This section explains the configuration of the Keras model that is applied to the dataset to predict a rating based on the features selected.
- In Keras, a Sequential model is simply a linear combination of layers, which are the following: Dense is used to define the layer types to a fully-connected layer within a deep neural network. Finally, Activation is used to convert the inputs from the features into an output that can be used as a prediction. There are many types of activation functions that can be used in a neural network; however, for this chapter, we will go with relu and softmax.
- The Sequential model is configured to include three Dense layers:
- The first layer has input_dim set to the number of features from ...