CompositeMarkovModel

A CompositeMarkovModel holds one model for each order from 1 up to maxOrder. When choosing the next symbol, it tries the highest order first and falls back to a shorter context when the longer one has not been seen. This keeps the style of a high-order model while still being able to continue from contexts it has not met before.

Train it with learn(), then make new sequences with generate().

Creating a CompositeMarkovModel

You can create a CompositeMarkovModel using the following function:

CompositeMarkovModel(maxOrder)

Parameter	Type	Default	Description
maxOrder	int	required	The highest order to use. The model holds every order from 1 up to this.

For example,

model = CompositeMarkovModel(4)

Functions

Once a CompositeMarkovModel model has been created, the following functions are available.

Function	Description
model.learn(listOfSymbols)	Learn the patterns in a sequence of symbols, at every order.
model.get(tupleOfSymbols)	Pick a random symbol to follow a context, using the longest order that fits.
model.generate()	Generate a new sequence in the style the model learned.
model.isConnected(context)	Report whether the model can continue from a context.