Ch 19 recurrent neural nets

Deep Learning
#1

Hello!
Notebook: 02_stacked_lstm_with_feature_embeddings.ipynb

In this notebook, we insert three inputs (tickers, returns, and months). Can I insert other inputs, for example, technical information generated from TALIB? Is there a risk of overfitting?

returns = Input(shape=(window_size, n_features),
                name='Returns')

tickers = Input(shape=(1,),
                name='Tickers')

months = Input(shape=(12,),
               name='Months')

Other question: in this piece of code, we train and test with the same data frame that we predict after. Can we have a problem with overfitting?

training = rnn.fit(X_train,
                   y_train,
                   epochs=50,
                   batch_size=32,
                   validation_data=(X_test, y_test),
                   callbacks=[early_stopping, checkpointer],
                   verbose=1)

Thanks.