手写数字识别-小数据集

Posted 2020-12-10 ccla

1.手写数字数据集

• from sklearn.datasets import load_digits
• digits = load_digits()

 

 

 

 

2.图片数据预处理

• x：归一化MinMaxScaler()
  

   

   

   

   

   

• y：独热编码OneHotEncoder()或to_categorical
  

   

   

   

   

   

• 张量结构
  

   

   

• 训练集测试集划分

 　　

 

 

 

3.设计卷积神经网络结构

• 绘制模型结构图，并说明设计依据。
  

   

   

  # 导入相关包
  # sequential设计层数
  from tensorflow.keras.models import Sequential
  from tensorflow.keras.layers import Dense,Dropout,Flatten,Conv2D,MaxPool2D
  
  #建立模型
  model=Sequential()
  
  ks=(3,3)
  ips=X_train.shape[1:]
  
  #一层卷积
  model.add(Conv2D(filters=16, kernel_size=ks, padding=‘same‘,  input_shape=ips,activation=‘relu‘))
  
  #池化层1
  model.add(MaxPool2D(pool_size=(2,2)))
  model.add(Dropout(0.25))
  
  #二层卷积
  model.add(Conv2D(filters=32,kernel_size=ks,padding=‘same‘,activation=‘relu‘))
  
  #池化层2
  model.add(MaxPool2D(pool_size=(2,2)))
  model.add(Dropout(0.25))
  
  #三层卷积
  model.add(Conv2D(filters=64,kernel_size=ks,padding=‘same‘,activation=‘relu‘))
  #四层卷积
  model.add(Conv2D(filters=128,kernel_size=ks,padding=‘same‘,activation=‘relu‘))
  
  #池化层3
  model.add(MaxPool2D(pool_size=(2,2)))
  model.add(Dropout(0.25))
  
  model.add(Flatten()) #平坦层
  model.add(Dense(128,activation=‘relu‘)) #dense全连接层
  model.add(Dropout(0.25))
  model.add(Dense(10,activation=‘softmax‘)) #激活函数  softmax分类
  
  model.summary()

  Model: "sequential"
  _________________________________________________________________
  Layer (type) Output Shape Param #
  =================================================================
  conv2d (Conv2D) (None, 8, 8, 16) 160
  _________________________________________________________________
  max_pooling2d (MaxPooling2D) (None, 4, 4, 16) 0
  _________________________________________________________________
  dropout (Dropout) (None, 4, 4, 16) 0
  _________________________________________________________________
  conv2d_1 (Conv2D) (None, 4, 4, 32) 4640
  _________________________________________________________________
  max_pooling2d_1 (MaxPooling2 (None, 2, 2, 32) 0
  _________________________________________________________________
  dropout_1 (Dropout) (None, 2, 2, 32) 0
  _________________________________________________________________
  conv2d_2 (Conv2D) (None, 2, 2, 64) 18496
  _________________________________________________________________
  conv2d_3 (Conv2D) (None, 2, 2, 128) 73856
  _________________________________________________________________
  max_pooling2d_2 (MaxPooling2 (None, 1, 1, 128) 0
  _________________________________________________________________
  dropout_2 (Dropout) (None, 1, 1, 128) 0
  _________________________________________________________________
  flatten (Flatten) (None, 128) 0
  _________________________________________________________________
  dense (Dense) (None, 128) 16512
  _________________________________________________________________
  dropout_3 (Dropout) (None, 128) 0
  _________________________________________________________________
  dense_1 (Dense) (None, 10) 1290
  =================================================================
  Total params: 114,954
  Trainable params: 114,954
  Non-trainable params: 0
  _________________________________________________________________

4.模型训练

• model.compile(loss=‘categorical_crossentropy‘, optimizer=‘adam‘, metrics=[‘accuracy‘])
• train_history = model.fit(x=X_train,y=y_train,validation_split=0.2, batch_size=300,epochs=10,verbose=2)
  

   

   

  
  

   

   

5.模型评价

• model.evaluate()
  
• 交叉表与交叉矩阵
• pandas.crosstab
  

   

   

• seaborn.heatmap
  

   

   

   
  补交以前的一份作业

  K均值算法--应用

  因为爬取天气数据用了很长时间。网址：https://www.cnblogs.com/ccla/p/12732967.html