Tongyi Qianwen Langchain adapter
Project description
灵积-通义千问 Langchain
探索 通义千问 Api 在 langchain 中的使用 参考借鉴 openai langchain 的实现 目前仅用于个人学习
pip install langchain-qianwen
llms
from langchain.callbacks.streaming_stdout import StreamingStdOutCallbackHandler
from langchain_qianwen import Qwen_v1
if __name__ == "__main__":
llm = Qwen_v1(
model_name="qwen-v1",
streaming=True,
callbacks=[StreamingStdOutCallbackHandler()],
)
question = "你好, 帮忙解释一下 Hello World 是什么意思"
llm(question)
chat_models
from langchain.callbacks.streaming_stdout import StreamingStdOutCallbackHandler
from langchain_qianwen import ChatQwen_v1
from langchain.schema import (
HumanMessage,
)
if __name__ == "__main__":
chat = ChatQwen_v1(
model_name="qwen-v1",
streaming=True,
callbacks=[StreamingStdOutCallbackHandler()],
)
chat([HumanMessage(content="举例说明一下 PHP 为什么是世界上最好的语言")])
使用 agent 增加网络查询功能
from langchain.agents import load_tools, AgentType, initialize_agent
from langchain_qianwen import Qwen_v1
if __name__ == "__main__":
llm = Qwen_v1(
##经测试只有 plus 模型才能正常使用 agent
model_name="qwen-plus",
)
## 需要去 serpapi 官网申请一个 api_key
tool_names = ["serpapi"]
tools = load_tools(tool_names)
agent = initialize_agent(tools=tools,
llm=llm,
agent=AgentType.ZERO_SHOT_REACT_DESCRIPTION,
verbose=True)
agent.run("最近在福岛海滩上发现哥斯拉了吗?")
使用 embedding 提取文档中的信息
from langchain.embeddings.dashscope import DashScopeEmbeddings
from langchain.vectorstores import Chroma
from langchain.text_splitter import CharacterTextSplitter
from langchain.chains import RetrievalQA
from langchain.document_loaders import DirectoryLoader
from langchain_qianwen import Qwen_v1
if __name__ == "__main__":
llm = Qwen_v1(
model_name="qwen-turbo",
)
loader = DirectoryLoader("./assets", glob="**/*.txt")
document = loader.load()
text_splitter = CharacterTextSplitter(chunk_size=2048, chunk_overlap=0)
texts = text_splitter.split_documents(document)
embeddings = DashScopeEmbeddings(
model="text-embedding-v1",
)
print(f"text length: {len(texts)}")
# 使用 embedding engion 将 text 转换为向量
db = Chroma.from_documents(texts, embeddings)
retriever = db.as_retriever()
# qa = RetrievalQA.from_chain_type(llm=llm, chain_type="stuff", retriever=retriever, return_source_documents=True)
qa = RetrievalQA.from_chain_type(llm=llm, chain_type="stuff", retriever=retriever)
query = "文中的工厂模式使用例子有哪些??"
rsp = qa.run({"query": query})
print(rsp)
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
langchain_qianwen-0.1.3.tar.gz
(10.4 kB
view hashes)
Built Distribution
Close
Hashes for langchain_qianwen-0.1.3-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | d509f2c3839f5fc1c5ebe3dbc3a22b680302c44e1fcbe05c9d0c0016d4f57be8 |
|
MD5 | c7b1e1d25ff42a7bc2d4768dfde063c6 |
|
BLAKE2b-256 | aa1be2aa20840c9f5a6cb7f9ff4dd34e5b4292d4e75506282d3acfc5925b0e41 |