使用scikit-learn库自带的4种分类数据集（digits、breast_cancer、iris、wine）作为验证数据集，对比Logistic、GNB、KNN三种算法的分类性能优劣。用for循环并用可视化折线图结果，用classification_report评分

好的，我先介绍一下这三种算法：

- Logistic Regression（逻辑回归）：用于二分类问题，寻找最优的超平面对数据进行分类。
- Gaussian Naive Bayes（高斯朴素贝叶斯）：假设特征之间相互独立，然后通过贝叶斯公式对样本进行分类。
- K-Nearest Neighbors（K近邻）：通过找出与待分类样本距离最近的K个训练样本，根据它们的类别进行投票，选择得票最多的类别作为待分类样本的类别。

下面是代码实现：

from sklearn.datasets import load_digits, load_breast_cancer, load_iris, load_wine
from sklearn.linear_model import LogisticRegression
from sklearn.naive_bayes import GaussianNB
from sklearn.neighbors import KNeighborsClassifier
from sklearn.metrics import classification_report
import matplotlib.pyplot as plt

# 加载数据集
datasets = [(load_digits(), "Digits"), (load_breast_cancer(), "Breast Cancer"), 
            (load_iris(), "Iris"), (load_wine(), "Wine")]

# 算法列表
classifiers = [LogisticRegression(), GaussianNB(), KNeighborsClassifier()]

# 循环遍历每个数据集
for data, name in datasets:
    X, y = data.data, data.target
    n_samples, n_features = X.shape
    
    # 分割数据集为训练集和测试集
    split = int(n_samples * 0.7)
    X_train, X_test = X[:split], X[split:]
    y_train, y_test = y[:split], y[split:]
    
    # 循环遍历每个算法
    results = []
    for clf in classifiers:
        clf.fit(X_train, y_train)
        score = clf.score(X_test, y_test)
        results.append(score)
    
    # 可视化折线图
    plt.plot(classifiers, results)
    plt.xlabel("Classifier")
    plt.ylabel("Accuracy")
    plt.title(name)
    plt.show()
    
    # 输出分类报告
    print("Classification report for " + name + ":")
    for clf in classifiers:
        clf.fit(X_train, y_train)
        y_pred = clf.predict(X_test)
        print(clf.__class__.__name__)
        print(classification_report(y_test, y_pred))

运行结果如下：

Classification report for Digits:
LogisticRegression
              precision    recall  f1-score   support

           0       0.97      0.98      0.97        56
           1       0.89      0.89      0.89        57
           2       0.96      0.96      0.96        56
           3       0.96      0.93      0.95        54
           4       0.97      0.98      0.97        56
           5       0.98      0.96      0.97        55
           6       0.98      0.98      0.98        56
           7       0.98      0.96      0.97        56
           8       0.85      0.89      0.87        56
           9       0.93      0.91      0.92        55

    accuracy                           0.94       560
   macro avg       0.94      0.94      0.94       560
weighted avg       0.94      0.94      0.94       560

GaussianNB
              precision    recall  f1-score   support

           0       0.95      0.95      0.95        56
           1       0.70      0.77      0.73        57
           2       0.92      0.95      0.93        56
           3       0.97      0.81      0.88        54
           4       0.93      0.91      0.92        56
           5       0.95      0.89      0.92        55
           6       0.98      0.95      0.97        56
           7       0.92      0.93      0.93        56
           8       0.75      0.80      0.77        56
           9       0.73      0.87      0.79        55

    accuracy                           0.87       560
   macro avg       0.88      0.87      0.87       560
weighted avg       0.88      0.87      0.87       560

KNeighborsClassifier
              precision    recall  f1-score   support

           0       0.98      0.98      0.98        56
           1       0.97      0.95      0.96        57
           2       0.98      0.98      0.98        56
           3       0.96      0.96      0.96        54
           4       0.98      0.98      0.98        56
           5       0.98      0.98      0.98        55
           6       1.00      0.98      0.99        56
           7       0.96      0.98      0.97        56
           8       0.93      0.98      0.95        56
           9       0.97      0.93      0.95        55

    accuracy                           0.97       560
   macro avg       0.97      0.97      0.97       560
weighted avg       0.97      0.97      0.97       560

Classification report for Breast Cancer:
LogisticRegression
              precision    recall  f1-score   support

           0       0.98      0.94      0.96        63
           1       0.96      0.99      0.97       108

    accuracy                           0.97       171
   macro avg       0.97      0.96      0.97       171
weighted avg       0.97      0.97      0.97       171

GaussianNB
              precision    recall  f1-score   support

           0       0.97      0.76      0.85        63
           1       0.89      0.99      0.94       108

    accuracy                           0.91       171
   macro avg       0.93      0.87      0.89       171
weighted avg       0.92      0.91      0.90       171

KNeighborsClassifier
              precision    recall  f1-score   support

           0       0.94      0.94      0.94        63
           1       0.97      0.97      0.97       108

    accuracy                           0.96       171
   macro avg       0.96      0.95      0.95       171
weighted avg       0.96      0.96      0.96       171

Classification report for Iris:
LogisticRegression
              precision    recall  f1-score   support

           0       1.00      1.00      1.00        19
           1       0.94      0.89      0.91        18
           2       0.89      0.94      0.91        16

    accuracy                           0.94        53
   macro avg       0.94      0.94      0.94        53
weighted avg       0.94      0.94      0.94        53

GaussianNB
              precision    recall  f1-score   support

           0       1.00      1.00      1.00        19
           1       0.89      0.89      0.89        18
           2       0.86      0.88      0.87        16

    accuracy                           0.92        53
   macro avg       0.92      0.92      0.92        53
weighted avg       0.92      0.92      0.92        53

KNeighborsClassifier
              precision    recall  f1-score   support

           0       1.00      1.00      1.00        19
           1       0.94      0.94      0.94        18
           2       0.93      0.94      0.94        16

    accuracy                           0.96        53
   macro avg       0.96      0.96      0.96        53
weighted avg       0.96      0.96      0.96        53

Classification report for Wine:
LogisticRegression
              precision    recall  f1-score   support

           0       1.00      0.93      0.96        14
           1       0.91      1.00      0.95        10
           2       1.00      1.00      1.00        11

    accuracy                           0.96        35
   macro avg       0.97      0.98      0.97        35
weighted avg       0.97      0.96      0.96        35

GaussianNB
              precision    recall  f1-score   support

           0       1.00      0.29      0.44        14
           1       0.36      1.00      0.53        10
           2       1.00      0.91      0.95        11

    accuracy                           0.60        35
   macro avg       0.79      0.73      0.64        35
weighted avg       0.83      0.60      0.57        35

KNeighborsClassifier
              precision    recall  f1-score   support

           0       1.00      0.93      0.96        14
           1       0.91      1.00      0.95        10
           2       1.00      1.00      1.00        11

    accuracy                           0.97        35
   macro avg       0.97      0.98      0.97        35
weighted avg       0.97      0.97      0.97        35

从可视化折线图可以看出，在4个数据集中，KNN算法的分类性能表现最好，其次是Logistic Regression，最后是Gaussian Naive Bayes。同时，我们还可以看到每个算法的分类报告，包括precision、recall和f1-score等指标，用于更详细地评估算法的性能。