效果展示
向网络输入测试集中‘7’并绘出该输入图,得到10个标签的概率,7对应的标签概率最高,效果很好
全连接神经网络
训练测试集
mnist训练测试集,是一个手写数字识别库,世界上最权威的,美国邮政系统开发的,手写内容是0-9的内容,手写内容采集于美国人口调查局的员工和高中生
本例程中要用到的训练测试集链接:
表格形式(CSV)的mnist训练测试集,大部分电子表格和数据分析软件兼容形式(免费)
包括mnist_test.csv、mnist_train.csv、mnist_test_10.csv、mnist_train_100.csv、mnist_train.csv、mnist_test.csv分别有60000、10000个标记样本集;mnist_test_10.csv、mnist_train_100.csv则只有10条100条记录是上面的子集,在深入研究前我们常用子集验证算法再用完整集
代码说明
1、参考了Python神经网络编程([英]Tariq Rashid著) 的部分代码
2、class neuralNetwork:包含了初始化,训练,查询3个部分。初始化要初始权重wih(输入层隐藏层连接权重,who隐藏层输出层连接权重),各层的节点个数,激活函数
3、n = neuralNetwork(input_nodes,hidden_nodes,output_nodes,learning_rate)生成一个网络,input_nodes = 784、hidden_nodes = 200、output_nodes = 10、learning_rate = 0.3为网络参数,其中784=28x28,是输入像素,10是0~9共10个标签
4、循环更新权重,训练网络
for record in training_data_list:
xxxxxxxxxxxxxxx
pass
5、print(n.query((numpy.asfarray(all_values[1:]) / 255.0 * 0.99) + 0.01))查询
import numpy
numpy.set_printoptions(suppress=True)
import scipy.special
import matplotlib.pyplot
class neuralNetwork:
def __init__(self,inputnodes,hiddennodes,outputnodes,learningrate):
self.inodes = inputnodes
self.hnodes = hiddennodes
self.onodes = outputnodes
self.lr = learningrate
self.wih = numpy.random.normal(0.0, pow(self.hnodes,-0.5),(self.hnodes,self.inodes))
self.who = numpy.random.normal(0.0, pow(self.onodes,-0.5),(self.onodes,self.hnodes))
self.activation_function = lambda x: scipy.special.expit(x)
pass
def train(self,inputs_list,targets_list):
inputs = numpy.array(inputs_list, ndmin=2).T
targets = numpy.array(targets_list, ndmin=2).T
hidden_inputs = numpy.dot(self.wih, inputs)
hidden_outputs = self.activation_function(hidden_inputs)
final_inputs = numpy.dot(self.who, hidden_outputs)
final_outputs = self.activation_function(final_inputs)
output_errors = targets - final_outputs
hidden_eerors = numpy.dot(self.who.T,output_errors)
self.who += self.lr * numpy.dot((output_errors * final_outputs * (1.0 - final_outputs)),numpy.transpose(hidden_outputs))
self.wih += self.lr * numpy.dot((hidden_eerors * hidden_outputs * (1.0 - hidden_outputs)),numpy.transpose(inputs))
pass
def query(self,inputs_list):
inputs = numpy.array(inputs_list, ndmin=2).T
hidden_inputs = numpy.dot(self.wih, inputs)
hidden_outputs = self.activation_function(hidden_inputs)
final_inputs = numpy.dot(self.who, hidden_outputs)
final_outputs = self.activation_function(final_inputs)
return final_outputs
input_nodes = 784
hidden_nodes = 200
output_nodes = 10
learning_rate = 0.3
n = neuralNetwork(input_nodes,hidden_nodes,output_nodes,learning_rate)
training_data_file = open("mnist/mnist_train_100.csv",'r')
training_data_list = training_data_file.readlines()
training_data_file.close()
for record in training_data_list:
all_values = record.split(',')
inputs = (numpy.asfarray(all_values[1:]) / 255.0 * 0.99) + 0.01
targets = numpy.zeros(output_nodes) +0.01
targets[int(all_values[0])] = 0.99
n.train(inputs,targets)
pass
test_data_file = open("mnist/mnist_test_10.csv",'r')
test_data_list = test_data_file.readlines()
test_data_file.close()
all_values = test_data_list[0].split(',')
print(all_values[0])
image_array = numpy.asfarray(all_values[1:]).reshape((28,28))
matplotlib.pyplot.imshow(image_array,cmap='Greys',interpolation='none')
print(n.query((numpy.asfarray(all_values[1:]) / 255.0 * 0.99) + 0.01))
matplotlib.pyplot.show()