描述

我正在尝试复制 Langchain tutorial 以使用 OllamaFunctions 进行网页抓取，就像在 Google Colab 环境中展示的那样。

代码

[1] %%capture
     !pip install langchain_experimental

[2] from langchain_experimental.llms.ollama_functions import OllamaFunctions

     lm = OllamaFunctions(model="llama2:13b",
                      base_url="http://localhost:11434",
                      temperature=0)

[3] %%capture
     !pip install -q langchain-openai langchain playwright beautifulsoup4
     !playwright install

[4] import nest_asyncio
     nest_asyncio.apply()

[5] from langchain.chains import create_extraction_chain
     schema = {
          "properties": {
           "news_article_title": {"type": "string"},
           "news_article_summary": {"type": "string"},
              },
           "required": ["news_article_title", "news_article_summary"],
      }

     def extract(content: str, schema: dict):
      return create_extraction_chain(schema=schema, llm=llm, verbose=True).invoke(content)

[6] import pprint
     from langchain.text_splitter import RecursiveCharacterTextSplitter
     from langchain_community.document_loaders import AsyncChromiumLoader
     from langchain_community.document_transformers import BeautifulSoupTransformer

    def scrape_with_playwright(urls, schema):
         loader = AsyncChromiumLoader(urls)
         docs = loader.load()
         bs_transformer = BeautifulSoupTransformer()
         docs_transformed = bs_transformer.transform_documents(
                docs, tags_to_extract=["span"]
         )
         print("Extracting content with LLM")
         # Grab the first 1000 tokens of the site
         splitter = RecursiveCharacterTextSplitter.from_tiktoken_encoder(
                   chunk_size=1000,
                    chunk_overlap=0,
                     separators=["\n"]
          )
         splits = splitter.split_documents(docs_transformed)
         print("Number of splits:", len(splits))  # Add this debugging statement
         if splits:  # Check if splits list is not empty
                # Process the first split
                extracted_content = extract(schema=schema, content=splits[0].page_content) #  Line where error occurs
                pprint.pprint(extracted_content)
                return extracted_content
         else:
                 print("No splits found")  # Add this debugging statement
                 return None```

[7] urls = ["https://www.nytimes.com/"]
     extracted_content = scrape_with_playwright(urls, schema=schema) python

错误

但是我遇到了以下错误：

ConnectionError: HTTPConnectionPool(host='localhost', port=11434): Max retries exceeded with url: /api/chat/ (Caused by NewConnectionError('<urllib3.connection.HTTPConnection object at 0x7f7b19911300>: Failed to establish a new connection: [Errno 111] Connection refused'))