from keras.datasets import mnist,cifar10，cifar100
(X_train, y_train), (X_valid, Y_valid) = cifar10.load_data() 
model.fit(X_train, Y_train, batch_size=batch_size, nb_epoch=nb_epoch, shuffle=True,
    verbose=1, validation_data=(X_valid, Y_valid), )

from keras.preprocessing.image import ImageDataGenerator
(trainX, trainY), (testX, testY) = cifar100.load_data()
trainX = trainX.astype('float32')
testX = testX.astype('float32')
trainX /= 255.
testX /= 255.
Y_train = np_utils.to_categorical(trainY, nb_classes)
Y_test = np_utils.to_categorical(testY, nb_classes)
generator = ImageDataGenerator(rotation_range=15,
        width_shift_range=5./32,
        height_shift_range=5./32)
generator.fit(trainX, seed=0)
model.fit_generator(generator.flow(trainX, Y_train, batch_size=batch_size),
     steps_per_epoch=len(trainX) // batch_size, epochs=nb_epoch,
     callbacks=callbacks,
     validation_data=(testX, Y_test),
     validation_steps=testX.shape[0] // batch_size, verbose=1)