MLflow 集成

pip install mlflow langchain-openai langgraph -qU

import os

# 如果您运行了 MLflow 跟踪服务器，请设置 MLflow 跟踪 URI
os.environ["MLFLOW_TRACKING_URI"] = ""
os.environ["OPENAI_API_KEY"] = ""

import mlflow

# 可选：设置一个实验来组织您的追踪
mlflow.set_experiment("LangChain MLflow Integration")

# 启用追踪
mlflow.langchain.autolog()

import mlflow
from langchain_core.output_parsers import StrOutputParser
from langchain_core.prompts import ChatPromptTemplate
from langchain_openai import ChatOpenAI

# 启用 MLflow 追踪
mlflow.langchain.autolog()

# 创建一个简单的链
llm = ChatOpenAI(model_name="gpt-5.4")

prompt = ChatPromptTemplate.from_messages(
    [
        (
            "system",
            "You are a helpful assistant that translates {input_language} to {output_language}.",
        ),
        ("human", "{input}"),
    ]
)

chain = prompt | llm | StrOutputParser()

# 运行链
result = chain.invoke(
    {
        "input_language": "English",
        "output_language": "German",
        "input": "I love programming.",
    }
)

import mlflow
from langchain.tools import tool
from langchain.agents import create_agent


# 启用 MLflow 追踪
mlflow.langchain.autolog()


# 定义一个工具
@tool
def count_words(text: str) -> str:
    """Counts the number of words in a text."""
    word_count = len(text.split())
    return f"This text contains {word_count} words."


# 创建一个 LangGraph 代理
llm = ChatOpenAI(model="gpt-5.4")
tools = [count_words]
graph = create_agent(llm, tools)

# 运行代理
result = graph.invoke(
    {"messages": [{"role": "user", "content": "Write me a 71-word story about a cat."}]}
)