Understanding the testing matrix

In this section, we will look at the testing matrix that we should consider in order to evaluate the trained ML models. For the baseline approach, we will be using the following five testing matrices:

  • Precision
  • Recall
  • F1-score
  • Support
  • Training accuracy

Before we understand these terms, let's cover some basic terms that will help us to understand the preceding terms.

  • True Positive (TP)—If the classifier predicts that the given movie review carries a positive sentiment and that movie review has a positive sentiment in an actual scenario, then these kinds of test cases are considered TP. So, you can define the TP as if the test result is one that detects the condition when the condition is actually present.
  • True Negative ...

Get Machine Learning Solutions 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.