.Classifier;
    106import weka.classifiers.Evaluation;
    11 //import weka.classifiers.trees.REPTree;
    127
    138// Based on:
    … …  
    2621    public static void main(String[] args)
    2722    {
    28     /*
    29     if (args.length != 2) {
    30         System.err.println("Error: incorrect number of command-line arguments");
    31         System.err.println("Usage: input_training_data.arff output-model.{model|ser}");
    32         System.exit(1);
    33         }*/
    34    
    3523    WekaUtil.checkUsageTraining(args);
    3624
    … …  
    5240    catch (Exception e) {
    5341        e.printStackTrace();
    54     }
    55 
    56     /* 
    57         System.out.println("Training on ARFF file: " + input_arff_filename);
    58 
    59     try {
    60         DataSource data_source = new DataSource(input_arff_filename);
    61         Instances data_instances = data_source.getDataSet();
    62 
    63         // *********
    64         Instances filtered_data_instances = WekaUtil.applyFilter(data_instances,"471"); // top-up with removal of 'arousal'
    65 
    66         // ****** comment varies
    67         // With removal of R471, last column is the one we want to predict => numAttributes()-1
    68         filtered_data_instances.setClassIndex(filtered_data_instances.numAttributes()-1);
    69        
    70 
    71         // Build scheme/classifier
    72         REPTree classifier = new REPTree(); // scheme
    73         String[] reptree_options = weka.core.Utils.splitOptions("-M 2 -V 0.001 -N 3 -S 1 -L -1 -I 0.0");
    74         classifier.setOptions(reptree_options);
    75 
    76         classifier.buildClassifier(filtered_data_instances);
    77         System.out.println(classifier);
    78 
    79        
    80         Evaluation eval = WekaUtil.evaluateClassifier(classifier,filtered_data_instances);
    81         System.out.println(eval.toSummaryString());
    82            
    83         System.out.println("Saving REPTree classifier model as: " + output_model_filename);
    84         weka.core.SerializationHelper.write(output_model_filename, classifier);
    85 
    86          
    87     }
    88     catch (Exception e) {
    89         e.printStackTrace();
    90     }
    91  
    92     */
    93    
     42    }   
    9443    }
    9544}
Note: See TracChangeset for help on using the changeset viewer.