microsoft/generative-ai-for-beginners：109k Stars 生成式AI入门完全指南

# 克隆课程仓库（不含翻译，减小体积）
git clone --filter=blob:none --sparse \
  https://github.com/microsoft/generative-ai-for-beginners.git

cd generative-ai-for-beginners

# 只下载英文版
git sparse-checkout set --no-cone \
  '/*' '!translations' '!translated_images'

# 安装依赖
pip install -r requirements.txt
# 或
npm install

# Python 示例：Azure OpenAI
from openai import AzureOpenAI

client = AzureOpenAI(
    api_key=os.getenv("AZURE_OPENAI_KEY"),
    api_version="2024-02-01",
    azure_endpoint=os.getenv("AZURE_OPENAI_ENDPOINT")
)

response = client.chat.completions.create(
    model="gpt-4",
    messages=[
        {"role": "user", "content": "写一首关于 AI 的诗"}
    ]
)
print(response.choices[0].message.content)

# 1. 文本转向量
from openai import OpenAI
client = OpenAI()

response = client.embeddings.create(
    model="text-embedding-3-small",
    input="要嵌入的文本"
)
vector = response.data[0].embedding

# 2. 存储到向量数据库
# 支持：Pinecone、Milvus、Chroma、FAISS

# 3. 相似度搜索
query_vector = get_embedding("搜索query")
results = vector_db.search(
    query_vector,
    top_k=5
)

# 定义可调用的函数
functions = [
    {
        "name": "get_weather",
        "description": "获取城市天气",
        "parameters": {
            "type": "object",
            "properties": {
                "city": {
                    "type": "string",
                    "description": "城市名称"
                }
            },
            "required": ["city"]
        }
    }
]

# LLM 自动选择调用
response = client.chat.completions.create(
    model="gpt-4-turbo",
    messages=messages,
    tools=functions,
    tool_choice="auto"
)

# 1. 文档加载和分块
from langchain.document_loaders import PDFLoader
loader = PDFLoader("document.pdf")
documents = loader.load()

# 2. 文本分块
from langchain.text_splitter import RecursiveCharacterTextSplitter
splitter = RecursiveCharacterTextSplitter(chunk_size=1000)
chunks = splitter.split_documents(documents)

# 3. 向量化并存储
from langchain.vectorstores import Chroma
db = Chroma.from_documents(chunks, embeddings)

# 4. 检索增强生成
retrieved = db.similarity_search(query)
context = "\n".join([doc.page_content for doc in retrieved])
prompt = f"基于以下上下文回答：{context}\n\n问题：{query}"

# 简化 Agent 框架
class SimpleAgent:
    def __init__(self, llm, tools):
        self.llm = llm
        self.tools = tools

    def run(self, task):
        # 1. 规划
        plan = self.llm Planner(task, self.tools)

        # 2. 执行
        for step in plan:
            result = self.execute_step(step)

        # 3. 返回结果
        return result

# LoRA 微调示例（使用 PEFT）
from peft import LoraConfig, get_peft_model

config = LoraConfig(
    r=8,
    lora_alpha=16,
    target_modules=["q_proj", "v_proj"],
    lora_dropout=0.05,
    bias="none",
    task_type="CAUSAL_LM"
)
model = get_peft_model(base_model, config)