pole

Documentation

Overview

Package pole provides definition of the pole balancing experiments is classic Reinforced Learning task proposed by Richard Sutton and Charles Anderson. In this experiment we will try to teach RF model of balancing pole placed on the moving cart.

Index

• func NewCartDoublePoleGenerationEvaluator(outDir string, markov bool, actionType ActionType) experiment.GenerationEvaluator
• func NewCartPoleGenerationEvaluator(outDir string, randomStart bool, winBalanceSteps int) experiment.GenerationEvaluator
• type ActionType
• type CartPole

Functions

func NewCartDoublePoleGenerationEvaluator

func NewCartDoublePoleGenerationEvaluator(outDir string, markov bool, actionType ActionType) experiment.GenerationEvaluator

NewCartDoublePoleGenerationEvaluator is the generations evaluator for double-pole balancing experiment: both Markov and non-Markov versions

func NewCartPoleGenerationEvaluator

func NewCartPoleGenerationEvaluator(outDir string, randomStart bool, winBalanceSteps int) experiment.GenerationEvaluator

NewCartPoleGenerationEvaluator is to create generations evaluator for single-pole balancing experiment. This experiment performs evolution on single pole balancing task in order to produce appropriate genome.

Types

type ActionType

type ActionType byte

ActionType The type of action to be applied to environment

const (
	// ContinuousAction The continuous action type meaning continuous values to be applied to environment
	ContinuousAction ActionType = iota
	// DiscreteAction The discrete action assumes that there are only discrete values of action (e.g. 0, 1)
	DiscreteAction
)

The supported action types

type CartPole

type CartPole struct {
	// contains filtered or unexported fields
}

CartPole The structure to describe cart pole emulation