public override void Learn(GameHistory history)
{
m_Trained = true;
BoardRepresentation[] looserBoards = new BoardRepresentation[history.Count() / 2];
int looserIdx = looserBoards.Length;
for (int i = history.Count() - 2; i >= 0; i -= 2)
{
looserBoards[--looserIdx] = history.Peek(i);
}
BoardRepresentation[] winnerBoards = new BoardRepresentation[history.Count() - looserBoards.Length];
int winnerIdx = winnerBoards.Length;
for (int i = history.Count() - 1; i >= 0; i -= 2)
{
winnerBoards[--winnerIdx] = history.Peek(i);
}
m_FuncCallibrator.Callibrate(winnerBoards, looserBoards);
}
public void UndoMove()
{
if (GameHistory.Count != 0)
{
List <int> move = GameHistory.Peek();
GameBoard.Board[move[0], move[1]] = move[2];
GameBoard.Board[move[3], move[4]] = move[5];
RedoStack.Push(GameHistory.Pop());
PlayerOnTurn = PlayerSwap();
if (!GameInProgress)
{
GameInProgress = true;
}
}
}
public override void Learn(GameHistory history)
{
BoardRepresentation[] looserBoards = new BoardRepresentation[history.Count() / 2];
int looserIdx = looserBoards.Length;
for (int i = history.Count() - 2; i >= 0; i -= 2)
{
looserBoards[--looserIdx] = history.Peek(i);
}
BoardRepresentation[] winnerBoards = new BoardRepresentation[history.Count() - looserBoards.Length];
int winnerIdx = winnerBoards.Length;
for (int i = history.Count() - 1; i >= 0; i -= 2)
{
winnerBoards[--winnerIdx] = history.Peek(i);
}
int wIdx = winnerBoards.Length - 1;
int lIdx = looserBoards.Length - 1;
float[] wTarget = Grade(winnerBoards[wIdx]);;
float[] lTarget = Grade(looserBoards[lIdx]);
if (history.WinType == 1)
{
wTarget[0] = wTarget[0] + (0.1f * (1.0f - wTarget[0]));
wTarget[1] = wTarget[1] + (0.1f * (0.0f - wTarget[1]));
wTarget[2] = wTarget[2] + (0.1f * (0.0f - wTarget[2]));
wTarget[3] = wTarget[3] + (0.1f * (0.0f - wTarget[3]));
wTarget[4] = wTarget[4] + (0.1f * (0.0f - wTarget[4]));
lTarget[0] = lTarget[0] + (0.1f * (0.0f - lTarget[0]));
lTarget[1] = lTarget[1] + (0.1f * (0.0f - lTarget[1]));
lTarget[2] = lTarget[2] + (0.1f * (0.0f - lTarget[2]));
lTarget[3] = lTarget[3] + (0.1f * (0.0f - lTarget[3]));
lTarget[4] = lTarget[4] + (0.1f * (0.0f - lTarget[4]));
}
else if (history.WinType == 2)
{
wTarget[0] = wTarget[0] + (0.1f * (0.0f - wTarget[0]));
wTarget[1] = wTarget[1] + (0.1f * (1.0f - wTarget[1]));
wTarget[2] = wTarget[2] + (0.1f * (0.0f - wTarget[2]));
wTarget[3] = wTarget[3] + (0.1f * (0.0f - wTarget[3]));
wTarget[4] = wTarget[4] + (0.1f * (0.0f - wTarget[4]));
lTarget[0] = lTarget[0] + (0.1f * (0.0f - lTarget[0]));
lTarget[1] = lTarget[1] + (0.1f * (0.0f - lTarget[1]));
lTarget[2] = lTarget[2] + (0.1f * (0.0f - lTarget[2]));
lTarget[3] = lTarget[3] + (0.1f * (1.0f - lTarget[3]));
lTarget[4] = lTarget[4] + (0.1f * (0.0f - lTarget[4]));
