A machine learning classifier is a function that classifies objects. It is created by training on a set of examples rather than by explicit programming. Training data can be numerical, textual, sounds and images, as well as combinations of these.

Collect training examples

Train a classifier to differentiate between dark and light colors. Examples will be hexadecimal RGB color values:

Here are some dark colors:

Here are some light colors:

Construct a training set where each training value is assigned the correct class:

Train a classifier

Use the training set to train a classifier:

Test the classifier

Give the classifier a new color. It will infer, based on the training data, how to classify the color:

Get the classifiers degree of confidence in what it has inferred, expressed as the probabilities of membership in each class:

Classify a list of colors:

In[9]:=
Click for copyable input
Out[9]=

Get information about the classifier

Get a report of the properties of the classifier: