当前位置: 首页 > news >正文

神经网络实验——MLP

目录

1 目的

2 方法

3 源代码

4 结果


1 目的

①熟悉 Python 的输入输出流;
②学会使用 matplotlib进行图像可视化;
③掌握神经网络的基本原理,学会使用 sklearn 库中的 MLPClassifier 函数构建基础的多层感知机神经网络分类器;
④学会使用网格查找进行超参数优化。

2 方法

①读取并解压 mnist.gz文件,并区分好训练集与测试集;
②查看数据结构,对手写字符进行可视化展示;
③构建多层感知机神经网络模型,并使用网格查找出最优参数;
④输出模型的最优参数以及模型的预测精度。

3 源代码

①启动 Spyder,新建.py 文件,加载试验所需模块

# 导入相关模块
from sklearn.neural_network import MLPClassifier
from sklearn.model_selection import GridSearchCV
import numpy as np
import pickle
import gzip
import matplotlib. pyplot as plt

②加载数据,数据文件保存在 mnist.gz 安装包中,因此需要对文件进行解压后对文件进行读取,且区分训练集、测试集与验证集

#解压数据并进行读取
with gzip.open(r"D:\大二下\数据挖掘\神经网络\mnist.gz") as fp:training_data, valid_data, test_data= pickle.load(fp,encoding='bytes') 
#区分训练集与测试集
X_training_data,y_training_data= training_data
X_valid_data,y_valid_data= valid_data
X_test_data, y_test_data= test_data

③查看数据的结构,为后续建模做准备:

#定义函数show_data_struct 用于展示数据的结构
def show_data_struct():print(X_training_data.shape,y_training_data.shape)print(X_valid_data.shape,y_valid_data.shape)print(X_test_data.shape,y_test_data.shape)print(X_training_data[0])print(y_training_data[0])
#使用show_data_struct 函数进行数据展示
show_data_struct()

④为了更好地了解数据的形态,对手写字符进行可视化展示

#定义函数用于可视化字符的原有图像
def show_image():plt.figure(1)for i in range(10):image=X_training_data[i]pixels=image.reshape((28,28))plt.subplot(5,2,i+1)plt.imshow(pixels,cmap='gray')plt.title(y_training_data[i])plt.axis('off')plt.subplots_adjust(top=0.92,bottom=0.08,left=0.10,right=0.95, hspace=0.45,wspace=0.85)plt.show()
#使用show_image函数进行图像展示
show_image()

⑤构建参数字典,用于后续使用网格查找进行超参数优化

#字典中用于存放的 MLPClassifier 函数的参数列表
mlp_clf__tuned_parameters= {"hidden_layer_sizes":[(100,),(100,30)],"solver":[' adam', 'sgd', 'bfgs'],"max_iter":[20],"verbose":[True]}

⑥使用MLPClassifier 丽数构建多层感知机神经网络,并使用GridSearchCV 网格查找进行超参数优化,找出最合适的参数

#构建多层感知机分类器
mlp=MLPClassifier()
#通过网格查找出最优参数
estimator= GridSearchCV(mlp,mlp_clf__tuned_parameters,n_jobs=6)
#拟合模型
estimator.fit(X_training_data, y_training_data)
#输出最优参数
print(estimator.best_params_)
#输出模型的预测精度
print(estimator.score(X_test_data, y_test_data))

4 结果

(50000, 784) (50000,)

(10000, 784) (10000,)

(10000, 784) (10000,)

[0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.01171875 0.0703125  0.0703125  0.0703125

 0.4921875  0.53125    0.68359375 0.1015625  0.6484375  0.99609375

 0.96484375 0.49609375 0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.1171875  0.140625   0.3671875  0.6015625

 0.6640625  0.98828125 0.98828125 0.98828125 0.98828125 0.98828125

 0.87890625 0.671875   0.98828125 0.9453125  0.76171875 0.25

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.19140625

 0.9296875  0.98828125 0.98828125 0.98828125 0.98828125 0.98828125

 0.98828125 0.98828125 0.98828125 0.98046875 0.36328125 0.3203125

 0.3203125  0.21875    0.15234375 0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.0703125  0.85546875 0.98828125

