private Intent GetBestIntent(TicTacTardStateWithAction action)
{
float maxValue = -1;
Intent intent = action.intent;
TicTacTardPlayer fakePlayer = new TicTacTardPlayer(2, "1");
for (int i = 5; i < 14; ++i)
{
Intent tempIntent = (Intent)i;
if (TicTacTardGame.CanPlayIntent(action, tempIntent))
{
TicTacTardState nextState = TicTacTardGame.PlayAction(action, fakePlayer, tempIntent, false);
TicTacTardState calculatedState = ticTacTardStateWithActions.Find(state => state.IsSameState(nextState));
if (calculatedState != null && calculatedState.value > maxValue)
{
maxValue = calculatedState.value;
intent = tempIntent;
}
}
}
return(intent);
}
public void InitIntent(bool isHuman)
{
isInit = false;
if (player.Count == 0)
{
switch (gameType)
{
case TicTacTardGameType.HumanVHuman:
for (int i = 0; i < 2; i++)
{
player.Add(new TicTacTardPlayer(i, i.ToString()));
}
break;
case TicTacTardGameType.HumanVBot:
player.Add(new TicTacTardAndroid(0, "0"));
player.Add(new TicTacTardPlayer(1, "1"));
TicTacTardStateWithAction currentStateWithAction =
new TicTacTardStateWithAction(currentState, GetRandomPossibleMove(currentState));
int safeLoopIteration = 0;
bool policyIsStable = false;
while (!policyIsStable && safeLoopIteration < 100)
{
++safeLoopIteration;
policyIsStable = (player[0] as TicTacTardAndroid).ComputeInitIntent(currentStateWithAction, true, true);
}
if (safeLoopIteration >= 100)
{
Debug.LogError("safeLoopIteration trigger : ExitComputeIntent");
}
Debug.Log("Bot generate " + (player[0] as TicTacTardAndroid).ticTacTardStateWithActions.Count + " states");
break;
case TicTacTardGameType.BotVBot:
for (int i = 0; i < 2; i++)
{
player.Add(new TicTacTardAndroid(i, i.ToString()));
}
break;
}
}
((TicTacTardPlayer)player[0]).ResetScore();
((TicTacTardPlayer)player[1]).ResetScore();
currentPlayer = (TicTacTardPlayer)player[0];
}
public TicTacTardPlayer(TicTacTardPlayer playerToClone)
{
id = playerToClone.id;
isHuman = playerToClone.isHuman;
token = playerToClone.token;
scores = new Dictionary <Direction, int>();
for (int i = 0; i < 8; ++i)
{
scores.Add((Direction)i, 0);
}
foreach (KeyValuePair <Direction, int> score in playerToClone.scores)
{
scores[score.Key] = score.Value;
}
}
private void ChangePlayer()
{
currentPlayer = (TicTacTardPlayer)(currentPlayer.ID == 0 ? player[1] : player[0]);
}
// 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);
}
private TicTacEpisode GenerateEpisodeFromState(TicTacTardStateWithAction state)
{
TicTacEpisode episode = new TicTacEpisode();
Intent initialIntent = EpsilonGreedy(state);
TicTacTardPlayer fakeOpponent1 = new TicTacTardPlayer(0, "0");
TicTacTardPlayer fakeOpponent2 = new TicTacTardPlayer(1, "1");
TicTacTardPlayer currentPlayer = fakeOpponent1;
string tokenCurrentPlayer = Token;
TicTacTardStateWithAction currentState = new TicTacTardStateWithAction(state, state.intent);
currentState.reward = 0;
currentState.WinScore = 0;
currentState.Visits = 0;
int safeLoopIteration = 0;
while (currentState.nbActionPlayed < 9 && !fakeOpponent1.playerWon && !fakeOpponent2.playerWon && safeLoopIteration < 200)
{
++safeLoopIteration;
TicTacTardState newState;
if (tokenCurrentPlayer == Token)
{
newState =
TicTacTardGame.PlayAction(currentState, currentPlayer, initialIntent, false);
if (newState == null)
{
initialIntent = TicTacTardGame.GetRandomPossibleMove(currentState);
continue;
}
}
else
{
newState =
TicTacTardGame.PlayAction(currentState, currentPlayer, TicTacTardGame.GetRandomPossibleMove(currentState), false);
if (newState == null)
{
continue;
}
}
TicTacTardStateWithAction existingState = ticTacTardStateWithActions.Find(stateSaved => newState.IsSameState(stateSaved));
if (existingState == null)
{
TicTacTardStateWithAction initNewState =
new TicTacTardStateWithAction(newState, TicTacTardGame.GetRandomPossibleMove(newState));
initNewState.prevState = currentState;
currentState = initNewState;
initialIntent = currentState.intent;
ticTacTardStateWithActions.Add(currentState);
}
else
{
currentState = existingState;
initialIntent = currentState.intent;
}
episode.EpisodeStates.Add(currentState);
currentPlayer = tokenCurrentPlayer == fakeOpponent1.Token ? fakeOpponent2 : fakeOpponent1;
tokenCurrentPlayer = currentPlayer.Token;
}
if (safeLoopIteration >= 200)
{
Debug.LogError("Safe loopIteration trigger : exit generate episode");
}
if (fakeOpponent1.playerWon)
{
currentState.reward = 1;
}
else
{
currentState.reward = 0;
}
return(episode);
}
