python学习打卡day31

DAY 31 文件的规范拆分和写法
今日的示例代码包含2个部分

notebook文件夹内的ipynb文件，介绍下今天的思路
项目文件夹中其他部分：拆分后的信贷项目

知识点回顾

规范的文件命名
规范的文件夹管理
机器学习项目的拆分
编码格式和类型注解

**作业：**尝试针对之前的心脏病项目ipynb，将他按照今天的示例项目整理成规范的形式，思考下哪些部分可以未来复用。

python 复制代码

import pandas as pd 
import seaborn as sns
import numpy as np
from typing import Tuple, Dict
from sklearn.preprocessing import StandardScaler, MinMaxScaler
from sklearn.model_selection import train_test_split
from sklearn.ensemble import RandomForestClassifier  # 随机森林分类器
from sklearn.metrics import make_scorer, accuracy_score, precision_score, recall_score, f1_score  # 用于评估分类器性能的指标
from sklearn.metrics import classification_report, confusion_matrix  # 用于生成分类报告和混淆矩阵
import warnings  # 用于忽略警告信息
import os

def load_data(file_path: str) -> pd.DataFrame:
    """加载数据文件

    Args:
        file_path: 数据文件路径

    Returns:
        加载的数据框
    """
    return pd.read_csv(file_path)

def encode_categorical_features(data: pd.DataFrame) -> Tuple[pd.DataFrame, Dict]:
    """对分类特征进行编码

    Args:
        data: 原始数据框

    Returns:
        编码后的数据框和编码映射字典
    """
    discrete_features = ['sex', 'cp', 'fbs', 'restecg', 'exang', 'slope', 'thal']
    continuous_features = ['age', 'trestbps', 'chol', 'thalach', 'oldpeak']
    data_encoded = data.copy()
    # Purpose 独热编码
    data_encoded = pd.get_dummies(data, columns=discrete_features)
    return data_encoded

def normalization(data: pd.DataFrame) -> Tuple[pd.DataFrame, Dict]:
    """对连续特征进行归一化

    Args:
        data: 连续数据框

    Returns:
        归一化后的数据
    """
    continuous_features = ['age', 'trestbps', 'chol', 'thalach', 'oldpeak']
    data_scaled = data.copy()
    min_max_scaler = MinMaxScaler()
    data_scaled[continuous_features] = min_max_scaler.fit_transform(data_scaled[continuous_features])
    
    return data_scaled

if __name__ == "__main__":
    # 测试代码
    file_path = r"C:\Users\zwj\Desktop\python\Python60DaysChallenge-main\1\text\day31\data\raw\heart.csv"
    if not os.path.exists(file_path):
        print(f"文件不存在，请检查路径：{file_path}")
    else:
        data = load_data(file_path)
        data_encoded = encode_categorical_features(data)
        data_clean = normalization(data_encoded)
        print("数据预处理完成！")

train.py

python 复制代码

# -*- coding: utf-8 -*-

import sys
import os
sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))

from sklearn.ensemble import RandomForestClassifier
from sklearn.model_selection import train_test_split
from sklearn.metrics import classification_report, confusion_matrix
import time
import joblib # 用于保存模型
from typing import Tuple # 用于类型注解

from data.preprocessing import load_data, encode_categorical_features, normalization

def prepare_data() -> Tuple:
    """准备训练数据

    Returns:
        训练集和测试集的特征和标签
    """
    # 加载和预处理数据
    data = load_data(r"C:\Users\zwj\Desktop\python\Python60DaysChallenge-main\1\text\day31\data\raw\heart.csv")
    data_encoded = encode_categorical_features(data)
    data_scaled = normalization(data_encoded)
    
    # 分离特征和标签
    X = data_scaled.drop(['target'], axis=1)
    y = data_scaled['target']
    
    # 划分训练集和测试集
    X_train, X_test, y_train, y_test = train_test_split(
        X, y, test_size=0.2, random_state=42
    )
    
    return X_train, X_test, y_train, y_test

def train_model(X_train, y_train, model_params=None) -> RandomForestClassifier:
    """训练随机森林模型

