学习笔记：使用Logistic回归和随机梯队下降预测良/恶性肿瘤

# -*- coding: utf-8 -*-
"""
Created on Wed Apr 25 17:07:55 2018

@author: eagle
"""

import pandas as pd
import numpy as np
import requests
import io
from sklearn.cross_validation import train_test_split
from sklearn.preprocessing import StandardScaler
from sklearn.linear_model import LogisticRegression
from sklearn.linear_model import SGDClassifier
from sklearn.metrics import classification_report


#创建特征列表
column_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']

#使用pandas.read_csv函数从互联网读取指定数据
data = pd.read_csv('https://archive.ics.uci.edu/ml/machine-learning-databases/breast-cancer-wisconsin/breast-cancer-wisconsin.data',names=column_names)
#print(data.shape)
#将?替换为标准缺失值null
data = data.replace('?',value = np.nan)
#丢弃缺失值(只要有一个维度上缺失)
data = data.dropna(how='any')
#查看data的数据量和维度
#data.shape

#准备训练、测试数据

#随机采用，25%用于测试，75%用于训练
X_train,X_test,y_train,y_test = train_test_split(data[column_names[1:10]],data[column_names[10]],test_size = 0.25,random_state = 33)
#训练数据类别
class_train = y_train.value_counts()
#测试数据类别
class_test = y_test.value_counts()
#print(data.shape)
#print(data)
#将?替换为标准缺失值
#data=data.replace(to_replace='?',np.nan)

# =============================================================================
# 训练及预测
# =============================================================================

#标准化数据,保证每个维度的特征数据方差为1，均值为0
ss =  StandardScaler()
X_train = ss.fit_transform(X_train)
X_test = ss.transform(X_test)

#初始化Logistic回归和SGDClassifier
lr = LogisticRegression()
sgdc = SGDClassifier()
#使用LogisticRegression训练模型参数并预测
lr.fit(X_train,y_train)
lr_y_predict = lr.predict(X_test)

#使用SGDClassfier训练并预测
sgdc.fit(X_train,y_train)
sgdc_y_predict = sgdc.predict(X_test)

# =============================================================================
# 性能分析（准确率、召回率）
# =============================================================================


#获得准确率（预测准确/所有样本）
print ('Accuracy of LR Classifier:',lr.score(X_test,y_test))
#获取其他三个指标(召回率、精确率及F1)
#召回率=真阳/(真阳+假阴)
#精确率=真阳/(真阳+假阳)
#F1=2/(1/召回率+1/精确率)
print(classification_report(y_test,lr_y_predict,target_names=['Benign','Malignant']))

print ('Accuracy of SGD Classifier:',sgdc.score(X_test,y_test))
#获取其他三个指标(召回率、精确率及F1)
#召回率=真阳/(真阳+假阴)
#精确率=真阳/(真阳+假阳)
#F1=2/(1/召回率+1/精确率)
print(classification_report(y_test,sgdc_y_predict,target_names=['Benign','Malignant']))

上述是实现《PYTHON机器学习及实践》书中的代码，不过从运行结果看，使用随机梯队下降，每次运行的结果是有差别的。