使用Scikit-learn实现逻辑回归

最近一边看模型，一边学习数学，还是自己实现一下好点

1. 导入数据集，可以直接使用Scikit-learn自带的胸癌数据集

from sklearn.datasets import load_breast_cancer

breast_cancer = load_breast_cancer()
samples = breast_cancer.data   #  【样本】(或者也叫【示例】)
label = breast_cancer.target   #  【标记】

2. 划分数据集

from sklearn.model_selection import train_test_split
# 将矩阵随机划分成训练集和测试集, test_size表示测试集的比例(即训练集：测试集=7:3)
sample_train, sample_test, label_train, label_test = train_test_split(samples, label, test_size=0.3)

3. 导入回归模型并训练

from sklearn.linear_model import LogisticRegression
classifier = LogisticRegression()  
classifier.fit(sample_train, label_train)  # 训练分类模型(Fit the model according to the given training data)

4. 查看准确率

count = 0
Length = len(label_test)

for i in range(Length):
    if classifier.predict(sample_test)[i] != label_test[i]:  # 预测测试样本
        count += 1

5. 所有代码如下：

from sklearn.datasets import load_breast_cancer
from sklearn.model_selection import train_test_split
from sklearn.linear_model import LogisticRegression

# load the dataset: breast_cancer
breast_cancer = load_breast_cancer()
samples = breast_cancer.data
label = breast_cancer.target

# 将矩阵随机划分成训练集和测试集, test_size表示测试集的比例(即训练集：测试集=7:3)
sample_train, sample_test, label_train, label_test = train_test_split(samples, label, test_size=0.3)

classifier = LogisticRegression()  # 使用类，参数全是默认的

classifier.fit(sample_train, label_train)  # 训练分类模型(Fit the model according to the given training data)

count = 0
Length = len(label_test)

for i in range(Length):
    if classifier.predict(sample_test)[i] != label_test[i]:  # 预测测试样本
        count += 1

6. 结果输出：（随机划分的训练集和测试集，所以每次预测效果可能都不一样）

8
0.9532163742690059