public IGameState GetNextState(List <IGameState> gameStates, Intent intentTrigger)
{
int x = gameState.GetPos().x;
int y = gameState.GetPos().y;
switch (intentTrigger)
{
case Intent.WantToGoBot:
y -= 1;
break;
case Intent.WantToGoLeft:
x -= 1;
break;
case Intent.WantToGoRight:
x += 1;
break;
case Intent.WantToGoTop:
y += 1;
break;
}
return(gameStates.Find(state =>
{
float epsilon = 0.00001f;
Vector2Int statePos = state.GetPos();
return statePos.x == x && statePos.y == y;
}));
}
public TicTacTardState(TicTacTardState state, Vector2Int newToken, string tokenToPlace, bool updateDisplay)
{
visits = state.visits;
winScore = state.winScore;
List <List <ICell> > oldGridBoard = state.Grid;
gridBoard = new List <List <ICell> >();
nbActionPlayed = state.nbActionPlayed + 1;
for (int i = 0; i < oldGridBoard.Count; ++i)
{
gridBoard.Add(new List <ICell>());
for (int j = 0; j < oldGridBoard[i].Count; ++j)
{
gridBoard[i].Add(new TicTacTardCell(oldGridBoard[i][j] as TicTacTardCell, updateDisplay));
if (i == newToken.x && j == newToken.y)
{
((TicTacTardCell)gridBoard[i][j]).token = tokenToPlace;
}
}
}
if (updateDisplay)
{
gridBoard[newToken.x][newToken.y].GetCellGameObject().transform.GetChild(0).GetComponent <TextMeshPro>().text = tokenToPlace;
}
}
public TicTacTardCell(TicTacTardCell cell, bool updateDisplay)
{
position = cell.position;
CellType = cell.CellType;
token = cell.token;
playerId = cell.playerId;
if (updateDisplay)
{
gameObject = cell.GetCellGameObject();
}
}
public void PolicyEvaluation()
{
float delta;
float gamma = 0.9f;
float tetha = 0.1f;
int safeLoopIteration = 0;
ICell cell;
do
{
delta = 0;
++safeLoopIteration;
foreach (GameStateWithAction gameStateWithAction in gameStateWithActions)
{
Vector2Int pos = gameStateWithAction.gameState.GetPos();
cell = worldCells[pos.x][pos.y];
if ((goalX == pos.x && goalY == pos.y) || cell.GetCellType() == CellType.Obstacle)
{
continue;
}
float temp = gameStateWithAction.gameState.GetValue();
float newValue = 0;
IGameState nextGameState = gameStateWithAction.GetNextState(gameStates, gameStateWithAction.intent);
float nextReward = worldCells[nextGameState.GetPos().x][nextGameState.GetPos().y].GetReward();
newValue += 1 * (nextReward + gamma * nextGameState.GetValue());
gameStateWithAction.gameState.SetValue(newValue);
delta = Math.Max(delta, Math.Abs(temp - gameStateWithAction.gameState.GetValue()));
}
} while (delta >= tetha && safeLoopIteration < 5000);
if (safeLoopIteration >= 5000)
{
Debug.LogError("Safe loop iteration trigger, exit policyEvaluation");
}
}
public bool PolicyImprovement()
{
bool policyStable = true;
IPlayer fakePlayer = new GridWoldPlayer();
foreach (GameStateWithAction gameStateWithAction in gameStateWithActions)
{
Vector2Int currentPos = gameStateWithAction.gameState.GetPos();
ICell cell = worldCells[currentPos.x][currentPos.y];
if ((goalX == currentPos.x && goalY == currentPos.y) || cell.GetCellType() == CellType.Obstacle)
{
continue;
}
fakePlayer.SetCell(worldCells[currentPos.x][currentPos.y]);
Intent intentToPlay = gameStateWithAction.intent;
int maxValue = 0;
Intent intentWithBestValue = GetIntentWithBestValue(gameStateWithAction, fakePlayer).intent;
gameStateWithAction.intent = intentWithBestValue;
if (intentToPlay != intentWithBestValue)
{
Debug.Log("State " + gameStateWithAction.gameState.GetPos() + " has " + intentToPlay + " redirect to " + intentWithBestValue);
policyStable = false;
}
}
if (!policyStable)
{
Debug.Log(">>>>>>>>> Call PolicyEvaluation");
PolicyEvaluation();
}
return(policyStable);
}
// return true if intent is correct
public static TicTacTardState PlayAction(TicTacTardState state, TicTacTardPlayer player, Intent intent, bool updateDisplay)
{
Vector2Int vector2Int = new Vector2Int(0, 0);
bool endGame = false;
List <Direction> directions = new List <Direction>();
switch (intent)
{
case Intent.BotCenter:
vector2Int.x = 1;
vector2Int.y = 0;
