深度学习05-keras实例mnist手写数字识别实例分步骤详解

数据准备

直接导入数据集mnist，把训练集和测试集导入。

from keras.datasets import mnist
(X_train,y_train),(X_test,y_test) = mnist.load_data()

数据集介绍

print('图像数据格式：',X_train.shape)
print('训练集：%2.0f,测试集：%3.0f' % (X_train.shape[0],X_test.shape[0]))

训练集一共60000条数据集，测试集一共包含10000数据集，每条记录是2828的格式。图像由2828个像素点（灰度值）构成。

import matplotlib.pyplot as plt
plt.imshow(X_train[6])
plt.imshow(X_train[6],cmap = 'Greys')

#画20个样本
for i in range(20):
    plt.subplot(4,5,i+1)
    plt.imshow(X_train[i])
#把横纵坐标标签去掉
for i in range(20):
    plt.subplot(4,5,i+1)
    plt.imshow(X_train[i])
    plt.xticks([])
    plt.yticks([])
#更改颜色
for i in range(20):
    plt.subplot(4,5,i+1)
    plt.imshow(X_train[i],cmap = 'Reds')#Blues
    plt.xticks([])
    plt.yticks([])

#画出值为5的前15个样本
fig = plt.figure(figsize = (8,5))
for i in range(15):
    ax = fig.add_subplot(3,5,i+1,xticks = [],yticks = [])
    ax.imshow(X_train[y_train ==5][i],cmap = 'Greys')

这个 是画出前15个因变量为5的值

数据整理

输入维度为（num，28，28），需要把后面的维度拼接起来，变换成784维。
将因变量转换为哑变量组

X_train = X_train.reshape(X_train.shape[0],
                          X_train.shape[1] * X_train.shape[2])
X_test = X_test.reshape(X_test.shape[0],
                        X_test.shape[1]*X_test.shape[2])
X_train[0].shape

#将因变量转换为哑变量组
from keras.utils import to_categorical
y_train = to_categorical(y_train)
y_test = to_categorical(y_test)

建立模型

第一步选择模型

#定义模型
from keras.models import Sequential
from keras.layers.core import Dense,Dropout,Activation
from keras.optimizers import SGD
import numpy

#第一步选择模型 
model = Sequential()

第二步：构建网络层

model.add(Dense(500,input_shape = (784,)))#输入层，784
model.add(Activation('tanh'))#激活函数是tanh
model.add(Dropout(0.5))#50%节点连接Dropout

model.add(Dense(300))#隐藏层节点300个
model.add(Activation('tanh'))#激活函数是tanh
model.add(Dropout(0.5))#50%节点连接Dropout

model.add(Dense(10))#输出10个结果，因变量维度为10
model.add(Activation('softmax'))#最后一层激活函数是tanh

#网络连接权重数：（784+1）*500+(500+1)*300+(300+1)*10
model.summary()

第三步：编译

#优化函数，设定学习等参数n
sgd = SGD(lr = 0.01,decay = 1e-e,momentum = 0.9,nesterov = True)
model.compile(loss = 'categorical_crossentropy',
              optimizer = sgd,
              metrics = ['accuracy'])
#模型训练
from keras.callbacks import ModelCheckpoint

checkpoint = ModelCheckpoint(r'F:\learning_kecheng\deenlearning\NEW\all_zjwj\mnist_BPbest.hdf5',
                                  monitor = 'val_acc',
                                  save_best_only=True,
                                  verbose = 1)

第四步：训练

model.fit(X_train,y_train,batch_size = 200,epochs = 100,
          shuffle = True,verbose = 1,
          validation_data = (X_test,y_test),
          callbacks = [checkpoint])
model.save(r'F:\learning_kecheng\deenlearning\NEW\all_zjwj\mnist_BPbest.hdf5')
from keras.models import load_model
model = load_model(r'F:\learning_kecheng\deenlearning\NEW\all_zjwj\mnist_BPbest.hdf5')

第五步：模型评估

score = model.evaluate(X_test,y_test, batch_size = 200,verbose = 1)
print('测试集损失函数：%f,预测准确率：%2.2f%%'%(score[0],score[1]*100))