Error Analysis

In the previous chapter, we explained the importance of error analysis. In this section, the different evaluation metrics will be calculated for all three models that were created in the previous activities so that we can compare them.

For learning purposes, we will compare the models using accuracy, precision, and recall metrics. This way, it will be possible to see that even though a model might be better in terms of one metric, it could be worse when measuring a different metric, which helps to emphasize the importance of choosing the right metric to measure your model according to the goal you wish to achieve.

Accuracy, Precision, and Recall

As a quick reminder, in order to measure performance and perform error analysis, it is required that you use the predict method for the different sets of data (training, validation, and testing). The following code snippets present a clean way of measuring all three metrics on our three sets at once:

Note

The following...