public void Put(Vector env, Vector gesture, double evaluation)
{
SceneItem item = Get(env);
if (item == null)
{
item = new SceneItem(env, gesture, evaluation);
items.Add(item);
}
else
{
item.evaluation = evaluation;
}
}
/// <summary>
/// 对环境做出评估
/// Agent对环境有个好恶，该好恶与目标任务无关，只与它在环境中接收的奖励或惩罚有关
///
/// </summary>
/// <returns></returns>
private double doEvaluateEnviornment(int time, PolicyState policyState)
{
var obs = net.GetReceoptorSplit();
Vector env = obs.env;
//在记忆库中查找当前环境
SceneItem sceneItem = this.scene.GetMatched(env);
if (sceneItem != null)
{
policyState.AddEnviormentEvaluation(0, env, sceneItem.evaluation);
return(sceneItem.evaluation);
}
double forcastReward = net.DoForcastReward(time, obs.gesture, env, 3, 0);
policyState.AddEnviormentEvaluation(1, env, forcastReward);
return(forcastReward);
}
