机器学习实现逻辑回归-癌症分类预测

1.数据来源

复制代码

https://archive.ics.uci.edu/ml/machine-learning-databases/breast-cancer-wisconsin/breast-cancer-wisconsin.data

2.代码示例(已添加注释)

python 复制代码

import pandas as pd
import numpy as np
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import StandardScaler
from sklearn.linear_model import LogisticRegression,Ridge
from sklearn.metrics import classification_report,roc_auc_score
#给列名字
names = ['Sample code number', 'Clump Thickness', 'Uniformity of Cell Size', 'Uniformity of Cell Shape','Marginal Adhesion', 'Single Epithelial Cell Size', 'Bare Nuclei', 'Bland Chromatin','Normal Nucleoli', 'Mitoses', 'Class']

data = pd.read_csv('https://archive.ics.uci.edu/ml/machine-learning-databases/breast-cancer-wisconsin/breast-cancer-wisconsin.data',names = names)

##数据预处理:去除以？表示的缺失值
data.replace('?',np.nan,inplace=True)
data.dropna(inplace=True)


##目标值和特征值
tezheng=data.iloc[:,1:10]
mubiao=data['Class']

##划分数据
x_train,x_test,y_train,y_test=train_test_split(tezheng,mubiao,random_state=22)

##特征工程:标准化
trans=StandardScaler()
x_train=trans.fit_transform(x_train)
x_test=trans.transform(x_test)

##生成模型
em = LogisticRegression(solver='sag')
em.fit(x_train,y_train)
y_pre=em.predict(x_test)
print(em.coef_,em.intercept_)
print(em.score(x_test,y_test))

##模型评估
print(y_test==y_pre)
report=classification_report(y_test,y_pre,labels=[2,4],target_names=['良性','恶性'])##求解准确率与召回率
print(report)
y_test1=np.where(y_test>3,1,0)##roc_auc函数的第一个参数要求必须是1，0分类才行，且1表示正例
print(roc_auc_score(y_test1,y_pre))##计算AUC指标，确定模型是否受样本不均衡影响，越接近1，越好，越接近0.5越差
##注意精确率，召回率，ROC_AUC指标是评估二分类的模型性能的指标