 0.98828125 0.98828125 0.98828125 0.98828125 0.7734375  0.7109375

 0.96484375 0.94140625 0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.3125     0.609375   0.41796875 0.98828125

 0.98828125 0.80078125 0.04296875 0.         0.16796875 0.6015625

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.0546875  0.00390625 0.6015625  0.98828125 0.3515625

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.54296875 0.98828125 0.7421875  0.0078125  0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.04296875

 0.7421875  0.98828125 0.2734375  0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.13671875 0.94140625

 0.87890625 0.625      0.421875   0.00390625 0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.31640625 0.9375     0.98828125

 0.98828125 0.46484375 0.09765625 0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.17578125 0.7265625  0.98828125 0.98828125

 0.5859375  0.10546875 0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.0625     0.36328125 0.984375   0.98828125 0.73046875

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.97265625 0.98828125 0.97265625 0.25       0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.1796875  0.5078125  0.71484375 0.98828125

 0.98828125 0.80859375 0.0078125  0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.15234375 0.578125

 0.89453125 0.98828125 0.98828125 0.98828125 0.9765625  0.7109375

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.09375    0.4453125  0.86328125 0.98828125 0.98828125 0.98828125

 0.98828125 0.78515625 0.3046875  0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.08984375 0.2578125  0.83203125 0.98828125

 0.98828125 0.98828125 0.98828125 0.7734375  0.31640625 0.0078125

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.0703125  0.66796875

 0.85546875 0.98828125 0.98828125 0.98828125 0.98828125 0.76171875

 0.3125     0.03515625 0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.21484375 0.671875   0.8828125  0.98828125 0.98828125 0.98828125

 0.98828125 0.953125   0.51953125 0.04296875 0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.53125    0.98828125

 0.98828125 0.98828125 0.828125   0.52734375 0.515625   0.0625

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.         0.         0.

 0.         0.         0.         0.        ]

5

通过观察数据结构可知,数据由 10000个样本组成,其中每一个样本是由784(28*28)个像素组成的图像,像素黑白用 0/1 进行表示,对应的label目标变量的每个字符图像的真实标签。

由图可知,MNIST数据由手写字符图像和标签组成。

通过对 MLP分类器的学习可见,模型经过 20次迭代,loss 不断减少0.2320587后达到拟合状态。

由输出结果可见,通过 GridSearchCV 网格查到的最优参数为:隐藏层数为(100,30),最大池化层为 20,激活函数为sgd;且此时多层感知机神经网络MNIST手写字符识别的准确率达到了 0.9347。

http://www.lryc.cn/news/539075.html

相关文章:

  • 配置Api自动生成
  • dify-AI 私有部署可修改前端页面
  • 使用 @Results 注解来手动指定字段映射
  • 数据治理中 大数据处理一般都遵循哪些原则
  • 从0到1:STM32温控系统开发踩坑指南
  • 修改时无条件,可以自定义id条件(通过查询)
  • 工业制造能耗管理新突破,漫途MTIC-ECM平台助力企业绿色转型!
  • 实现一个简单的协同过滤推荐算法
  • eNSP防火墙综合实验
  • 操作系统知识(二)
  • 图论:tarjan 算法求解强连通分量
  • Spring中Bean的四种实例化方法
  • 专利申请要求
  • 解锁 JavaScript 异步编程:Promise 链式操作、async/await 与 Promise.all 深度剖析
  • Centos虚拟机扩展磁盘空间
  • 记录一次部署PC端网址全过程
  • 利用 OpenCV 进行棋盘检测与透视变换
  • Java Spring boot 篇:常用注解
  • #渗透测试#批量漏洞挖掘#Apache Log4j反序列化命令执行漏洞
  • 【Linux】Linux 文件系统——关于inode 不足的相关案例
  • k8s集群如何赋权普通用户仅管理指定命名空间资源
  • 工控网络安全介绍 工控网络安全知识题目
  • AIGC(生成式AI)试用 21 -- Python调用deepseek API
  • 跨平台AES/DES加密解密算法【超全】
  • Webpack 基础入门
  • deepseek-v3在阿里云和腾讯云的使用中的差异
  • Mathtype安装入门指南
  • 使用 Apache PDFBox 提取 PDF 中的文本和图像
  • 【js逆向_入门】图灵爬虫练习平台 第四题
  • Redis7——基础篇(三)