eTok = TokenizerFactory.createTokenizer(fs, eLang, tokenizerFile, null);
sLogger.info("Tokenizer and vocabs created successfully.");
eScoreFn = (ScoringModel) new Bm25();
eScoreFn.setAvgDocLength(lang2AvgSentLen.get(eLang)); //average sentence length = heuristic based on De-En data
eScoreFn.setDocCount(env.readCollectionDocumentCount());
dict = new DefaultFrequencySortedDictionary(new Path(env.getIndexTermsData()), new Path(env.getIndexTermIdsData()), new Path(env.getIndexTermIdMappingData()), fs);
dfTable = new DfTableArray(new Path(env.getDfByTermData()), fs);
//for backward compatibility