Summary

In this chapter, we've explored various advanced concepts, tools, and best practices that added more tools to our toolbox, ranging from advanced techniques for PixieApps (Streaming, how to implement a route by integrating third-party libraries with @captureOutput, PixieApp events, and better modularity with pd_app), to essential developer tools like the PixieDebugger. We've also covered the details of how to create your own custom visualization using the PixieDust display() API. We also discussed pixiedust_node, which is an extension of the PixieDust framework that lets developers who are more comfortable with JavaScript work with data in their favorite language.

Throughout the remainder of this book, we are going to put all these lessons learned to good use by building industry use case data pipelines, starting with a Deep Learning Visual Recognition application in Chapter 6, Analytics Study: AI and Image Recognition with TensorFlow.

A developer quick-reference...