# 1. 基础配置：屏蔽警告、Mac中文适配、导入所有工具库
import warnings
warnings.filterwarnings("ignore")
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
from sklearn.datasets import load_iris
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import StandardScaler
from sklearn.linear_model import LogisticRegression
from sklearn.metrics import accuracy_score, classification_report, confusion_matrix, roc_curve, roc_auc_score

# Mac系统中文显示适配，解决乱码问题
plt.rcParams["font.sans-serif"] = ["PingFang SC", "Arial Unicode MS"]
plt.rcParams["axes.unicode_minus"] = False
# 全局图表清晰度设置
plt.rcParams["figure.dpi"] = 100

# 2. 加载替代数据集：鸢尾花二分类数据集（与题目乳腺癌数据集完全不同）
# 仅选取山鸢尾（0）和变色鸢尾（1）两个类别，构建二分类任务
iris = load_iris()
X = iris.data[iris.target != 2]  # 剔除第三类，仅保留二分类数据
y = iris.target[iris.target != 2]
class_names = iris.target_names[:2]  # 类别名称：['setosa', 'versicolor']
feature_names = iris.feature_names
print(f"替代数据集特征维度：{X.shape}")
print(f"替代数据集类别名称：{class_names}")

# 3. 拆分训练集与测试集：测试集占比30%，随机种子42，分层抽样
X_train, X_test, y_train, y_test = train_test_split(
    X, y, test_size=0.3, random_state=42, stratify=y
)

# 4. 特征数据标准化处理
scaler = StandardScaler()
X_train_scaled = scaler.fit_transform(X_train)
X_test_scaled = scaler.transform(X_test)

# 5. 定义LogisticRegression模型，设置参数
model = LogisticRegression(random_state=42, max_iter=1000)

# 6. 在训练集上完成模型训练
model.fit(X_train_scaled, y_train)

# 7. 对测试集完成类别预测与正类概率预测
y_test_pred = model.predict(X_test_scaled)
y_test_score = model.predict_proba(X_test_scaled)[:, 1]

# 8. 计算并输出所有分类评估指标
print("="*60)
print(f"测试集分类准确率：{accuracy_score(y_test, y_test_pred):.4f}")
print("="*60)
print("\n=== 测试集完整分类报告 ===")
print(classification_report(y_test, y_test_pred, target_names=class_names))
print("="*60)
print("\n=== 测试集混淆矩阵 ===")
cm = confusion_matrix(y_test, y_test_pred)
print(cm)
print("="*60)
print(f"测试集ROC曲线AUC值：{roc_auc_score(y_test, y_test_score):.4f}")
print("="*60)

# 9. 绘制可视化图表
# 图表1：混淆矩阵热力图
plt.figure(figsize=(8, 6))
sns.heatmap(
    cm, annot=True, fmt="d", cmap="Blues",
    xticklabels=class_names, yticklabels=class_names,
    cbar=True, linewidths=0.5
)
plt.xlabel("预测类别", fontsize=12)
plt.ylabel("真实类别", fontsize=12)
plt.title("鸢尾花二分类任务混淆矩阵热力图", fontsize=14)
plt.tight_layout()
plt.show()

# 图表2：ROC曲线
plt.figure(figsize=(8, 6))
# 计算ROC曲线坐标
fpr, tpr, _ = roc_curve(y_test, y_test_score)
auc_value = roc_auc_score(y_test, y_test_score)
# 绘制ROC曲线
plt.plot(fpr, tpr, color="#1f77b4", linewidth=2, label=f"ROC曲线 (AUC={auc_value:.4f})")
# 绘制随机猜测参考线
plt.plot([0, 1], [0, 1], color="#ff7f0e", linewidth=2, linestyle="--", label="随机猜测 (AUC=0.5)")
# 填充曲线下区域
plt.fill_between(fpr, tpr, alpha=0.2, color="#1f77b4")
# 图表元素设置
plt.xlabel("假阳性率 (FPR)", fontsize=12)
plt.ylabel("真阳性率 (TPR)", fontsize=12)
plt.title("鸢尾花二分类任务ROC曲线", fontsize=14)
plt.legend(fontsize=12)
plt.grid(alpha=0.3)
plt.tight_layout()
plt.show()

# 10. 输出模型的特征权重与偏置项
print("\n=== 模型特征权重数值 ===")
for feature_name, weight in zip(feature_names, model.coef_[0]):
    print(f"{feature_name}: {weight:.6f}")

print("\n=== 模型偏置项（截距项）数值 ===")
print(f"偏置项: {model.intercept_[0]:.6f}")