Variational autoencoder in Keras

In Keras, building the variational autoencoder is much easier and with lesser lines of code. The Keras variational autoencoders are best built using the functional style. So far we have used the sequential style of building the models in Keras, and now in this example, we will see the functional style of building the VAE model in Keras. The steps to build a VAE in Keras are as follows:

1. Define the hyper-parameters and the number of neurons in the hidden layers and the latent variables layer:

import keras
from keras.layers import Lambda, Dense, Input, Layer
from keras.models import Model
from keras import backend as K

learning_rate = 0.001
batch_size = 100
n_batches = int(mnist.train.num_examples/batch_size)
# number of pixels in the MNIST image as number of inputs
n_inputs = 784
n_outputs = n_inputs
# number of hidden layers
n_layers = 2
# neurons in each hidden layer
n_neurons = [512,256]
# the dimensions of latent variables
n_neurons_z = 128

2. Build the input...