8 Training and Testing
In this chapter we’ll look at training, the process of taking a system that’s been initialized with default or random values and gradually improving it so that it’s tuned to the data we want to understand. When we’re done training, we can estimate how well our system will evaluate new data it hasn’t seen before, a process known as testing.
We illustrate the ideas in this chapter using a supervised classifier, which we teach with labeled data. Most of the techniques we discuss are general and can be applied to almost all types of learners.
Training
When we train a classifier with supervised learning, every sample has ...
Get Deep Learning now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
