Keras实现卷积神经网络

 # -*- coding: utf-8 -*-
 """
 Created on Sun Jan 20 11:25:29 2019

 @author: zhen
 """

 import numpy as np
 from keras.datasets import mnist
 from keras.models import Sequential
 from keras.layers import Dense
 from keras.layers import Dropout
 from keras.layers import Flatten
 from keras.layers.convolutional import Conv2D
 from keras.layers.convolutional import MaxPooling2D

 # 加载数据
 (x_train, y_train), (x_test, y_test) = mnist.load_data("../test_data_home")
 # 转化训练数据为四维张量形式
 x_train = x_train.reshape(x_train.shape[0], 28, 28, 1).astype("float32")
 x_test = x_test.reshape(x_test.shape[0], 28, 28, 1).astype("float32")
 # 归一化
 x_train /= 255
 x_test /= 255

 #转化为one hot 编码
 def to_one_hot(y):
     y_one_hot = np.zeros(10) # 生成全零向量
     y_one_hot[y] = 1
     return y_one_hot

 # 重置标签
 y_train_one_hot = np.array([to_one_hot(y_train[i]) for i in range(len(y_train))])
 y_test_one_hot = np.array([to_one_hot(y_test[i]) for i in range(len(y_test))])
 # 搭建卷积神经网络
 model = Sequential()
 model.add(Conv2D(filters=32, kernel_size=(3, 3), strides=(1, 1), padding='same', input_shape=(28, 28, 1),
                  activation='relu'))
 # 添加最大池化层
 model.add(MaxPooling2D(pool_size=(2, 2)))
 # 添加Dropout层
 model.add(Dropout(0.2))
 # 构建深度网络
 model.add(Conv2D(64, kernel_size=(3, 3), strides=(1, 1), padding='same', activation='relu'))
 model.add(MaxPooling2D(pool_size=(2, 2)))
 model.add(Dropout(0.2))
 model.add(Conv2D(128, kernel_size=(3, 3), strides=(1, 1), padding='same', activation='relu'))
 model.add(MaxPooling2D(pool_size=(2, 2)))
 model.add(Dropout(0.2))
 # 展开
 model.add(Flatten())
 # 构造全连接层
 model.add(Dense(128, activation='relu'))
 model.add(Dense(64, activation='relu'))
 model.add(Dense(32, activation='relu'))
 model.add(Dense(10, activation='softmax'))
 # 定义损失函数
 model.compile(loss='categorical_crossentropy', optimizer='adagrad',
               metrics=['accuracy'])
 # 训练
 model.fit(x_train, y_train_one_hot, validation_data=(x_test, y_test_one_hot),
           epochs=2, batch_size=128)
 # 评估
 # verbose : 0表示不显示数据,1表示显示进度
 scores = model.evaluate(x_test, y_test_one_hot, verbose=0)
 print(scores)

结果：