keras多层感知器识别手写数字执行预测代码_感知机模型多层神经网络

import numpy as np                             #导入相关包
import pandas as pd
from keras.utils import np_utils
from keras.datasets import mnist
np.random.seed(10)
(X_train_image,y_train_label),(X_test_image,y_test_label)=mnist.load_data() #读取MNIST数据
import matplotlib.pyplot as plt
def plot_image(image):                     #定义显示某个数字图像的函数
fig=plt.gcf()
fig.set_size_inches(2,2)
plt.imshow(image,cmap='binary')
plt.show()
def plot_images_labels_prediction(images,labels,prediction,idx,num=10):   #定义显示features，labels和prediction的函数
fig=plt.gcf()
fig.set_size_inches(12,14)
if num>25:
num=25
for i in range(0,num):
ax=plt.subplot(5,5,1+i)
ax.imshow(images[idx],cmap='binary')
title="label="+str(labels[idx])
if len(prediction)>0:
title+=",prediction="+str(prediction[idx])
ax.set_title(title,fontsize=10)
ax.set_xticks([])
ax.set_yticks([])
idx+=1
plt.show()
X_Train=X_train_image.reshape(60000,784).astype('float32')   #将特征值转换为一维向量
X_Test=X_test_image.reshape(10000,784).astype('float32')
X_Train_normalize=X_Train/255  #将一维向量的数字标准化
X_Test_normalize=X_Test/255
y_TrainOneHot=np_utils.to_categorical(y_train_label)  #将labels转换为One-Hot Encoding
y_TestOneHot=np_utils.to_categorical(y_test_label)

from keras.models import Sequential
from keras.layers import Dense
from keras.layers import Dropout

model=Sequential()                    #建立模型
model.add(Dense(units=256,            #定义”隐藏层“神经元个数为256
input_dim=784,        #定义”输入层“神经元个数为784，因为28*28=784
kernel_initializer='normal',  #使用normal distribution正态分布来初始化权重(weight)和偏差
activation='relu'))  #定义激活函数为relu
model.add(Dropout(0.5))              #加入Doupout功能，防止过拟合
model.add(Dense(units=10,            #定义”输出层“神经元个数为10，代表0～9个数
kernel_initializer='normal',
activation='softmax')) #激活函数为softmax

model.compile(loss='categorical_crossentropy',optimizer='adam',metrics=['accuracy'])

train_history=model.fit(x=X_Train_normalize,y=y_TrainOneHot,validation_split=0.2,epochs=10,batch_size=200,verbose=2)

def show_train_history(train_histroy,train,validation): #输入参数为train_history和train_histroy.history中的索引train，validation
plt.plot(train_histroy.history[train])            #train_histroy.history是一个字典
plt.plot(train_history.history[validation])
plt.title('Train Histroy')
plt.ylabel(train)
plt.xlabel('epoch')
plt.legend(['train','validation'],loc='upper left')
plt.show()

show_train_history(train_history,'acc','val_acc')

show_train_history(train_history,'loss','val_loss')

scores=model.evaluate(X_Test_normalize,y_TestOneHot)
print(scores)

prediction=model.predict_classes(X_Test)

plot_images_labels_prediction(X_test_image,y_test_label,prediction,idx=1,num=25)

def show_wrong(images,labels,prediction):
fig=plt.gcf()
fig.set_size_inches(20,60)
array=[]
for i in range(0,10000):
if(str(labels[i])!=str(prediction[i])):            #把测试集的标签和预测结果进行比较，不相等的就是预测错误的，用一个数组保存不相等的索引
array.append(i)
for i in range(0,len(array)):
ax=plt.subplot(30,8,1+i)
ax.imshow(images[array[i]],cmap='binary')
title="label="+str(labels[array[i]])+",prediction="+str(prediction[array[i]])
ax.set_title(title,fontsize=12)
ax.set_xticks([])
ax.set_yticks([])
plt.show()
print("预测错误的一共有"+str(len(array))+"个")

show_wrong(x_Test,y_Test,prediction)

import pandas as pd
pd.crosstab(y_test_label,prediction,rownames=['lable'],colnames=['prediction'])