In the old version:
from import merge
merge6 = merge([layer1,layer2], mode = 'concat', concat_axis = 3)
In the new version:
from import concatenate
merge = concatenate([layer1, layer2], axis=3)
Additional knowledge:Methods for keras input data: and model.fit_generator
1. The first, ordinary without data enhancement
from import mnist,cifar10，cifar100
(X_train, y_train), (X_valid, Y_valid) = cifar10.load_data()
(X_train, Y_train, batch_size=batch_size, nb_epoch=nb_epoch, shuffle=True,
verbose=1, validation_data=(X_valid, Y_valid), )
2. The second type, ImageDataGenerator with data enhancement, can rotate the angle, panning and other operations.
from import ImageDataGenerator
(trainX, trainY), (testX, testY) = cifar100.load_data()
trainX = ('float32')
testX = ('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)
(trainX, seed=0)
model.fit_generator((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=[0] // batch_size, verbose=1)
This description of the usage in keras above is all I have to share with you, I hope it will give you a reference and I hope you will support me more.