keras CNN

1、数据预处理

import keras
from keras.datasets import mnist
from keras.models import Sequential
from keras.layers import Dense, Flatten, Conv2D, MaxPooling2D
from keras import backend as K

num_classes = 10
img_rows, img_cols = 28, 28
 
# 通过Keras封装好的API加载MNIST数据。其中trainX就是一个60000 * 28 * 28的数组，
# trainY是每一张图片对应的数字。
(trainX, trainY), (testX, testY) = mnist.load_data()

# 根据对图像编码的格式要求来设置输入层的格式。
if K.image_data_format() == 'channels_first':
    trainX = trainX.reshape(trainX.shape[0], 1, img_rows, img_cols)
    testX = testX.reshape(testX.shape[0], 1, img_rows, img_cols)
    input_shape = (1, img_rows, img_cols)
else:
    trainX = trainX.reshape(trainX.shape[0], img_rows, img_cols, 1)
    testX = testX.reshape(testX.shape[0], img_rows, img_cols, 1)
    input_shape = (img_rows, img_cols, 1)

trainX = trainX.astype('float32')
testX = testX.astype('float32')
trainX /= 255.0
testX /= 255.0
 
# 将标准答案转化为需要的格式（one-hot编码）。
trainY = keras.utils.to_categorical(trainY, num_classes)
testY = keras.utils.to_categorical(testY, num_classes)

 

2、定义卷积神经网络. 

 A）：Sequential model：序列模型是一个线性的层次堆栈。 可以通过传递一系列 layer 实例给构造器来创建一个序列模型.可以简单的添加 layer 通过 .add() 函数。

B）：Compile函数：训练之前需要通过compile对学习过程进行配置，相关参数有三个，包括优化器/损失函数/metrics
Before training a model, you need to configure the learning process, which is done via the compile method. It receives three arguments:

An optimizer. This could be the string identifier of an existing optimizer (such as rmsprop or adagrad), or an instance of the Optimizer class. See: optimizers.
A loss function. This is the objective that the model will try to minimize. It can be the string identifier of an existing loss function (such as categorical_crossentropy or mse), or it can be an objective function. See: losses.
A list of metrics. For any classification problem you will want to set this to metrics=['accuracy']. A metric could be the string identifier of an existing metric or a custom metric function.

C）：Dense()方法

keras.layers.Dense(units, 
                  activation=None, 
                  use_bias=True, 
                  kernel_initializer='glorot_uniform', 
                  bias_initializer='zeros', 
                  kernel_regularizer=None, 
                  bias_regularizer=None, 
                  activity_regularizer=None, 
                  kernel_constraint=None, 
                  bias_constraint=None)

参数说明如下：

units:
该层有几个神经元

activation:
该层使用的激活函数

use_bias:
是否添加偏置项

kernel_initializer:
权重初始化方法

bias_initializer:
偏置值初始化方法

kernel_regularizer:
权重规范化函数

bias_regularizer:
偏置值规范化方法

activity_regularizer:
输出的规范化方法

kernel_constraint:
权重变化限制函数

bias_constraint:
偏置值变化限制函数
 

# 使用Keras API定义模型。
model = Sequential()
model.add(Conv2D(32, kernel_size=(5, 5), activation='relu', input_shape=input_shape))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Conv2D(64, (5, 5), activation='relu'))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Flatten())      #拉成一惟向量
model.add(Dense(500, activation='relu'))
model.add(Dense(num_classes, activation='softmax'))
 
# 定义损失函数、优化函数和评测方法。
model.compile(loss=keras.losses.categorical_crossentropy,
              optimizer=keras.optimizers.SGD(),
              metrics=['accuracy'])

3、训练模型并测试准确率

model.fit(trainX, trainY,
          batch_size=128,
          epochs=10,
          validation_data=(testX, testY))
 
# 在测试数据上计算准确率。
score = model.evaluate(testX, testY)
print('Test loss:', score[0])
print('Test accuracy:', score[1])