}
else if (history.WinType == 3)
{
wTarget[0] = wTarget[0] + (0.1f * (0.0f - wTarget[0]));
wTarget[1] = wTarget[1] + (0.1f * (0.0f - wTarget[1]));
wTarget[2] = wTarget[2] + (0.1f * (1.0f - wTarget[2]));
wTarget[3] = wTarget[3] + (0.1f * (0.0f - wTarget[3]));
wTarget[4] = wTarget[4] + (0.1f * (0.0f - wTarget[4]));
lTarget[0] = lTarget[0] + (0.1f * (0.0f - lTarget[0]));
lTarget[1] = lTarget[1] + (0.1f * (0.0f - lTarget[1]));
lTarget[2] = lTarget[2] + (0.1f * (0.0f - lTarget[2]));
lTarget[3] = lTarget[3] + (0.1f * (0.0f - lTarget[3]));
lTarget[4] = lTarget[4] + (0.1f * (1.0f - lTarget[4]));
}
while (lIdx >= 0 || wIdx >= 0)
{
if (wIdx >= 0)
{
Train(winnerBoards[wIdx], wTarget);
if (wIdx > 0)
{
float[] prevWState = Grade(winnerBoards[wIdx - 1]);
wTarget[0] = prevWState[0] + (0.1f * (wTarget[0] - prevWState[0]));
wTarget[1] = prevWState[1] + (0.1f * (wTarget[1] - prevWState[1]));
wTarget[2] = prevWState[2] + (0.1f * (wTarget[2] - prevWState[2]));
wTarget[3] = prevWState[3] + (0.1f * (wTarget[3] - prevWState[3]));
wTarget[4] = prevWState[4] + (0.1f * (wTarget[4] - prevWState[4]));
}
wIdx--;
}
if (lIdx >= 0)
{
Train(looserBoards[lIdx], lTarget);
if (lIdx > 0)
{
float[] prevLState = Grade(looserBoards[lIdx - 1]);
lTarget[0] = prevLState[0] + (0.1f * (lTarget[0] - prevLState[0]));
lTarget[1] = prevLState[1] + (0.1f * (lTarget[1] - prevLState[1]));
lTarget[2] = prevLState[2] + (0.1f * (lTarget[2] - prevLState[2]));
lTarget[3] = prevLState[3] + (0.1f * (lTarget[3] - prevLState[3]));
lTarget[4] = prevLState[4] + (0.1f * (lTarget[4] - prevLState[4]));
}
lIdx--;
}
}
m_GamesTrained++;
if (m_GamesTrained % 100 == 0)
{
Save(SAVE_FILE);
}
}
public override void Learn(GameHistory history)
{
BoardRepresentation[] boards = new BoardRepresentation[history.Count()];
for (int i = boards.Length - 1; i >= 0; i--)
{
boards[i] = history.Peek(i);
}
int idx = boards.Length - 1;
int firstRaceBoardIdx = idx;
for (int i = 0; i < boards.Length; i++)
{
if (Race(boards[i]))
{
firstRaceBoardIdx = i;
break;
}
}
float[] wTarget = new float[5];
float[] lTarget = new float[5];
Grade(boards[idx], idx > firstRaceBoardIdx, wTarget);
Grade(boards[idx - 1], (idx - 1) > firstRaceBoardIdx, lTarget);
if (history.WinType == 1)
{
wTarget[0] = wTarget[0] + (STEP_SIZE * (1.0f - wTarget[0]));
wTarget[1] = wTarget[1] + (STEP_SIZE * (0.0f - wTarget[1]));
wTarget[2] = wTarget[2] + (STEP_SIZE * (0.0f - wTarget[2]));
wTarget[3] = wTarget[3] + (STEP_SIZE * (0.0f - wTarget[3]));
wTarget[4] = wTarget[4] + (STEP_SIZE * (0.0f - wTarget[4]));
lTarget[0] = lTarget[0] + (STEP_SIZE * (0.0f - lTarget[0]));
lTarget[1] = lTarget[1] + (STEP_SIZE * (0.0f - lTarget[1]));
lTarget[2] = lTarget[2] + (STEP_SIZE * (0.0f - lTarget[2]));
