mlpy.mdp.distrib.ProbabilityDistribution

class mlpy.mdp.distrib.ProbabilityDistribution(proba_calc_method=None)

Bases: object

Probability Distribution.

This class handles evaluation of empirically derived states and calculates the probability distribution from them.

Parameters:

proba_calc_method : str

The method used to calculate the probability distribution for the initial state. Defaults to ‘defaultprobacalcmethod’.

Methods

add_state(state)	Adds a state to the states list.
clear()	Clear the probability distribution.
get()	Retrieve the probability distribution.
iadd(state, proba)	In-place addition of the probability to the states probability.
sample()	Returns a next state according to the probability distribution.