Supervised Learning

Models are trained on labeled data, where each input has a known correct output.

• We have a dataset with one or more "features" (or independent variables, or just variables) (X) and one or more results (y).
• We would like to create a function that given a new set of values for X will predict the value(s) of y.

It is called "supervised" learning because for the given data-set we know whay is the true result (aka. labels).

We can divide the supervised learning problems into two types:

• Regression (continuous values)

• Classification (discrete labels)

• Input variables: Independent variables, aka. features

• Output variables: Dependent variables, target, labels

The model is supposed to predict the output from the input.