cahya commited on
Commit
2cf743a
1 Parent(s): 15a59f8

remove unigrams

Browse files
Files changed (1) hide show
  1. run_evaluation.py +1 -3
run_evaluation.py CHANGED
@@ -22,9 +22,7 @@ class KenLM:
22
  vocab_dict = tokenizer.get_vocab()
23
  self.vocabulary = [x[0] for x in sorted(vocab_dict.items(), key=lambda x: x[1], reverse=False)]
24
  self.vocabulary = self.vocabulary[:-1]
25
- with open(unigrams, "r") as f:
26
- unigrams_text = f.read()
27
- self.decoder = build_ctcdecoder(self.vocabulary, model_name, unigrams=unigrams_text)
28
 
29
  @staticmethod
30
  def lm_postprocess(text):
 
22
  vocab_dict = tokenizer.get_vocab()
23
  self.vocabulary = [x[0] for x in sorted(vocab_dict.items(), key=lambda x: x[1], reverse=False)]
24
  self.vocabulary = self.vocabulary[:-1]
25
+ self.decoder = build_ctcdecoder(self.vocabulary, model_name)
 
 
26
 
27
  @staticmethod
28
  def lm_postprocess(text):