Python LangChain Patterns | Generated by AI
Here are a few concise, illustrative Python snippets showing common LangChain patterns. You can mix and match these pieces to build more complex workflows.
1. Simple LLM Chain with a PromptTemplate
from langchain import OpenAI, LLMChain, PromptTemplate
# 1. Initialize the LLM
llm = OpenAI(temperature=0.7)
# 2. Create a prompt template
template = PromptTemplate(
input_variables=["product"],
template="Write a short marketing blurb for the product: {product}."
)
# 3. Build the chain
chain = LLMChain(llm=llm, prompt=template)
# 4. Run it
result = chain.run({"product": "wireless noise-cancelling headphones"})
print(result)
2. Conversation Chain with Memory
from langchain import OpenAI, ConversationChain
from langchain.memory import ConversationBufferMemory
# 1. LLM and memory
llm = OpenAI(temperature=0)
memory = ConversationBufferMemory()
# 2. Build the conversation chain
conv = ConversationChain(llm=llm, memory=memory)
# 3. Simulate dialogue
print(conv.predict(input="Hi there!"))
print(conv.predict(input="Can you remind me what we talked about before?"))
3. Retrieval + QA over Documents (FAISS)
from langchain import OpenAI, RetrievalQA
from langchain.vectorstores import FAISS
from langchain.embeddings.openai import OpenAIEmbeddings
from langchain.text_splitter import RecursiveCharacterTextSplitter
from langchain.document_loaders import TextLoader
# 1. Load and split documents
loader = TextLoader("my_docs/guide.txt")
docs = loader.load()
splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=100)
chunks = splitter.split_documents(docs)
# 2. Embed and index
embeddings = OpenAIEmbeddings()
vectorstore = FAISS.from_documents(chunks, embeddings)
# 3. Build a RetrievalQA chain
qa = RetrievalQA.from_chain_type(
llm=OpenAI(temperature=0),
chain_type="stuff",
retriever=vectorstore.as_retriever()
)
# 4. Ask questions
answer = qa.run("What are the primary benefits of our API?")
print(answer)
4. Agent with Tools
from langchain import OpenAI, SerpAPIWrapper, LLMMathChain, initialize_agent
from langchain.agents import Tool
# 1. Define tools
search = SerpAPIWrapper()
tools = [
Tool(
name="Search",
func=search.run,
description="Useful for web searches"
),
Tool(
name="Calculator",
func=LLMMathChain.from_llm(OpenAI(temperature=0)).run,
description="Useful for math calculations"
)
]
# 2. Create agent
agent = initialize_agent(
tools,
llm=OpenAI(temperature=0),
agent="zero-shot-react-description",
verbose=True
)
# 3. Run agent
response = agent.run("What's the capital of France plus two times five?")
print(response)
Feel free to adapt any of these snippets—swap in different LLMs, change retrievers (e.g., Chroma, Pinecone), or add your own tools and memory types as needed!