先跑出第一个结果，
再问模型怎么学会的。

pip install numpy scikit-learn

# -*- coding: utf-8 -*-
"""10 分钟训练你的第一个模型 —— 不用下载任何数据"""
from sklearn.datasets import load_breast_cancer
from sklearn.model_selection import train_test_split
from sklearn.linear_model import LogisticRegression
from sklearn.tree import DecisionTreeClassifier
from sklearn.preprocessing import StandardScaler

# ===== 第 1 步：拿数据（sklearn 自带，569 个肿瘤样本，30 个特征）=====
X, y = load_breast_cancer(return_X_y=True)
print(f"数据形状：{X.shape}  →  {X.shape[0]} 个样本，每个 {X.shape[1]} 个特征")
print(f"标签：{set(y)}   （0=恶性, 1=良性）")

# ===== 第 2 步：切分 —— 训练集用来学，测试集用来考试 =====
X_tr, X_te, y_tr, y_te = train_test_split(
    X, y, test_size=0.3, random_state=42, stratify=y)
print(f"训练集 {len(X_tr)} 个，测试集 {len(X_te)} 个")

# ===== 第 3 步：训练两个模型 =====
# 模型A：逻辑回归（线性模型，需要标准化）
scaler = StandardScaler().fit(X_tr)
lr = LogisticRegression(max_iter=1000)
lr.fit(scaler.transform(X_tr), y_tr)

# 模型B：决策树，故意不限制深度
tree = DecisionTreeClassifier(random_state=42)
tree.fit(X_tr, y_tr)

# ===== 第 4 步：看结果 =====
print("\n" + "=" * 46)
print(f"{'模型':<16}{'训练集':>10}{'测试集':>10}{'差距':>10}")
print("=" * 46)
for name, m, Xa, Xb in [
    ("逻辑回归", lr, scaler.transform(X_tr), scaler.transform(X_te)),
    ("决策树(无限制)", tree, X_tr, X_te),
]:
    tr = m.score(Xa, y_tr)
    te = m.score(Xb, y_te)
    print(f"{name:<16}{tr:>9.1%}{te:>10.1%}{tr-te:>10.1%}")
print("=" * 46)

# ===== 第 5 步：给决策树限制深度，再看 =====
print("\n决策树深度对效果的影响：")
print(f"{'最大深度':<10}{'训练集':>10}{'测试集':>10}{'差距':>10}")
for d in [1, 2, 3, 5, 10, None]:
    t = DecisionTreeClassifier(max_depth=d, random_state=42).fit(X_tr, y_tr)
    tr, te = t.score(X_tr, y_tr), t.score(X_te, y_te)
    label = str(d) if d else "不限制"
    print(f"{label:<10}{tr:>9.1%}{te:>10.1%}{tr-te:>10.1%}")