吴裕雄--天生自然TensorFlow高层封装：Keras-返回值

# 1. 数据预处理。
import keras
from keras.models import Model
from keras.datasets import mnist
from keras.layers import Input, Dense
from tflearn.layers.core import fully_connected

num_classes = 10
img_rows, img_cols = 28, 28

# 通过Keras封装好的API加载MNIST数据。
(trainX, trainY), (testX, testY) = mnist.load_data()
trainX = trainX.reshape(trainX.shape[0], img_rows * img_cols)
testX = testX.reshape(testX.shape[0], img_rows * img_cols)

# 将图像像素转化为0到1之间的实数。
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. 通过返回值的方式定义模型。
inputs = Input(shape=(784,))

x = Dense(500, activation='relu')(inputs)
predictions = Dense(10, activation='softmax')(x)

model = Model(inputs=inputs, outputs=predictions)
model.compile(loss=keras.losses.categorical_crossentropy,optimizer=keras.optimizers.SGD(),metrics=['accuracy'])

# 3. 训练模型。
model.fit(trainX, trainY,batch_size=32,epochs=10,validation_data=(testX, testY))