感知器及其在python中的实现_感知器基础原理及python实现过程详解

阳光穿透心脏的1/2处 2023-01-09 01:28 89阅读 0赞

简单版本，按照李航的《统计学习方法》的思路编写

![Image 1][]

数据采用了著名的sklearn自带的iries数据，最优化求解采用了SGD算法。

预处理增加了标准化操作。

'''

perceptron classifier

created on 2019.9.14

author: vince

'''

import pandas

import numpy

import logging

import matplotlib.pyplot as plt

from sklearn.datasets import load\_iris

from sklearn.model\_selection import train\_test\_split

from sklearn.metrics import accuracy\_score

'''

perceptron classifier

Attributes

w: ld-array = weights after training

l: list = number of misclassification during each iteration

'''

class Perceptron:

def \_\_init\_\_(self,eta = 0.01,iter\_num = 50,batch\_size = 1):

'''

eta: float = learning rate (between 0.0 and 1.0).

iter\_num: int = iteration over the training dataset.

batch\_size: int = gradient descent batch number,if batch\_size == 1,used SGD;

if batch\_size == 0,use BGD;

else MBGD;

'''

self.eta = eta;

self.iter\_num = iter\_num;

self.batch\_size = batch\_size;

def train(self,X,Y):

'''

train training data.

X:\{array-like\},shape=\[n\_samples,n\_features\] = Training vectors,where n\_samples is the number of training samples and

n\_features is the number of features.

Y:\{array-like\},share=\[n\_samples\] = traget values.

'''

self.w = numpy.zeros(1 + X.shape\[1\]);

self.l = numpy.zeros(self.iter\_num);

for iter\_index in range(self.iter\_num):

for sample\_index in range(X.shape\[0\]):

if (self.activation(X\[sample\_index\]) != Y\[sample\_index\]):

logging.debug("%s: pred(%s),label(%s),%s,%s" % (sample\_index,self.net\_input(X\[sample\_index\]),Y\[sample\_index\],X\[sample\_index,0\],1\]));

self.l\[iter\_index\] += 1;

for sample\_index in range(X.shape\[0\]):

if (self.activation(X\[sample\_index\]) != Y\[sample\_index\]):

self.w\[0\] += self.eta \* Y\[sample\_index\];

self.w\[1:\] += self.eta \* numpy.dot(X\[sample\_index\],Y\[sample\_index\]);

break;

logging.info("iter %s: %s,%s" %

(iter\_index,self.w\[0\],self.w\[1\],self.w\[2\],self.l\[iter\_index\]));

def activation(self,x):

return numpy.where(self.net\_input(x) >= 0.0,1,-1);

def net\_input(self,x):

return numpy.dot(x,self.w\[1:\]) + self.w\[0\];

def predict(self,x):

return self.activation(x);

def main():

logging.basicConfig(level = logging.INFO,format = '%(asctime)s %(filename)s\[line:%(lineno)d\] %(levelname)s %(message)s',datefmt = '%a,%d %b %Y %H:%M:%S');

iris = load\_iris();

features = iris.data\[:99,\[0,2\]\];

\# normalization

features\_std = numpy.copy(features);

for i in range(features.shape\[1\]):

features\_std\[:,i\] = (features\_std\[:,i\] - features\[:,i\].mean()) / features\[:,i\].std();

labels = numpy.where(iris.target\[:99\] == 0,-1,1);

\# 2/3 data from training,1/3 data for testing

train\_features,test\_features,train\_labels,test\_labels = train\_test\_split(

features\_std,labels,test\_size = 0.33,random\_state = 23323);

logging.info("train set shape:%s" % (str(train\_features.shape)));

p = Perceptron();

p.train(train\_features,train\_labels);

test\_predict = numpy.array(\[\]);

for feature in test\_features:

predict\_label = p.predict(feature);

test\_predict = numpy.append(test\_predict,predict\_label);

score = accuracy\_score(test\_labels,test\_predict);

logging.info("The accruacy score is: %s "% (str(score)));

\#plot

x\_min,x\_max = train\_features\[:,0\].min() - 1,train\_features\[:,0\].max() + 1;

y\_min,y\_max = train\_features\[:,1\].min() - 1,1\].max() + 1;

plt.xlim(x\_min,x\_max);

plt.ylim(y\_min,y\_max);

plt.xlabel("width");

plt.ylabel("heigt");

plt.scatter(train\_features\[:,1\],c = train\_labels,marker = 'o',s = 10);

k = - p.w\[1\] / p.w\[2\];

d = - p.w\[0\] / p.w\[2\];

plt.plot(\[x\_min,x\_max\],\[k \* x\_min + d,k \* x\_max + d\],"go-");

plt.show();

if \_\_name\_\_ == "\_\_main\_\_":

main();

![Image 1][]

以上就是本文的全部内容，希望对大家的学习有所帮助，也希望大家多多支持我们。

[Image 1]:

感知器及其在python中的实现_感知器基础原理及python实现过程详解

发表评论取消回复

还没有评论，来说两句吧...

相关阅读

相关 Python中感知器算法的完整指南

相关 Python编写多层感知器分类器算法

相关感知器及其在python中的实现_感知器基础原理及python实现过程详解

相关机器学习_感知器算法学习(Python实现)

相关感知器Perceptron

相关感知哈希原理及实现

相关 Python Machine Learning - 感知器算法

相关机器学习二：感知器算法实现

相关机器学习（二）：感知器算法剖析及Python实现

相关 Rosenblatt感知器