    Args:
        X_train: 训练特征
        y_train: 训练标签
        model_params: 模型参数字典

    Returns:
        训练好的模型
    """
    if model_params is None:
        model_params = {'random_state': 42}
    
    model = RandomForestClassifier(**model_params)
    model.fit(X_train, y_train)
    return model

def evaluate_model(model, X_test, y_test) -> None:
    """评估模型性能

    Args:
        model: 训练好的模型
        X_test: 测试特征
        y_test: 测试标签
    """
    y_pred = model.predict(X_test)
    print("\n分类报告：")
    print(classification_report(y_test, y_pred))
    print("\n混淆矩阵：")
    print(confusion_matrix(y_test, y_pred))

def save_model(model, model_path: str) -> None:
    """保存模型

    Args:
        model: 训练好的模型
        model_path: 模型保存路径
    """
    os.makedirs(os.path.dirname(model_path), exist_ok=True)
    joblib.dump(model, model_path)
    print(f"\n模型已保存至: {model_path}")

if __name__ == "__main__":
    # 准备数据
    X_train, X_test, y_train, y_test = prepare_data()
    
    # 记录开始时间
    start_time = time.time()
    
    # 训练模型
    model = train_model(X_train, y_train)
    
    # 记录结束时间
    end_time = time.time()
    print(f"\n训练耗时: {end_time - start_time:.4f} 秒")
    
    # 评估模型
    evaluate_model(model, X_test, y_test)
    
    # 保存模型
    save_model(model, "models/random_forest_model.joblib")

plots.py

python 复制代码

import matplotlib.pyplot as plt
import seaborn as sns
import shap
import numpy as np
from typing import Any
import sys
import os
sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))

from sklearn.ensemble import RandomForestClassifier
from sklearn.model_selection import train_test_split
from sklearn.metrics import classification_report, confusion_matrix
import time
import joblib # 用于保存模型
from typing import Tuple # 用于类型注解
from data.preprocessing import load_data, encode_categorical_features, normalization
def plot_feature_importance_shap(model: Any, X_test, save_path: str = None) -> None:
    """绘制SHAP特征重要性图

    Args:
        model: 训练好的模型
        X_test: 测试数据
        save_path: 图片保存路径
    """
    # 初始化SHAP解释器
    explainer = shap.TreeExplainer(model)
    shap_values = explainer.shap_values(X_test)
    
    # 绘制特征重要性条形图
    plt.figure(figsize=(12, 8))
    shap.summary_plot(shap_values[:, :, 0], X_test, plot_type="bar", show=False)
    plt.title("SHAP特征重要性")
    
    if save_path:
        plt.savefig(save_path)
        print(f"特征重要性图已保存至: {save_path}")
    plt.show()

def plot_confusion_matrix(y_true, y_pred, save_path: str = None) -> None:
    """绘制混淆矩阵热力图

    Args:
        y_true: 真实标签
        y_pred: 预测标签
        save_path: 图片保存路径
    """
    plt.figure(figsize=(8, 6))
    cm = confusion_matrix(y_true, y_pred)
    sns.heatmap(cm, annot=True, fmt='d', cmap='Blues')
    plt.title('混淆矩阵')
    plt.ylabel('真实标签')
    plt.xlabel('预测标签')
    
    if save_path:
        plt.savefig(save_path)
        print(f"混淆矩阵图已保存至: {save_path}")
    plt.show()

def set_plot_style():
    """设置绘图样式"""
    #plt.style.use('seaborn')
    sns.set()  # 使用 seaborn 的样式
    plt.rcParams['font.sans-serif'] = ['SimHei']
    plt.rcParams['axes.unicode_minus'] = False

if __name__ == "__main__":
    # 设置绘图样式
    set_plot_style()
    # 准备数据和模型
    from sklearn.datasets import load_iris
    from sklearn.ensemble import RandomForestClassifier
    from sklearn.model_selection import train_test_split

    data = load_data(r"C:\Users\zwj\Desktop\python\Python60DaysChallenge-main\1\text\day31\data\raw\heart.csv")
    data_encoded = encode_categorical_features(data)
    data_scaled = normalization(data_encoded)
    
    # 分离特征和标签
    X = data_scaled.drop(['target'], axis=1)
    y = data_scaled['target']
    
    # 划分训练集和测试集
    X_train, X_test, y_train, y_test = train_test_split(
        X, y, test_size=0.2, random_state=42
    )
    # 训练模型
    model = RandomForestClassifier(random_state=42)
    model.fit(X_train, y_train)
    plot_feature_importance_shap(model, X_test)
    # 这里可以添加测试代码
    print("可视化模块加载成功！")

@浙大疏锦行