Sequential Data
Sequential data is information that happens in a sequence and is related to past and future data. An example of sequential data is time series data; as you perceive it, time only travels in one direction.
Suppose you have a ball (as in Figure 9.2), and you want to predict where this ball will travel next. If you have no prior information about the direction from which the ball was thrown, you will simply have to guess. However, if in addition to the ball's current location, you also had information about its previous location, the problem would be much simpler. To be able to predict the ball's next location, you need the previous location information in a sequential (or ordered) form to make a prediction about future events.
RNNs function in a way that allows the sequence of the information to retain value with the help of internal memory.
You'll take a look at some examples of sequential...