Measuring Model Performance

Description
After fitting a classifier and using it to predict the labels of previously unseen data, you then must figure out how to measure it's performance. That is, you need a metric. The speaker here asks the question, which data should be used to compute accuracy?

In this video, the speaker walks you through a step-by-step method on how to measure the performance of your model. He recommends splitting your data into 2 sets: a training set and a test set. This is to see how well your model generalizes to unseen data. The video takes you through the process of how to do this.

Length: 4 minutes 40 seconds