File "D:\llm\VideoLingo-1.7.0\st.py", line 117, in
main()
File "D:\llm\VideoLingo-1.7.0\st.py", line 113, in main
text_processing_section()
File "D:\llm\VideoLingo-1.7.0\st.py", line 30, in text_processing_section
process_text()
File "D:\llm\VideoLingo-1.7.0\st.py", line 48, in process_text
step3_2_splitbymeaning.split_sentences_by_meaning()
File "D:\llm\VideoLingo-1.7.0\core\step3_2_splitbymeaning.py", line 118, in split_sentences_by_meaning
sentences = parallel_split_sentences(sentences, max_length=load_key("max_split_length"), max_workers=load_key("max_workers"), nlp=nlp, retry_attempt=retry_attempt)
File "D:\llm\VideoLingo-1.7.0\core\step3_2_splitbymeaning.py", line 100, in parallel_split_sentences
split_result = future.result()
File "d:\Users\NINGMEI\anaconda3\envs\videolingo\lib\concurrent\futures_base.py", line 445, in result
return self.__get_result()
File "d:\Users\NINGMEI\anaconda3\envs\videolingo\lib\concurrent\futures_base.py", line 390, in __get_result
raise self.exception
File "d:\Users\NINGMEI\anaconda3\envs\videolingo\lib\concurrent\futures\thread.py", line 52, in run
result = self.fn(*self.args, **self.kwargs)
File "D:\llm\VideoLingo-1.7.0\core\step3_2_splitbymeaning.py", line 60, in split_sentence
best_split = response_data[f"split{response_data['best']}"]
KeyError: 'split_Due to slight differences in flow between the two methods outlined above (considering not just length but also the readability of them as independent units), method 1 might generally be preferred because it retains each point as a separate unit, which is clearer for context and smoother reading by preserving the natural progression of ideas within subtitles.'
这是 llm 响应格式出错，请问使用的什么 llm
ollama:qwen
---- 回复的原邮件 ----
| 发件人 | @.> |
| 日期 | 2024年11月10日 09:54 |
| 收件人 | @.> |
| 抄送至 | @.>@.> |
| 主题 | Re: [Huanshere/VideoLingo] 这是什么问题，方法第一种好还第二种好？？ (Issue #243) |
这是 llm 响应格式出错，请问使用的什么 llm
—
Reply to this email directly, view it on GitHub, or unsubscribe.
You are receiving this because you authored the thread.Message ID: @.***>
不建议使用小模型，在安装文档中有说明，小模型的格式遵循不好
Huanshere
