Added an option to include a header to the output file
Changed the default Stream Reader to RTG (a synthetic data stream).
Adding the header required changes to other classes, such as Evaluator, StreamWritter and BasicClassificationEvaluator
To test these changes:
This will run RTG with MultinomialNaiveBayes and output the header in log_RTG.txt
./spark.sh "EvaluatePrequential -l (org.apache.spark.streamdm.classifiers.bayes.MultinomialNaiveBayes) -s (org.apache.spark.streamdm.streams.generators.RandomTreeGenerator) -h" 1> log_RTG.txt 2> error_RTG.txt
Similar to the previous test, but this will yield the default (previous) approach: no header.
./spark.sh "EvaluatePrequential -l (org.apache.spark.streamdm.classifiers.bayes.MultinomialNaiveBayes) -s (org.apache.spark.streamdm.streams.generators.RandomTreeGenerator)" 1> log_RTG.txt 2> error_RTG.txt
This will use the new default stream reader, i.e., RTG
./spark.sh "EvaluatePrequential -l (org.apache.spark.streamdm.classifiers.bayes.MultinomialNaiveBayes) -h" 1> log.txt 2> error.txt
To test these changes:
This will run RTG with MultinomialNaiveBayes and output the header in log_RTG.txt
./spark.sh "EvaluatePrequential -l (org.apache.spark.streamdm.classifiers.bayes.MultinomialNaiveBayes) -s (org.apache.spark.streamdm.streams.generators.RandomTreeGenerator) -h" 1> log_RTG.txt 2> error_RTG.txtSimilar to the previous test, but this will yield the default (previous) approach: no header.
./spark.sh "EvaluatePrequential -l (org.apache.spark.streamdm.classifiers.bayes.MultinomialNaiveBayes) -s (org.apache.spark.streamdm.streams.generators.RandomTreeGenerator)" 1> log_RTG.txt 2> error_RTG.txtThis will use the new default stream reader, i.e., RTG
./spark.sh "EvaluatePrequential -l (org.apache.spark.streamdm.classifiers.bayes.MultinomialNaiveBayes) -h" 1> log.txt 2> error.txt