服务器之家

服务器之家 > 正文

python实现神经网络感知器算法

时间:2020-12-27 00:10     来源/作者:海峰-清欢

现在我们用python代码实现感知器算法

?
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
# -*- coding: utf-8 -*-
import numpy as np
 
 
class Perceptron(object):
 """
 eta:学习率
 n_iter:权重向量的训练次数
 w_:神经分叉权重向量
 errors_:用于记录神经元判断出错次数
 """
 
 def __init__(self, eta=0.01, n_iter=2):
  self.eta = eta
  self.n_iter = n_iter
  pass
 
 def fit(self, X, y):
  """
  输入训练数据培训神经元
  X:神经元输入样本向量
  y: 对应样本分类
  X:shape[n_samples,n_features]
  x:[[1,2,3],[4,5,6]]
  n_samples = 2 元素个数
  n_features = 3 子向量元素个数
  y:[1,-1]
  初始化权重向量为0
  加一是因为前面算法提到的w0,也就是步调函数阈值
  """
  self.w_ = np.zeros(1 + X.shape[1])
  self.errors_ = []
  for _ in range(self.n_iter):
   errors = 0
   """
   zip(X,y) = [[1,2,3,1],[4,5,6,-1]]
   xi是前面的[1,2,3]
   target是后面的1
   """
   for xi, target in zip(X, y):
    """
    predict(xi)是计算出来的分类
    """
    update = self.eta * (target - self.predict(xi))
    self.w_[1:] += update * xi
    self.w_[0] += update
    print update
    print xi
    print self.w_
    errors += int(update != 0.0)
    self.errors_.append(errors)
    pass
 
 def net_input(self, X):
  """
  z = w0*1+w1*x1+....Wn*Xn
  """
  return np.dot(X, self.w_[1:]) + self.w_[0]
 
 def predict(self, X):
  return np.where(self.net_input(X) >= 0, 1, -1)
 
 
if __name__ == '__main__':
 datafile = '../data/iris.data.csv'
 import pandas as pd
 
 df = pd.read_csv(datafile, header=None)
 import matplotlib.pyplot as plt
 import numpy as np
 
 y = df.loc[0:100, 4].values
 y = np.where(y == "Iris-setosa", 1, -1)
 X = df.iloc[0:100, [0, 2]].values
 # plt.scatter(X[:50, 0], X[:50, 1], color="red", marker='o', label='setosa')
 # plt.scatter(X[50:100, 0], X[50:100, 1], color="blue", marker='x', label='versicolor')
 # plt.xlabel("hblength")
 # plt.ylabel("hjlength")
 # plt.legend(loc='upper left')
 # plt.show()
 
 pr = Perceptron()
 pr.fit(X, y)

其中数据为

python实现神经网络感知器算法

控制台输出为

python实现神经网络感知器算法

你们跑代码的时候把n_iter设置大点,我这边是为了看每次执行for循环时方便查看数据变化。

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服务器之家。

原文链接:http://blog.csdn.net/u013692888/article/details/76999252

相关文章

热门资讯

2020微信伤感网名听哭了 让对方看到心疼的伤感网名大全
2020微信伤感网名听哭了 让对方看到心疼的伤感网名大全 2019-12-26
Intellij idea2020永久破解,亲测可用!!!
Intellij idea2020永久破解,亲测可用!!! 2020-07-29
歪歪漫画vip账号共享2020_yy漫画免费账号密码共享
歪歪漫画vip账号共享2020_yy漫画免费账号密码共享 2020-04-07
背刺什么意思 网络词语背刺是什么梗
背刺什么意思 网络词语背刺是什么梗 2020-05-22
电视剧《琉璃》全集在线观看 琉璃美人煞1-59集免费观看地址
电视剧《琉璃》全集在线观看 琉璃美人煞1-59集免费观看地址 2020-08-12
返回顶部