In this chapter, we covered sequence-to-sequence networks and wrote a language translator using a series of known sentence-by-sentence translations as a training set. We were introduced to RNNs as a base for our work and likely crossed the threshold of big data as we trained using a 20 GB set of training data.
Next, we'll jump into tabular data and make predictions on economic and financial data. We'll use parts of our prior work so we can hit the ground running, namely the initial pipeline work we've written so far to download and prepare training data. However, we'll focus on a time series problem, so it will be quite different from the image and text work we've done to date.