August 5, 2020

TESTING OUR MODEL

1. Training Set

2. Validation Set

3. Test Set

Making predictions and classifications.

1. Training Set is the data the algorithm will learn from.

2. Validation Set is use to compute the accuracy or error of the classifier.

   1. Ways of checking accuracy:

      1. Validation Error

      2. Precision Recall

      3. F1 Score

   2. In general, using 80% of your data for Training and 20% for Validation is a good place to start.

3. Test Set - once we are happy with our model’s performance, we use this as a final validation and test of our model to see the accuracy.