As mentioned in Chapter 10, Deploying on a Distributed System, in the Importing Python Models in the JVM with DL4J section, when doing DL in Python, an alternative to TensorFlow is Keras. It can be used as a high-level API on top of a TensorFlow backed. In this section, we are going to learn how to do sentiment analysis in Keras, and finally we will make a comparison between this implementation and the previous one in TensorFlow.
We are going to use the exact same IMDB dataset (25,000 samples for training and 25,000 for test) as for the previous implementations through DL4J and TensorFlow. The prerequisites for this example are the same as for the TensorFlow example (Python 2.7.x, the PIP package manager, and Tensorflow), plus of course Keras. The Keras code module has that dataset built in:
from keras.datasets import imdb
So, we...
