Machine Learning in Elixir

Wrapping Up

In this chapter, you implemented two types of neural networks to predict the prices of AAPL stock in the future. You created and trained both a convolutional and recurrent neural network to perform single-step time-series prediction on AAPL stock prices. You learned a bit about the challenges of modeling time-series data. You used Explorer for normalizing your input data and VegaLite for visualizing your input data.

In the previous two chapters, you spent a lot of time working with recurrent neural networks. Recurrent neural networks are powerful, but in the past five years, they’ve been completely blown away by an even more powerful class of models.

In the next chapter, you’ll come face-to-face with the transformer, the architecture ...