当前位置：首页 > article >正文

LangChain pdf的读取以及向量数据库的使用

article 2025/1/16 12:40:50

以下使用了3399.pdf， Rockchip RK3399 TRM Part1

import ChatGLM
from langchain.chains import LLMChain
from langchain_core.output_parsers import StrOutputParser
from langchain_core.prompts import ChatPromptTemplate
from langchain.chains import SimpleSequentialChain
from langchain_core.runnables import RunnablePassthrough
from operator import itemgetter
from langchain_community.document_loaders import PyPDFLoader
import ChatGLM
from langchain.chains import LLMChain
from langchain_core.output_parsers import StrOutputParser
from langchain_core.prompts import ChatPromptTemplate
from langchain_community.tools.tavily_search import TavilySearchResults
from langchain.chains import LLMMathChain
from langchain_core.output_parsers import StrOutputParser
from langchain_core.prompts import ChatPromptTemplate
from langchain_community.document_loaders import TextLoader
from langchain.text_splitter import CharacterTextSplitter
from langchain_community.vectorstores import Chroma
from langchain_core.output_parsers import StrOutputParser
from langchain_core.prompts import ChatPromptTemplate
from langchain_core.runnables import RunnableParallel, RunnablePassthrough
from langchain_community.embeddings import JinaEmbeddings

# https://jina.ai/embeddings/
# https://python.langchain.com/docs/integrations/text_embedding/jina
# demo:  https://python.langchain.com/cookbook



llm = ChatGLM.ChatGLM_LLM()
loader = PyPDFLoader("3399.pdf")
documents = loader.load_and_split()

embeddings = JinaEmbeddings(
    jina_api_key="jina_fa2c341a2f634f1381f7cfec767150caSconYmQA2XRAcVKfZ7-Zboaqeydu", model_name="jina-embeddings-v2-base-en"
)

vectorstore = Chroma.from_documents(documents, embeddings)
retriever = vectorstore.as_retriever()

template = """Answer the question based only on the following context:
{context}

Question: {question}
"""
prompt = ChatPromptTemplate.from_template(template)
llm = ChatGLM.ChatGLM_LLM()
output_parser = StrOutputParser()
setup_and_retrieval = RunnableParallel(
    {"context": retriever, "question": RunnablePassthrough()}
)
chain = setup_and_retrieval | prompt | llm | output_parser

print(chain.invoke("eFuse Function Description"))