So, Keras is just a wrapper around TensorFlow and other deep learning frameworks. It's really good for prototyping and getting things built very quickly, but on the other hand, it gives you less control over your code. We'll take a chance to implement this sentiment analysis model in Keras so that you get a hands-on implementation in both TensorFlow and Keras. You can use Keras for fast prototyping and TensorFlow for a production-ready system.

More interesting news for you is that you don't have to switch to a totally different environment. You can now access Keras as a module in TensorFlow and import packages just like the following:

from tensorflow.python.keras.models 
import Sequential
from tensorflow.python.keras.layers 
import Dense, GRU, Embedding
from tensorflow.python.keras.optimizers 
import Adam
from tensorflow.python.keras.preprocessing.text 
import Tokenizer
from tensorflow.python.keras.preprocessing.sequence 
import pad_sequences

So, let's go ahead and use what we can now call a more abstracted module inside TensorFlow that will help us to prototype deep learning solutions very fast. This is because we will get to write full deep learning solutions in just a few lines of code.