Practice First, Then Test

Practice, Test, Get Better

One of the most important parts of machine learning is how we teach computers to recognise patterns.

We do this by giving the computer examples to learn from (training data) and then testing it with new examples it hasn't seen before (testing data).

How it works

1. 1

   Training

   We show the computer many examples. The right answer comes with each one.

2. 2

   Learning

   The computer finds patterns in the training data

3. 3

   Test

   Now it gets new examples. It has never seen them before.

4. 4

   Counting Up

   We count how many answers were right.

You do the same thing. Before a test you practice at home. In the test you get new questions. There you show what you have learned.