[已解决]人工智能中关于逻辑回归问题

isdkz

1、基于Sklearn机器学习工具或自编码实现经典逻辑回归（无正则化）、带有L1正则化和L2正则化的逻辑回归方法；并通过调整学习参数，比较其对分类精度的影响。



经典逻辑回归（无正则化）的实现：

1. from sklearn.linear_model import LogisticRegression
   
   
2. from sklearn.datasets import load_iris
   
   
3. from sklearn.model_selection import train_test_split
   
   
4. 
   
   
5. # 加载数据集
   
   
6. iris = load_iris()
   
   
7. X = iris.data[:, :2]
   
   
8. y = iris.target
   
   
9. 
   
   
10. # 划分数据集
    
    
11. X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)
    
    
12. 
    
    
13. # 构建逻辑回归模型
    
    
14. lr = LogisticRegression(penalty='none', solver='lbfgs', max_iter=1000)
    
    
15. 
    
    
16. # 训练模型
    
    
17. lr.fit(X_train, y_train)
    
    
18. 
    
    
19. # 预测测试集
    
    
20. y_pred = lr.predict(X_test)
    
    
21. 
    
    
22. # 计算准确率
    
    
23. accuracy = lr.score(X_test, y_test)
    
    
24. print("Accuracy:", accuracy)
    
    

复制代码

带有L1正则化的逻辑回归方法实现：

1. from sklearn.linear_model import LogisticRegression
   
   
2. from sklearn.datasets import load_iris
   
   
3. from sklearn.model_selection import train_test_split
   
   
4. 
   
   
5. # 加载数据集
   
   
6. iris = load_iris()
   
   
7. X = iris.data[:, :2]
   
   
8. y = iris.target
   
   
9. 
   
   
10. # 划分数据集
    
    
11. X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)
    
    
12. 
    
    
13. # 构建逻辑回归模型
    
    
14. lr = LogisticRegression(penalty='l1', solver='saga', max_iter=1000)
    
    
15. 
    
    
16. # 训练模型
    
    
17. lr.fit(X_train, y_train)
    
    
18. 
    
    
19. # 预测测试集
    
    
20. y_pred = lr.predict(X_test)
    
    
21. 
    
    
22. # 计算准确率
    
    
23. accuracy = lr.score(X_test, y_test)
    
    
24. print("Accuracy:", accuracy)
    
    

复制代码

带有L2正则化的逻辑回归方法实现：

1. from sklearn.linear_model import LogisticRegression
   
   
2. from sklearn.datasets import load_iris
   
   
3. from sklearn.model_selection import train_test_split
   
   
4. 
   
   
5. # 加载数据集
   
   
6. iris = load_iris()
   
   
7. X = iris.data[:, :2]
   
   
8. y = iris.target
   
   
9. 
   
   
10. # 划分数据集
    
    
11. X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)
    
    
12. 
    
    
13. # 构建逻辑回归模型
    
    
14. lr = LogisticRegression(penalty='l2', solver='lbfgs', max_iter=1000)
    
    
15. 
    
    
16. # 训练模型
    
    
17. lr.fit(X_train, y_train)
    
    
18. 
    
    
19. # 预测测试集
    
    
20. y_pred = lr.predict(X_test)
    
    
21. 
    
    
22. # 计算准确率
    
    
23. accuracy = lr.score(X_test, y_test)
    
    
24. print("Accuracy:", accuracy)
    
    

复制代码

通过调整学习参数，比较其对分类精度的影响：

可以通过调整学习率、正则化系数等参数来影响分类精度。例如，可以使用GridSearchCV来搜索最优的参数组合：

1. from sklearn.model_selection import GridSearchCV
   
   
2. 
   
   
3. # 定义参数范围
   
   
4. param_grid = {'penalty': ['none', 'l1', 'l2'],
   
   
5.               'C': [0.001, 0.01, 0.1, 1, 10, 100],
   
   
6.               'solver': ['lbfgs', 'saga']}
   
   
7. 
   
   
8. # 构建逻辑回归模型
   
   
9. lr = LogisticRegression(max_iter=1000)
   
   
10. 
    
    
11. # 使用GridSearchCV搜索最优参数
    
    
12. grid_search = GridSearchCV(estimator=lr, param_grid=param_grid, cv=5, n_jobs=-1)
    
    
13. grid_search.fit(X_train, y_train)
    
    
14. 
    
    
15. # 输出最优参数和准确率
    
    
16. print("Best parameters:", grid_search.best_params_)
    
    
17. print("Accuracy:", grid_search.best_score_)
    
    

复制代码

2、数据集：Iris数据集（选择两个特征，自己划分训练集和测试集）；



加载数据集并选择两个特征：

1. from sklearn.datasets import load_iris
   
   
2. from sklearn.model_selection import train_test_split
   
   
3. 
   
   
4. # 加载数据集
   
   
5. iris = load_iris()
   
   
6. X = iris.data[:, :2]
   
   
7. y = iris.target
   
   
8. 
   
   
9. # 划分数据集
   
   
10. X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)
    
    

复制代码

自己划分训练集和测试集，可以使用train_test_split函数。