办理llama_index中利用Ollama出现timed out 问题

征象：

1.   File "~/anaconda3/envs/leo_py38/lib/python3.8/site-packages/httpx/_transports/default.py", line 86, in map_httpcore_exceptions
2.     raise mapped_exc(message) from exc
3. httpx.ReadTimeout: timed out

复制代码

代码：

1. from llama_index.core import VectorStoreIndex, SimpleDirectoryReader, Settings
2. from llama_index.embeddings.huggingface import HuggingFaceEmbedding
3. from llama_index.llms.ollama import Ollama
5. ###定义使用的llm，embedding 模型
6. llm = Ollama(model="yi:34b")
7. embed_model = HuggingFaceEmbedding(model_name="BAAI/bge-large-zh-v1.5")
8. Settings.llm = llm
9. Settings.embed_model = embed_model
11. ## 加载文档
12. documents = SimpleDirectoryReader("../../data").load_data()
13. print("documents: ", len(documents))
15. ## 构建index
16. index = VectorStoreIndex.from_documents(documents, show_progress=True)
18. ## 构建query engine
19. query_engine = index.as_query_engine()
20. query = "身长九尺，髯长二尺的人是谁？"
22. ## query
23. response = query_engine.query(query)
24. print(f"query:{query}")
25. print(f"查询结果：{response}")

复制代码

办理办法是增加timeout时长

1. llm = Ollama(model="yi:34b", request_timeout=500)

复制代码

免责声明：如果侵犯了您的权益，请联系站长，我们会及时删除侵权内容，谢谢合作！更多信息从访问主页：qidao123.com:ToB企服之家，中国第一个企服评测及商务社交产业平台。