+ " entries.");
Path outputPath = new Path(outputDirName);
HadoopUtil.delete(conf, outputPath);
CorpusVectorizer corpus = new BnsCorpusVectorizer(new HDFSCorpusLoaderFactory());
log.info("Vectorizing train documents...");
corpus.convertToSequenceFile(conf, trainDir, outputDirName + "/train.seq",
new BnsCorpusLineParser());
log.info("Vectorizing test documents...");
corpus.convertToSequenceFile(conf, testDir, outputDirName + "/test.seq",
new BnsCorpusLineParser());
log.info("BNS Vectorization successful!");
return Job.SUCCESS;
}