Using the keras API completely is a bad way to showcase tensorflow 2.0. This exact same code works on 1.X, the real goodies come from eager execution, tf.function, being able to make your own training loop, etc.
Yes true. Although keras is not the perfect way to show the capabilities of tensorflow 2.0, But this is something to begin with. There is lot more to go.
We're a place where coders share, stay up-to-date and grow their careers.
We strive for transparency and don't collect excess data.