Summary

In this chapter, we have demonstrated how to train a stock price prediction model on the H2O platform. We have explained how to build a deployment workflow that calls two separate workflows: one to access historical prices and another to generate a price forecast. Finally, we have shown how to export the results into a .svg image file and a .csv file and send them via email automatically.

You have learned about the challenges of stock price prediction and ways to adjust the application accordingly. You have also learned how to connect to the H2O platform and how to use H2O nodes for the fast and accurate processing of machine learning tasks. You have also learned how to access stock market prices from the internet via the Python pandas-datareader package. Finally, you have learned how to orchestrate multiple workflows from one caller workflow that consumes their results.

You have acquired the necessary skills to build workflows running on the H2O platform. They work...