public void MyTestInitialize()
{
int height = 500;
int width = 500;
List<PlantSpecies> species = new List<PlantSpecies>() { new PlantSpecies(0) { Radius = 5, Reward = 100, Count = 1 } };
List<Predator> predators = new List<Predator>();
_agent = new ActionListAgent(0, new List<float[]>()) { X = 250, Y = 250 };
List<ForagingAgent> agents = new List<ForagingAgent>() { _agent };
_world = new World(agents, height, width, species, predators, PlantLayoutStrategies.Uniform);
_world.SensorLookup = new SensorDictionary((int)_world.AgentHorizon, _world.Width, _world.Height);
_world.AgentHorizon = 50;
}
public DiversityAnalyzer(World world)
{
_world = world;
locations = new int[numlocations][];
for (int i = 0; i < numlocations; i++)
locations[i] = new int[2];
orientations = new int[100];
velocities = new float[100];
for (int i = 0; i < numlocations; i++)
{
locations[i][0] = _random.Next() % world.Width;
locations[i][1] = _random.Next() % world.Height;
orientations[i] = _random.Next() % 360;
velocities[i] = (float)_random.NextDouble() * 5f;
}
}
/// <summary>
/// Creates a new Q-Learning teacher.
/// </summary>
/// <param name="id">The unique ID of this teacher.</param>
/// <param name="brain">The neural network value function for this teacher. It should have (2 + # of sensors) input nodes and 1 output node.</param>
/// <param name="numOrientationActions">The number of buckets to discretize the orientation action spacer into.</param>
/// <param name="numVelocityActions">The number of buckets to discretize the velocity action spacer into.</param>
/// <param name="world">The world this teacher will be evaluated in.</param>
public QLearningAgent(int id, int speciesId, IBlackBox brain, bool agentsNavigate, bool agentsHide,
int numOrientationActions, int numVelocityActions, World world)
: base(id, speciesId, brain, agentsNavigate, agentsHide)
{
Debug.Assert(brain.OutputCount == 1, "Incorrect number of outputs in neural network!");
_numVelocityActions = numVelocityActions;
_numOrientationActions = numOrientationActions;
_random = new Random();
_prevState = new double[brain.InputCount];
_observedValue = new double[1];
world.PlantEaten += new World.PlantEatenHandler(world_PlantEaten);
MaxReward = 200;
LearningRate = DEFAULT_LEARNING_RATE;
DiscountFactor = DEFAULT_DISCOUNT_FACTOR;
Epsilon = DEFAULT_EPSILON;
// The backprop learning rate is equivalent to the Q-Learning learning rate.
((FastCyclicNetwork)Brain).BackpropLearningRate = LearningRate;
}
/// <summary>
/// Initialize the experiment with some optional XML configutation data.
/// </summary>
public void Initialize(string name, XmlElement xmlConfig)
{
_name = name;
_populationSize = XmlUtils.GetValueAsInt(xmlConfig, "PopulationSize");
_specieCount = XmlUtils.GetValueAsInt(xmlConfig, "SpecieCount");
_activationScheme = ExperimentUtils.CreateActivationScheme(xmlConfig, "Activation");
_complexityRegulationStr = XmlUtils.TryGetValueAsString(xmlConfig, "ComplexityRegulationStrategy");
_complexityThreshold = XmlUtils.TryGetValueAsInt(xmlConfig, "ComplexityThreshold");
_description = XmlUtils.TryGetValueAsString(xmlConfig, "Description");
_timeStepsPerGeneration = (ulong)XmlUtils.GetValueAsInt(xmlConfig, "TimeStepsPerGeneration");
_stepReward = XmlUtils.GetValueAsInt(xmlConfig, "StepReward");
_agentType =(AgentTypes) Enum.Parse(typeof(AgentTypes), XmlUtils.TryGetValueAsString(xmlConfig, "AgentType"));
_plantLayout = (PlantLayoutStrategies)Enum.Parse(typeof(PlantLayoutStrategies), XmlUtils.TryGetValueAsString(xmlConfig, "PlantLayout"));
