首页 > 专利信息

简单python代码实现三层神经网络识别手写数字

简单python代码实现三层神经⽹络识别⼿写数字准备

这个过程⾮常简单，就是⽤到了很多的矩阵运算。

训练数据集，

测试数据集，

数据格式

.csv格式数据的每⼀⾏都是⼀个28*28像素的⼿写数字图⽚，每⼀⾏的第⼀个像素是数字的值，从第⼆个数字开始时像素值

import matplotlib.pyplot

import pylab

import numpy

# 读⼊训练数据

training_data_file = open("G:\神经⽹络数据\mnist_train.csv", "r")

training_data_list = training_adlines()

training_data_file.close()

# 图⽚展⽰

aist = training_data_list[1].split(",")

aist = numpy.asfarray(aist[1:]).reshape((28, 28))

matplotlib.pyplot.imshow(aist, interpolation="nearest")

pylab.show()

效果图展⽰

神经⽹络代码

1. 对象

import scipy.special

class NeuralNetWork:

# 初始化

def __init__(self, inputnodfes, hiddennodes, outputnodes, learningrate):

self.innodes = inputnodfes

self.hnodes = hiddennodes

self.lr = learningrate

self.wih = al(0.0, pow(self.innodes, -0.5), (self.hnodes, self.innodes)) self.w

ho = al(0.0, pow(self.hnodes, -0.5), (des, self.hnodes)) # 抑制函数

self.activation_function = lambda x: pit(x)

pass

# 训练

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_errors = numpy.dot(self.who.T, output_errors)

self.who += self.lr * numpy.dot((output_errors * final_outputs * (1.0 - final_outputs)),

self.wih += self.lr * numpy.dot((hidden_errors * hidden_outputs * (1.0 - hidden_outputs)), anspose(inputs))

pass

# 测试

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

1. 训练和测试代码

from neuralNetwork import NeuralNetWork

input_nodes = 784

hidden_nodes = 100

output_nodes = 10

learning_rate = 0.2

# 读⼊训练数据

training_data_file = open("G:\神经⽹络数据\mnist_train.csv", "r")

training_data_list = training_adlines()

training_data_file.close()

# 初始化神经⽹络

b = NeuralNetWork(input_nodes, hidden_nodes, output_nodes, learning_rate)

for record in training_data_list:

all_values = record.split(',')

inputs = (numpy.asfarray(all_values[1:]) / 255.0 * 0.9) + 0.01

targets = s(output_nodes) + 0.01

targets[int(all_values[0])] = 0.99

pass

print("训练完成")

# 使⽤测试数据测试准确性

test_data_file = open("G:\神经⽹络数据\mnist_test.csv", "r")

test_data_list = test_adlines()

test_data_file.close()

# 使⽤算法逐项对⽐测试数据是否准确，然后统计

scorecard = []

for record in test_data_list:

all_values = record.split(',')

correct_label = int(all_values[0])

inputs = (numpy.asfarray(all_values[1:]) / 255.0 * 0.99) + 0.01

outputs = b.query(inputs)

label = numpy.argmax(outputs)

if label == correct_label:

scorecard.append(1)

else:

scorecard.append(0)

pass

scorecard_array = numpy.asarray(scorecard)

print("准确率", scorecard_array.sum() / scorecard_array.size)

结果

最后运⾏的结果显⽰准确率达到94%

本文发布于:2024-09-21 21:48:01，感谢您对本站的认可！

本文链接：https://www.17tex.com/tex/2/89302.html

上一篇：数字图像处理总结汇总

下一篇：如何查看屏幕分辨率

标签：数据测试数据神经数字训练像素是否达到

留言与评论（共有 0 条评论）