感知机简单python实现

参考链接

from functools import reduce
from itertools import tee

class Perceptron:
	def __init__(self,input_num,activator):
		self.activator = activator
		self.weights = [0.0 for _ in range(input_num)]
		self.bias = 0.0
	def __str__(self):
		return 'weights\t:%s\nbias\t:%f\n' % (self.weights, self.bias)
	def predict(self, input_vec):
		return self.activator(\
			reduce(lambda a,b:a+b,\
				map(lambda x_w:x_w[0]*x_w[1], \
					zip(input_vec,self.weights)) \
					,0.0)+self.bias)

	def _update_weight(self,input_vec,output,label,rate):
		delta = label - output
		self.weights = list(map(
		lambda x_w:x_w[1]+rate*delta*x_w[0],
		zip(input_vec,self.weights)))
		self.bias += rate * delta

	def _one_iteration(self,input_vecs,labels,rate):
		samples = zip(input_vecs,labels)
		#print(list(samples))
		for(input_vec,label) in samples:
			output = self.predict(input_vec)
			self._update_weight(input_vec,output,label,rate)

	def train(self,input_vecs,labels,iteration,rate):
		for i in range(iteration):
			self._one_iteration(input_vecs,labels,rate)


def and_f(x):
	return 1 if x>0 else 0

if __name__ == '__main__':
	x = [[0,0],[0,1],[1,0],[1,1]]
	y = [0,0,0,1]
	and_perception = Perceptron(2,and_f)
	and_perception.train(x,y,10,0.1)
	print(and_perception)
	print('1 and 1 = %d'%and_perception.predict([1,1]))
	print('0 and 0 = %d'%and_perception.predict([0,0]))
	print('1 and 0 = %d'%and_perception.predict([1,0]))
	print('0 and 1 = %d'%and_perception.predict([0,1]))

对应可视化

from matplotlib import pyplot
from matplotlib import cm
from mpl_toolkits.mplot3d import Axes3D
import numpy as np

if __name__ == '__main__':
	# 训练数据
	x = np.array([[0,0,0],[1,1,1],[1,1,0],[1,0,1],[0,1,1]])
	y = [1,0,0,0,0]
	and_perception = Perceptron(3,and_f)
	and_perception.train(x,y,10,0.1)
	print(and_perception)
	fig = pyplot.figure()
	pyplot.xlim(0, 1)
	pyplot.ylim(0, 1)
	ax = Axes3D(fig)
	ax.set_zlim3d(0,1)
	length = len(y)
	for i in range(length):
		if y[i]==1:
			ax.scatter(x[0:,0:1][i],x[0:,1:2][i],x[0:,2:3][i], c = 'r', marker='*')
		else:
			ax.scatter(x[0:,0:1][i],x[0:,1:2][i],x[0:,2:3][i], c = 'b', marker='*')

	x2 = np.linspace(0,1,10)
	y2 = np.linspace(0,1,10)
	X,Y = np.meshgrid(x2,y2)
	Z = (and_perception.weights[0]*X + and_perception.weights[1]*Y +  + and_perception.bias)/(-and_perception.weights[2])
	for idx1, zz in enumerate(Z):
		for idx2, zzz in enumerate(zz):
			if Z[idx1][idx2]<0 :			
				Z[idx1][idx2] = and_perception.bias/(-and_perception.weights[2])#Z[idx1][idx2] if Z[idx1][idx2]>0 else np.nan
				X[idx1][idx2] = 0
				Y[idx1][idx2] = 0
	surf = ax.plot_surface(X, Y, Z, rstride = 1, cstride = 1,cmap = 'viridis', edgecolor = 'none',vmin=np.nanmin(Z), vmax=np.nanmax(Z))
	#使用模型对100个随机三维点进行分类
	test = np.random.randint(0,1000,(100,3))/1000
	for t in test:
		if and_perception.predict(t)==1:
			ax.scatter(t[0],t[1],t[2], c = 'r', marker='*')
		else:
			ax.scatter(t[0],t[1],t[2], c = 'b', marker='*')

	ax.set_xlabel('X-axis')
	ax.set_ylabel('Y-axis')
	ax.set_zlabel('Z-axis')
	pyplot.show()