lTarget[3] = lTarget[3] + (STEP_SIZE * (0.0f - lTarget[3]));
lTarget[4] = lTarget[4] + (STEP_SIZE * (0.0f - lTarget[4]));
}
else if (history.WinType == 2)
{
wTarget[0] = wTarget[0] + (STEP_SIZE * (0.0f - wTarget[0]));
wTarget[1] = wTarget[1] + (STEP_SIZE * (1.0f - wTarget[1]));
wTarget[2] = wTarget[2] + (STEP_SIZE * (0.0f - wTarget[2]));
wTarget[3] = wTarget[3] + (STEP_SIZE * (0.0f - wTarget[3]));
wTarget[4] = wTarget[4] + (STEP_SIZE * (0.0f - wTarget[4]));
lTarget[0] = lTarget[0] + (STEP_SIZE * (0.0f - lTarget[0]));
lTarget[1] = lTarget[1] + (STEP_SIZE * (0.0f - lTarget[1]));
lTarget[2] = lTarget[2] + (STEP_SIZE * (0.0f - lTarget[2]));
lTarget[3] = lTarget[3] + (STEP_SIZE * (1.0f - lTarget[3]));
lTarget[4] = lTarget[4] + (STEP_SIZE * (0.0f - lTarget[4]));
}
else if (history.WinType == 3)
{
wTarget[0] = wTarget[0] + (STEP_SIZE * (0.0f - wTarget[0]));
wTarget[1] = wTarget[1] + (STEP_SIZE * (0.0f - wTarget[1]));
wTarget[2] = wTarget[2] + (STEP_SIZE * (1.0f - wTarget[2]));
wTarget[3] = wTarget[3] + (STEP_SIZE * (0.0f - wTarget[3]));
wTarget[4] = wTarget[4] + (STEP_SIZE * (0.0f - wTarget[4]));
lTarget[0] = lTarget[0] + (STEP_SIZE * (0.0f - lTarget[0]));
lTarget[1] = lTarget[1] + (STEP_SIZE * (0.0f - lTarget[1]));
lTarget[2] = lTarget[2] + (STEP_SIZE * (0.0f - lTarget[2]));
lTarget[3] = lTarget[3] + (STEP_SIZE * (0.0f - lTarget[3]));
lTarget[4] = lTarget[4] + (STEP_SIZE * (1.0f - lTarget[4]));
}
bool isCurrentBoardWinner = true;
float[] prevWState = new float[5];
float[] prevLState = new float[5];
while (idx >= 0)
{
if (isCurrentBoardWinner)
{
Train(boards[idx], idx > firstRaceBoardIdx, wTarget);
idx--;
if (idx > 0)
{
Grade(boards[idx - 1], idx - 1 > firstRaceBoardIdx, prevWState);
wTarget[0] = prevWState[0] + (STEP_SIZE * (wTarget[0] - prevWState[0]));
wTarget[1] = prevWState[1] + (STEP_SIZE * (wTarget[1] - prevWState[1]));
wTarget[2] = prevWState[2] + (STEP_SIZE * (wTarget[2] - prevWState[2]));
wTarget[3] = prevWState[3] + (STEP_SIZE * (wTarget[3] - prevWState[3]));
wTarget[4] = prevWState[4] + (STEP_SIZE * (wTarget[4] - prevWState[4]));
}
}
else
{
Train(boards[idx], idx > firstRaceBoardIdx, lTarget);
idx--;
if (idx > 0)
{
Grade(boards[idx - 1], idx - 1 > firstRaceBoardIdx, prevLState);
lTarget[0] = prevLState[0] + (STEP_SIZE * (lTarget[0] - prevLState[0]));
lTarget[1] = prevLState[1] + (STEP_SIZE * (lTarget[1] - prevLState[1]));
lTarget[2] = prevLState[2] + (STEP_SIZE * (lTarget[2] - prevLState[2]));
lTarget[3] = prevLState[3] + (STEP_SIZE * (lTarget[3] - prevLState[3]));
lTarget[4] = prevLState[4] + (STEP_SIZE * (lTarget[4] - prevLState[4]));
}
}
isCurrentBoardWinner = !isCurrentBoardWinner;
}
m_GamesTrained++;
if (m_GamesTrained % 100 == 0)
{
Save(SAVE_FILE);
}
}