PrompTart commited on
Commit
3ae9bc6
β€’
1 Parent(s): 5b29e42

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +1 -1
README.md CHANGED
@@ -38,7 +38,7 @@ tokenizer.src_lang = "en"
38
  encoded = tokenizer(text.split('. '), return_tensors="pt", padding=True)
39
  generated_tokens = model.generate(**encoded, forced_bos_token_id=tokenizer.get_lang_id("ko"))
40
  outputs = tokenizer.batch_decode(generated_tokens, skip_special_tokens=True)
41
- print(' '.join(outputs))
42
  # => "이 λͺ¨λΈμ€ 지식 증λ₯˜ 기법(knowledge distillation techniques)을 μ‚¬μš©ν•˜μ—¬ λ―Έμ„Έ μ‘°μ •λ˜μ—ˆμŠ΅λ‹ˆλ‹€.
43
  # ν›ˆλ ¨ 데이터셋(training dataset)은 λŒ€ν˜• μ–Έμ–΄ λͺ¨λΈ(large language models)을 기반으둜 ν•œ ν˜‘μ—… 닀쀑 μ—μ΄μ „νŠΈ ν”„λ ˆμž„μ›Œν¬(collaborative multi-agent framework)λ₯Ό μ‚¬μš©ν•˜μ—¬ μƒμ„±λ˜μ—ˆμŠ΅λ‹ˆλ‹€."
44
 
 
38
  encoded = tokenizer(text.split('. '), return_tensors="pt", padding=True)
39
  generated_tokens = model.generate(**encoded, forced_bos_token_id=tokenizer.get_lang_id("ko"))
40
  outputs = tokenizer.batch_decode(generated_tokens, skip_special_tokens=True)
41
+ print('\n'.join(outputs))
42
  # => "이 λͺ¨λΈμ€ 지식 증λ₯˜ 기법(knowledge distillation techniques)을 μ‚¬μš©ν•˜μ—¬ λ―Έμ„Έ μ‘°μ •λ˜μ—ˆμŠ΅λ‹ˆλ‹€.
43
  # ν›ˆλ ¨ 데이터셋(training dataset)은 λŒ€ν˜• μ–Έμ–΄ λͺ¨λΈ(large language models)을 기반으둜 ν•œ ν˜‘μ—… 닀쀑 μ—μ΄μ „νŠΈ ν”„λ ˆμž„μ›Œν¬(collaborative multi-agent framework)λ₯Ό μ‚¬μš©ν•˜μ—¬ μƒμ„±λ˜μ—ˆμŠ΅λ‹ˆλ‹€."
44