程序员练级攻略：机器学习和人工智能-[2026重制版]

from sklearn.model_selection import train_test_split
from sklearn.ensemble import RandomForestClassifier
from sklearn.metrics import accuracy_score, classification_report
import pandas as pd
 
# 1. 加载数据
df = pd.read_csv('customer_data.csv')
X = df.drop('churn', axis=1)  # 特征
y = df['churn']              # 标签
 
# 2. 划分训练集/测试集
X_train, X_test, y_train, y_test = train_test_split(
    X, y, test_size=0.2, random_state=42
)
 
# 3. 训练模型
model = RandomForestClassifier(
    n_estimators=100,
    max_depth=10,
    random_state=42
)
model.fit(X_train, y_train)
 
# 4. 预测与评估
y_pred = model.predict(X_test)
print(f"准确率: {accuracy_score(y_test, y_pred):.2%}")
print(classification_report(y_test, y_pred))
 
# 5. 特征重要性
importances = pd.Series(model.feature_importances_, index=X.columns)
print(importances.sort_values(ascending=False).head(10))

import torch
import torch.nn as nn
import torch.optim as optim
 
# 定义简单的神经网络
class SimpleNet(nn.Module):
    def __init__(self, input_size, hidden_size, num_classes):
        super().__init__()
        self.fc1 = nn.Linear(input_size, hidden_size)
        self.relu = nn.ReLU()
        self.dropout = nn.Dropout(0.3)
        self.fc2 = nn.Linear(hidden_size, num_classes)
 
    def forward(self, x):
        x = self.fc1(x)
        x = self.relu(x)
        x = self.dropout(x)
        x = self.fc2(x)
        return x
 
# 初始化模型
model = SimpleNet(input_size=784, hidden_size=128, num_classes=10)
criterion = nn.CrossEntropyLoss()
optimizer = optim.Adam(model.parameters(), lr=0.001)
 
# 训练循环
for epoch in range(10):
    model.train()
    for batch_x, batch_y in train_loader:
        optimizer.zero_grad()
        outputs = model(batch_x)
        loss = criterion(outputs, batch_y)
        loss.backward()
        optimizer.step()
 
    print(f'Epoch {epoch+1}, Loss: {loss.item():.4f}')

# 结构化 Prompt 模板
SYSTEM_PROMPT = """你是一个专业的技术文档撰写专家。
你的任务是将用户提供的原始内容重写为高质量的技术文章。
 
## 输出要求：
1. 字数不少于3000字
2. 包含至少2张Mermaid图表
3. 使用中文输出
4. 保持专业但易懂的语气
 
## 重写规范：
- 标题格式：[原编号]-[原标题]-[2026重制版]
- 更新至2026年最新技术栈
- 添加代码示例和实践建议"""
 
def generate_article(original_content: str) -> str:
    prompt = f"""请根据以下原文，按照上述规则重写为2026年重制版：
 
## 原文内容：
{original_content}
 
请开始重写："""
    
    # 调用 LLM API
    response = client.chat.completions.create(
        model="gpt-4o",
        messages=[
            {"role": "system", "content": SYSTEM_PROMPT},
            {"role": "user", "content": prompt}
        ],
        temperature=0.7,
        max_tokens=8000
    )
    return response.choices[0].message.content

问题：一个篮子里有5个苹果，小明拿了2个，又放了3个进去。
现在篮子里有几个苹果？
 
❌ 直接回答：6个
 
✅ CoT 思维过程：
初始状态：篮子有5个苹果
步骤1：小明拿走2个 → 5 - 2 = 3个
步骤2：放入3个 → 3 + 3 = 6个
答案：6个

from langchain_community.vectorstores import PGVector
from langchain_openai import OpenAIEmbeddings, ChatOpenAI
from langchain.text_splitter import RecursiveCharacterTextSplitter
from langchain.chains import create_retrieval_chain
from langchain.prompts import ChatPromptTemplate
 
# 1. 加载并切分文档
with open("tech_doc.md") as f:
    text = f.read()
 
text_splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=200)
chunks = text_splitter.split_text(text)
 
# 2. 创建向量存储（使用pgvector扩展的PostgreSQL）
embeddings = OpenAIEmbeddings(model="text-embedding-3-small")
vectorstore = PGVector.from_texts(
    texts=chunks,
    embedding=embeddings,
    connection_string="postgresql://user:pass@localhost:5432/dbname",
    collection_name="tech_docs"
)
 
# 3. 创建RAG链
llm = ChatOpenAI(model="gpt-4o", temperature=0)
 
prompt = ChatPromptTemplate.from_template("""
基于以下上下文信息回答问题。如果不知道答案，就说不知道。
 
上下文：
{context}
 
问题：{question}
""")
 
retrieval_chain = create_retrieval_chain(
    llm=llm,
    retriever=vectorstore.as_retriever(search_kwargs={"k": 3}),
    prompt=prompt
)
 
# 4. 使用
response = retrieval_chain.invoke({"question": "什么是异步I/O模型？"})
print(response["answer"])

from langchain.agents import create_tool_calling_agent, Tool
from langchain_openai import ChatOpenAI
from langchain_community.tools import DuckDuckGoSearchRun
 
# 定义工具集
tools = [
    Tool(
        name="search",
        func=DuckDuckGoSearchRun(),
        description="搜索互联网获取最新信息"
    ),
    Tool(
        name="calculator",
        func=lambda x: eval(x),
        description="进行数学计算"
    )
]
 
# 创建 Agent
llm = ChatOpenAI(model="gpt-4o", temperature=0)
agent = create_tool_calling_agent(llm, tools, verbose=True)
 
# 运行
result = agent.invoke({"input": "2026年Python最新版本是什么？"})
print(result["output"])