mlpy.learners.offline.IOfflineLearner.learn

IOfflineLearner.learn()

Learn a policy from the experience.

Perform the learning step to derive a new policy taking the latest experience into account.

Parameters:

experience : Experience

The agent’s experience consisting of the previous state, the action performed in that state, the current state and the reward awarded.

Raises:

NotImplementedError

If the child class does not implement this function.