Keras Deep Learning Cookbook

Instantiate a sequential model

Next, we instantiate a sequential model and add the following layers:

• A simple RNN
• A dense layer with one output

Following are the steps of a simple RNN:

1. The next code listing shows model creation and compilation:

model = Sequential() model.add(SimpleRNN(n_neurons, batch_input_shape=(n_batch, X.shape[1], X.shape[2]), stateful=True)) model.add(Dense(1))

2. Then, we compile the model using model. compile(..) with loss and optimizers, shown as follows:

model.compile(loss='mean_squared_error', optimizer='adam')

We are using mean squared error (MSE) as the loss function and Adam as an optimizer. MSE is a loss function that uses a sum of squared difference between the predicted value and actual value divided b ...