public PawnMove TakeTurn(TicTacToeBoard t)
{
if (Policies.ContainsKey(t.GetBoardState()))
{
PawnMove moveToTake = Policies[t.GetBoardState()].OrderByDescending(x => x.Utility).ToList()[0];
if (Rnd.Range(1, 101) <= ExplorationRate)
{
moveToTake = t.GetAvailableMoves()[Rnd.Range(0, t.GetAvailableMoves().Count)];
}
TakenMoves.Add(moveToTake);
if (Spectation)
{
string res = "My available moves: \n";
foreach (PawnMove item in Policies[t.GetBoardState()].OrderByDescending(x => x.Utility).ToList())
{
res += $"[{item.X},{item.Y}]({item.Utility}) \n";
}
Console.WriteLine(res);
}
return(moveToTake);
}
else
{
Policies.Add(t.GetBoardState(), t.GetAvailableMoves());
PawnMove moveToTake;
moveToTake = Policies[t.GetBoardState()].OrderByDescending(x => x.Utility).ToList()[0];
TakenMoves.Add(moveToTake);
return(moveToTake);
}
}
public PawnMove TakeTurn(TicTacToeBoard t)
{
List <PawnMove> moves = t.GetAvailableMoves();
return(moves[Rnd.Range(0, moves.Count)]);
}
public void PlayTicTacToe()
{
TicTacToeBoard t = new TicTacToeBoard();
RandomAI randy = new RandomAI();
RandomAI randy2 = new RandomAI();
QLearnAITICTACTOE Qubert = new QLearnAITICTACTOE();
QLearnAITICTACTOE Qubert1 = new QLearnAITICTACTOE();
int player1win = 0;
int player2win = 0;
int stalemate = 0;
int player1winBM = 0;
int player2winBM = 0;
int stalemateBM = 0;
List <WinRatio> winratios = new List <WinRatio>();
for (int i = 0; i < 1000000; i++)
{
if (i % 10000 == 0)
{
winratios.Add(new WinRatio()
{
Player1 = player1winBM, Player2 = player2winBM, Draw = stalemateBM
});
stalemate++;
player1winBM = 0;
player2winBM = 0;
stalemateBM = 0;
}
t = new TicTacToeBoard();
while (t.CheckForWinner() == 0)
{
PawnMove mov = randy.TakeTurn(t);
t.TTTBoard[mov.X, mov.Y] = 1;
if (t.CheckForWinner() == 0)
{
mov = Qubert.TakeTurn(t);
t.TTTBoard[mov.X, mov.Y] = 2;
}
}
if (t.CheckForWinner() == 1)
{
// t.DrawBoard();
// Console.WriteLine("Player 1 Win!");
// Console.WriteLine();
player1win++;
player1winBM++;
Qubert.Reward(-1);
Qubert1.Reward(1);
}
if (t.CheckForWinner() == 2)
{
//t.DrawBoard();
//Console.WriteLine("Player 2 Win!");
//Console.WriteLine();
player2win++;
player2winBM++;
Qubert.Reward(1);
Qubert1.Reward(-1);
}
if (t.CheckForWinner() == 3)
{
// t.DrawBoard();
//Console.WriteLine("Noone");
// Console.WriteLine();
stalemate++;
stalemateBM++;
Qubert.Reward(.1f);
Qubert1.Reward(.1f);
}
}
Console.WriteLine("Hello World!");
Console.WriteLine("Player 1 victories: " + player1win);
Console.WriteLine("Player 2 victories: " + player2win);
Console.WriteLine("Draws " + stalemate);
string res = "Player 1 stats: ";
foreach (WinRatio item in winratios)
{
res += item.Player1 + "/";
}
Console.WriteLine(res);
res = "Player 2 stats: ";
foreach (WinRatio item in winratios)
{
res += item.Player2 + "/";
}
Console.WriteLine(res);
res = "Draws: ";
Console.WriteLine(Qubert.Policies.Count + "/19663 Known boardstates");
Console.WriteLine("" + ((float)Qubert.Policies.Count / 19663) * 100 + "% discovered");
foreach (WinRatio item in winratios)
{
res += item.Draw + "/";
}
Console.WriteLine(res);
Qubert.ExplorationRate = 0;
Qubert1.ExplorationRate = 0;
Qubert.Spectation = true;
for (int i = 0; i < 10; i++)
{
t = new TicTacToeBoard();
while (t.CheckForWinner() == 0)
{
t.DrawBoard();
Console.WriteLine("Choose X coordinate");
int playerX = int.Parse(Console.ReadLine());
Console.WriteLine("Choose Y coordinate");
int playerY = int.Parse(Console.ReadLine());
t.TTTBoard[playerX, playerY] = 1;
if (t.CheckForWinner() == 0)
{
PawnMove mov = Qubert.TakeTurn(t);
t.TTTBoard[mov.X, mov.Y] = 2;
}
}
if (t.CheckForWinner() == 1)
{
t.DrawBoard();
Console.WriteLine("Player 1 Win!");
Console.WriteLine();
player1win++;
player1winBM++;
Qubert.Reward(-1);
Qubert1.Reward(1);
}
if (t.CheckForWinner() == 2)
{
t.DrawBoard();
Console.WriteLine("Player 2 Win!");
Console.WriteLine();
player2win++;
player2winBM++;
Qubert.Reward(1);
Qubert1.Reward(-1);
}
if (t.CheckForWinner() == 3)
{
t.DrawBoard();
Console.WriteLine("Noone");
Console.WriteLine();
stalemate++;
stalemateBM++;
Qubert.Reward(-0.01f);
Qubert1.Reward(-0.01f);
}
}
}
