机器学习实战专题（三）LogisticRegression

1.理论快速回顾

1.1 模型

当y>0.5时，z为TRUE，else 为False

模型输出时，根据以上：1--TRUE，0--FALSE

1.2 损失函数

 

1.3 优化：梯度下降法

求导：

向量化

2.实战

2.1 实现逻辑回归类

​
import numpy as np
from sklearn.metrics import accuracy_score

class LogisticRegression(object):
    
    def __init__(self):
        """  初始化Logistic Regression模型"""
        self.coef=None
        self.intercept=None
        self._theta=None
        
    def sigmoid(self,t):
        return 1./(1.+np.exp(-t))
    def fit(self,X_train,y_train,eta=0.01,n_iters=1e4):
        '''使用梯度下降法训练logistic Regression模型 '''
        assert X_train.shape[0]==y_train.shape[0]
        
        ''' 定义损失函数'''
        
        def J(theta,X_b,y):
            y_hat=self.sigmoid(X_b.dot(theta))
            try:
                return -np.sum(y*np.log(y_hat)+(1-y)*np.log(1-y_hat))/len(y)
            except:
                return float('inf')
            
        ''' 损失函数求导'''
        def dJ(theta,X_b,y):
            '''向量化后的公式 '''
            return X_b.T.dot(self.sigmoid(X_b.dot(theta))-y)/len(y)
        
        '''梯度下降函数'''
        def gradient_descent(X_b,y,initial_theta,eta,n_iters=1e4,epsilon=1e-8):
            theta=initial_theta
            cur_iter=0
            
            while cur_iter < n_iters:
                gradient=dJ(theta,X_b,y)
                last_theta=theta
                theta=theta-eta*gradient
                if abs(J(theta,X_b,y)-J(last_theta,X_b,y))<epsilon:
                    break
                cur_iter+=1
            return theta
        X_b=np.hstack([np.ones((len(X_train),1)),X_train])
        initial_theta=np.zeros(X_b.shape[1])
        self._theta=gradient_descent(X_b,y_train,initial_theta,eta,n_iters)
        
        #截距
        self.intercept=self._theta[0]
        # x_i前的参数
        self.coef=self._theta[1:]
        return self
    
    ''' 预测'''
    def predict_proba(self,X_predict):
        '''给定待预测数据集X_predict，返回表示X_predict的结果概率向量 '''
        assert self.intercept is not None and self.coef is not None
        '''must fit before predict '''
        assert X_predict.shape[1]==len(self.coef)
        '''the feature number of X_predict must be equal to X_train '''
        X_b=np.hstack([np.ones((len(X_predict),1)),X_predict])
        return self.sigmoid(X_b.dot(self._theta))   
    
    def predict(self,X_predict):
        '''给定待预测数据集X_predict，返回表示X_predict的结果向量 '''
        assert self.intercept is not None and self.coef is not None
        assert X_predict.shape[1]==len(self.coef)
        prob=self.predict_proba(X_predict)
        return np.array(prob>=0.5,dtype='int')
    
    def score(self,X_test,y_test):
        """根据测试数据集X_test和y_test确定当前模型的准确度"""
        y_predict=self.predict(X_test)
        return accuracy_score(y_test,y_predict)
    def __repr__(self):
        return "LogisticRegression()"

​

2.2用小数据集进行测试

import matplotlib.pyplot as plt
from sklearn import datasets
from sklearn.model_selection import train_test_split

#取鸢尾花数据集
iris=datasets.load_iris()
X=iris.data
y=iris.target
# 筛选特征
X=X[y<2,:2]
y=y[y<2]
#绘制出图像
plt.scatter(X[y==0,0],X[y==0,1],color='red')
plt.scatter(X[y==1,0],X[y==1,1],color='blue')
plt.show()

# 切分数据集
X_train,X_test,y_train,y_test=train_test_split(X,y,random_state=666)

# 调用我们自己的逻辑回归函数
log_reg=LogisticRegression()
log_reg.fit(X_train,y_train)
print('final score is :%s'% log_reg.score(X_test,y_test))
print('actual prob is:')
print(log_reg.predict_proba(X_test))