protected virtual void SinglePixelRLEvaluateStep()
{
SchoolWorld.ActionInput.SafeCopyToHost();
bool wasTargetDetected = SchoolWorld.ActionInput.Host[ControlMapper.Idx("forward")] != 0;
bool wasTargetPresent = m_previousObjectType == ObjectType.Target;
if (wasTargetDetected && wasTargetPresent)
{
m_targetsDetected++;
WrappedWorld.Reward.Host[0] = 1f;
}
else if (m_previousObjectType != ObjectType.None)
{
if (wasTargetDetected && !wasTargetPresent)
{
m_targetsMisdetected++;
WrappedWorld.Reward.Host[0] = -1f;
}
else if (wasTargetPresent && !wasTargetDetected)
{
WrappedWorld.Reward.Host[0] = -1f;
}
else
{
WrappedWorld.Reward.Host[0] = 0f;
}
}
else
{
WrappedWorld.Reward.Host[0] = 0;
}
}
public void MapWorldInputs()
{
// Copy data from wrapper to world (inputs) - SchoolWorld validation ensures that we have something connected
School.ActionInput.SafeCopyToHost();
ControlsAdapterTemp.Host[0] = School.ActionInput.Host[ControlMapper.Idx("left")]; // A
ControlsAdapterTemp.Host[1] = School.ActionInput.Host[ControlMapper.Idx("backward")]; // S
ControlsAdapterTemp.Host[2] = School.ActionInput.Host[ControlMapper.Idx("right")]; // D
ControlsAdapterTemp.SafeCopyToDevice();
}
public void EvaluateSinglePixelRLStep()
{
SchoolWorld.ActionInput.SafeCopyToHost();
int action = 0;
if (SchoolWorld.ActionInput.Host[ControlMapper.Idx("forward")] != 0)
{
action = 1;
}
else if (SchoolWorld.ActionInput.Host[ControlMapper.Idx("backward")] != 0)
{
action = 2;
}
else if (SchoolWorld.ActionInput.Host[ControlMapper.Idx("left")] != 0)
{
action = 3;
}
else if (SchoolWorld.ActionInput.Host[ControlMapper.Idx("right")] != 0)
{
action = 4;
}
int expectedAction = 0;
if (m_lastTransition != null)
{
expectedAction = m_lastTransition.action;
//MyLog.WARNING.WriteLine("Action taken: " + action+" expected action: "+expectedAction);
}
if (m_lastTransition == null)
{
m_stepIsCorrect = true;
WrappedWorld.Reward.Host[0] = 0f;
}
else if (action == expectedAction)
{
m_stepIsCorrect = true;
if (m_importantActions.Contains(action))
{
m_importantActionsTaken++;
WrappedWorld.Reward.Host[0] = 1f;
}
else
{
WrappedWorld.Reward.Host[0] = 0f;
}
}
else
{
m_stepIsCorrect = false;
WrappedWorld.Reward.Host[0] = -1f;
}
}
public override void Execute()
{
// Process FOF controls
Owner.ActionInput.SafeCopyToHost();
float fof_up = Owner.ActionInput.Host[ControlMapper.Idx("fof_up")]; // I
float fof_left = Owner.ActionInput.Host[ControlMapper.Idx("fof_left")]; // J
float fof_down = Owner.ActionInput.Host[ControlMapper.Idx("fof_down")]; // K
float fof_right = Owner.ActionInput.Host[ControlMapper.Idx("fof_right")]; // L
FofX = ConvertBiControlToUniControl(fof_left, fof_right);
FofY = ConvertBiControlToUniControl(fof_down, fof_up);
Owner.CurrentWorld.MapWorldInputs();
}
protected override bool DidTrainingUnitComplete(ref bool wasUnitSuccessful)
{
int correction = m_currentObjectType == ObjectType.Target ? 1 : 0;
wasUnitSuccessful = (m_targetsShown - correction == m_targetsDetected && m_targetsMisdetected == 0);
bool acted = SchoolWorld.ActionInput.Host[ControlMapper.Idx("forward")] != 0;
MyLog.Writer.WriteLine(MyLogLevel.WARNING, "did act: " + acted + ", wasTarget: " + (m_previousObjectType == ObjectType.Target) +
", was allowed to act: " + m_previousTargetWithAction);
MyLog.Writer.WriteLine(MyLogLevel.WARNING, "m_targetsShown:" + m_targetsShown + ", m_targetsDetected: " + m_targetsDetected +
"correction: " + correction + ", targetsMisdetected: " + m_targetsMisdetected);
return(m_targetsShown - correction == m_targetsPerTU);
// - correction, because when ending, ExecuteStep is done before the last EvaluateStep - the ExecuteStep may prepare a new state which should be ignored by EvaluateStep
}