3.3 Summary
In this chapter, we finally dug into the trenches of practice instead of merely looking out from the towers of theory. Previously, we saw that NumPy arrays are the ideal tools for numeric computations, especially linear algebra. Now, we use them to provide fast and elegant implementations of what we learned in the previous chapter: norms, distances, dot products, and the Gram-Schmidt process.
Besides vectors, we also finally introduced matrices, one of the most important tools of machine learning. This time, we introduced, in a practical manner, viewing matrices as a table of numbers. Matrices can be transposed and added together, and unlike vectors, they can be multiplied with each other as well.
Speaking of our “from scratch” approach, before looking into how to actually work with matrices in practice, we created our very own Matrix implementation in vanilla Python. Closing the chapter, we dealt with the fundamentals and best practices of two-dimensional NumPy...
