virtual protected void Train(bool p_validTurn, List <GridDto> p_grid, List <GameGoalDto> p_goals, int p_powerupScore)
{
Vector.Release(_state1);
_state1 = _encoder.EncodeState(p_grid, p_powerupScore, _agent.GetParam(BaseAgent.STATE_DIM), boost);
float reward = GetReward(p_validTurn, p_goals);
_accReward += reward;
int finished = IsFinished(p_validTurn, p_goals);
if (!_test)
{
_agent.Train(_state0, _state1, reward, finished != 0);
}
if (finished != 0)
{
//if (finished > 0) _wins++;
//if (finished < 0) _loses++;
if (_test)
{
if (_histogram.ContainsKey(_validMoves))
{
_histogram[_validMoves]++;
}
else
{
_histogram[_validMoves] = 1;
}
}
Console.WriteLine("Epoch " + (_wins + _loses) + ", seed " + _seed + " >> moves (" + _validMoves + " / " + _moves + ") , win rate (" + _wins + " / " + _loses + ") : reward " + _accReward);
_logger.Log(_wins + ";" + _loses + ";" + _seed + ";" + _validMoves + ";" + _moves + ";" + _accReward);
started = false;
}
/*
* if (_moves > _DEBUG_limit && _agent.Epsilon > 0.5d)
* {
* DEBUG_LogError();
* started = false;
* }
*/
Vector.Release(_state0);
_state0 = Vector.Copy(_state1);
}