_paradigm = (EvolutionParadigm)Enum.Parse(typeof(EvolutionParadigm), XmlUtils.TryGetValueAsString(xmlConfig, "EvolutionParadigm"));
bool? diverse = XmlUtils.TryGetValueAsBool(xmlConfig, "LogDiversity");
if (diverse.HasValue && diverse.Value)
_logDiversity = true;
if (_agentType == AgentTypes.Social)
{
var memSection = xmlConfig.GetElementsByTagName("Memory")[0] as XmlElement;
_memory = (MemoryParadigm)Enum.Parse(typeof(MemoryParadigm), XmlUtils.TryGetValueAsString(memSection, "Paradigm"));
SocialAgent.DEFAULT_MEMORY_SIZE = XmlUtils.GetValueAsInt(memSection, "Size");
if (_memory == MemoryParadigm.IncrementalGrowth)
{
_memGens = XmlUtils.GetValueAsInt(memSection, "GrowthGenerations");
_maxMemorySize = XmlUtils.GetValueAsInt(memSection, "MaxSize");
}
_teaching = (TeachingParadigm)Enum.Parse(typeof(TeachingParadigm), XmlUtils.TryGetValueAsString(xmlConfig, "TeachingParadigm"));
}
var species = new List<PlantSpecies>();
var plants = xmlConfig.GetElementsByTagName("Plant");
for (int i = 0; i < plants.Count; i++)
{
var plant = plants[i] as XmlElement;
species.Add(new PlantSpecies(i)
{
Name = XmlUtils.GetValueAsString(plant, "Name"),
Radius = XmlUtils.GetValueAsInt(plant, "Radius"),
Reward = XmlUtils.GetValueAsInt(plant, "Reward"),
Count = XmlUtils.GetValueAsInt(plant, "Count")
});
}
Random random = new Random();
var agents = new List<ForagingAgent>();
const int NUM_AGENTS = 10;
for (int i = 0; i < NUM_AGENTS; i++)
{
agents.Add(new SpinningAgent(i) { X = random.Next(500), Y = random.Next(500), Orientation = random.Next(360) });
}
List<Predator> predators = new List<Predator>();
_predCount = XmlUtils.GetValueAsInt(xmlConfig, "Predators");
var predStr = XmlUtils.TryGetValueAsString(xmlConfig, "PredatorDistribution");
if (predStr != null)
PredatorDistribution = (PredatorDistributionTypes)Enum.Parse(typeof(PredatorDistributionTypes), predStr, true);
_predTypes = XmlUtils.GetValueAsInt(xmlConfig, "PredatorTypes");
if (PredatorDistribution == PredatorDistributionTypes.Alternating)
_predGens = XmlUtils.GetValueAsDouble(xmlConfig, "PredatorGenerations");
_distinguishPreds = XmlUtils.GetValueAsBool(xmlConfig, "DistinguishPredators");
_world = new World(agents, XmlUtils.GetValueAsInt(xmlConfig, "WorldHeight"), XmlUtils.GetValueAsInt(xmlConfig, "WorldHeight"), species, predators)
{
AgentHorizon = XmlUtils.GetValueAsInt(xmlConfig, "AgentHorizon"),
PlantLayoutStrategy = _plantLayout,
StepReward = _stepReward,
PredatorTypes = _predTypes
};
var outputs = XmlUtils.TryGetValueAsInt(xmlConfig, "Outputs");
var navigation = XmlUtils.TryGetValueAsBool(xmlConfig, "AgentsNavigate");
var hiding = XmlUtils.TryGetValueAsBool(xmlConfig, "AgentsHide");
_navigationEnabled = navigation.HasValue ? navigation.Value : false;
_hidingEnabled = hiding.HasValue ? hiding.Value : false;
if (!outputs.HasValue)
{
if (_navigationEnabled || _hidingEnabled)
_outputs = (_navigationEnabled ? 2 : 0) + (_hidingEnabled ? _predTypes + 1 : 0);
else
_outputs = outputs.HasValue ? outputs.Value : 2;
}
else
_outputs = outputs.Value;
var inputs = XmlUtils.TryGetValueAsInt(xmlConfig, "Inputs");
_inputs = inputs.HasValue ? inputs.Value : _world.PlantTypes.Count() * World.SENSORS_PER_OBJECT_TYPE + (_distinguishPreds ? _predTypes : 1) * World.SENSORS_PER_OBJECT_TYPE + 1;
_eaParams = new NeatEvolutionAlgorithmParameters();
_eaParams.SpecieCount = _specieCount;
_neatGenomeParams = new NeatGenomeParameters()
{
ActivationFn = PlainSigmoid.__DefaultInstance
};
if (_teaching != TeachingParadigm.EgalitarianEvolvedAcceptability)
_neatGenomeParams.InitialInterconnectionsProportion = 0.1;
}