directions.Add(Direction.Line1);
directions.Add(Direction.Column2);
break;
case Intent.BotLeft:
vector2Int.x = 0;
vector2Int.y = 0;
directions.Add(Direction.Line1);
directions.Add(Direction.Column1);
directions.Add(Direction.Diagonal1);
break;
case Intent.BotRight:
vector2Int.x = 2;
vector2Int.y = 0;
directions.Add(Direction.Line1);
directions.Add(Direction.Column3);
directions.Add(Direction.Diagonal2);
break;
case Intent.MidCenter:
vector2Int.x = 1;
vector2Int.y = 1;
directions.Add(Direction.Line2);
directions.Add(Direction.Column2);
directions.Add(Direction.Diagonal1);
directions.Add(Direction.Diagonal2);
break;
case Intent.MidLeft:
vector2Int.x = 0;
vector2Int.y = 1;
directions.Add(Direction.Line2);
directions.Add(Direction.Column1);
break;
case Intent.MidRight:
vector2Int.x = 2;
vector2Int.y = 1;
directions.Add(Direction.Line2);
directions.Add(Direction.Column3);
break;
case Intent.TopCenter:
vector2Int.x = 1;
vector2Int.y = 2;
directions.Add(Direction.Line3);
directions.Add(Direction.Column2);
break;
case Intent.TopLeft:
vector2Int.x = 0;
vector2Int.y = 2;
directions.Add(Direction.Line3);
directions.Add(Direction.Column1);
directions.Add(Direction.Diagonal2);
break;
case Intent.TopRight:
vector2Int.x = 2;
vector2Int.y = 2;
directions.Add(Direction.Line3);
directions.Add(Direction.Column3);
directions.Add(Direction.Diagonal1);
break;
}
TicTacTardCell currentCell = state.Grid[vector2Int.x][vector2Int.y] as TicTacTardCell;
if (currentCell?.token == "-1")
{
TicTacTardState newState = new TicTacTardState(state, vector2Int, player.Token, updateDisplay);
player.IncrementScore(directions);
return(newState);
}
else
{
if (updateDisplay)
{
Debug.Log("Token déjà placé");
}
}
return(null);
}
public TicTacTardCell(Vector2Int position, CellType CellType)
{
this.position = position;
this.CellType = CellType;
token = "-1";
}
public void SetPostion(Vector2Int position)
{
this.position = position;
}
public SooooookolatCell(float reward, CellType cellType, Vector2Int position)
{
this.reward = reward;
this.cellType = cellType;
this.position = position;
}
public void ValueIteration()
{
IPlayer fakePlayer = new GridWoldPlayer();
float tetha = 0.1f;
float delta;
float gamma = 0.8f;
int safeLoopIteration = 0;
do
{
++safeLoopIteration;
delta = 0;
foreach (GameStateWithAction gameStateWithAction in gameStateWithActions)
{
Vector2Int currentPos = gameStateWithAction.gameState.GetPos();
ICell currentCell = worldCells[currentPos.x][currentPos.y];
fakePlayer.SetCell(currentCell);
if ((goalX == currentPos.x && goalY == currentPos.y) || currentCell.GetCellType() == CellType.Obstacle)
{
continue;
}
float temp = gameStateWithAction.gameState.GetValue();
float newValue = 0;
for (int i = 1; i < 5; ++i)
{
if (GridWORDOGame.CanMove(fakePlayer, worldCells, (Intent)i))
{
IGameState nextGameState = gameStateWithAction.GetNextState(gameStates, (Intent)i);
float nextReward = worldCells[nextGameState.GetPos().x][nextGameState.GetPos().y].GetReward();
float tempValue = 1 * nextReward + (gamma * nextGameState.GetValue());
if (tempValue > newValue)
{
newValue = tempValue;
}
}
}
gameStateWithAction.gameState.SetValue(newValue);
delta = Math.Max(delta, Math.Abs(temp - gameStateWithAction.gameState.GetValue()));
}
} while (delta >= tetha && safeLoopIteration < 5000);
if (safeLoopIteration >= 5000)
{
Debug.LogError("Safe loop iteration trigger, exit valueIteration");
return;
}
foreach (GameStateWithAction gameStateWithAction in gameStateWithActions)
{
Vector2Int currentPos = gameStateWithAction.gameState.GetPos();
fakePlayer.SetCell(worldCells[currentPos.x][currentPos.y]);
gameStateWithAction.intent = GetIntentWithBestValue(gameStateWithAction, fakePlayer).intent;
}
}
public void SetPostion(Vector2Int position)
{
throw new System.NotImplementedException();
}
public GridWorldCell(Vector2Int pos, CellType type, float reward)
{
position = pos;
cellType = type;
this.reward = reward;
}
