感知机与adaline算法

原理

相对于感知机,Adaline算法有趣的多,因为在学习Adaline的过程中涉及到机器学习中一个重要的概念:定义、最小化损失函数。学习Adaline为以后学习更复杂高端的算法(比如逻辑斯蒂回归、SVM等)起到抛砖引玉的作用。
Adaline和感知机的一个重要区别是Adaline算法中权重参数更新按照线性激活函数而不是单位阶跃函数。

算法对比

感知机

import numpy as npclass Perceptron(object):    def __init__(self,eta = 0.1,n_iter = 10):        self.eta = eta        self.n_iter = n_iter    def fit(self,X,y):        self.w_ = np.zeros(1+X.shape[1])        self.errors = []        for _ in range(self.n_iter):            errors = 0            for feature, target in zip(X, y):                update =target - self.predict(feature)                self.w_[1:] += update*feature                self.w_[0]  += update                errors += int(update != 0.0)            self.errors.append(errors)    def net_input(self,X):        return np.dot(X,self.w_[1:]+self.w_[0])    def predict(self,X):        return np.where(self.net_input(X)>0.0,1,-1)

adaline算法

import numpy as npclass Adaline(object):    def __init__(self,eta = 0.1,n_iter = 10):        self.eta = eta        self.n_iter = n_iter    def fit(self,X,y):        self.w_ = np.zeros(1+X.shape[1])        self.cost_ = []        for i in range(self.n_iter):            output = self.predit(X)            errors = y - output            self.w_[1:] = self.eta * X.T.dot(errors)            self.w_[0] = self.eta * errors.sum()            cost = (errors ** 2).sum()/2.0            self.cost_.append(cost)        return self    def predit(self,X):        return np.dot(X,self.w_[1:])+self.w_[0]    def predict(self, X):        return np.where(self.predit(X) > 0.0, 1, -1)