public DevelopmentTest()
{
for (int i = 0; i < 32; i++)
{
uint stone = (uint)1 << i;
Console.WriteLine(((Stones)stone).ToString() + ", " + GetPoint(stone));
}
Board board = new Board(Board.Tables.Outer, Board.Tables.Left, true);
PrimaryUcb primaryUcb = new PrimaryUcb();
Mcts mcts = new Mcts(primaryUcb);
mcts.Init(board);
while (!board.IsMyWin && !board.IsYoWin)
{
var t = mcts.SearchNextAsync();
Console.WriteLine("Thinking ... ");
t.Wait();
Node node = t.Result;
board.MoveNext(node.prevMove);
mcts.SetMove(node.prevMove);
board.PrintStones();
Console.WriteLine("\n");
}
}
void runner()
{
userWaiter = new AutoResetEvent(false);
Move move = new Move();
isRunning = true;
while (isRunning)
{
if (mainBoard.IsMyTurn)
{
move = waitInput(myController);
}
else
{
move = waitInput(yoController);
}
if (!isRunning)
{
break;
}
//mcts의 상태를 변경해준다. board와는 따로 동작한다.
//반드시 보드보다 먼저 실행해야 한다. 좀 복잡하네 ㅋㅋ
mcts?.SetMove(move);
//현재 게임의 상태를 변경.
mainBoard.MoveNext(move);
if (mainBoard.IsMyWin)
{
MessageBox.Show("내가 이겼다.");
break;
}
else if (mainBoard.IsYoWin)
{
MessageBox.Show("상대가 이겼다.");
break;
}
TreeViewerMain.Load(mcts);
}
}
/// <summary>
/// 두 네트워크가 서로 싸움을 해서 그 기보를 학습 데이터로 내보낸다.
/// </summary>
void genReinforce()
{
//Mcts mcts1 = new Mcts(new OnlyPolicy(tcpCommClient));
Mcts mcts1 = new Mcts(new RealYame(tcpCommClient));
Mcts mcts2 = new Mcts(new RealYame(tcpCommClient));
mcts1.MaxVisitCount = 500;
mcts2.MaxVisitCount = 500;
//bool isMyFirst = Global.Rand.Next() % 2 == 0;
Board board = new Board((Board.Tables)Global.Rand.Next(4), (Board.Tables)Global.Rand.Next(4), true, false);
mcts1.Init(new Board(board));
mcts2.Init(new Board(board));
List <Tuple <Board, Move> > moves1 = new List <Tuple <Board, Move> >();
List <Tuple <Board, Move> > moves2 = new List <Tuple <Board, Move> >();
bool isMyWin = false;
for (int i = 0; i < 80; i++)
{
Move move;
if (board.IsMyTurn)
{
var task = mcts1.SearchNextAsync();
task.Wait();
move = task.Result.prevMove;
moves1.Add(new Tuple <Board, Move>(board, move));
}
else
{
var task = mcts2.SearchNextAsync();
task.Wait();
move = task.Result.prevMove;
moves2.Add(new Tuple <Board, Move>(board, move));
}
mcts1.SetMove(move);
mcts2.SetMove(move);
board = board.GetNext(move);
//겜이 끝났는지 확인
if (board.IsFinished)
{
isMyWin = board.IsMyWin;
break;
}
//mcts2.MaxVisitCount += 200;
}
//턴제한으로 끝났으면 점수로
if (!board.IsFinished)
{
isMyWin = (board.Point > 0);
}
lock (dataPolicy)
{
if (isMyWin)
{
dataPolicy.AddRange(moves1);
}
else
{
dataPolicy.AddRange(moves2);
}
}
lock (dataValue)
{
if (isMyWin)
{
var vals1 = from e in moves1 select new Tuple <Board, float>(e.Item1, 1);
var vals2 = from e in moves2 select new Tuple <Board, float>(e.Item1.GetOpposite(), 0);
dataValue.AddRange(vals1);
dataValue.AddRange(vals2);
}
else
{
var vals1 = from e in moves1 select new Tuple <Board, float>(e.Item1, 0);
var vals2 = from e in moves2 select new Tuple <Board, float>(e.Item1.GetOpposite(), 1);
dataValue.AddRange(vals1);
dataValue.AddRange(vals2);
}
}
}
void genRanPseudo()
{
Mcts mcts = new Mcts(new PseudoYame());
mcts.MaxVisitCount = maxVisitedCount;
Board board = new Board((Board.Tables)Global.Rand.Next(4), (Board.Tables)Global.Rand.Next(4), true, false);
mcts.Init(board);
List <Tuple <Board, Move> > moves1 = new List <Tuple <Board, Move> >();
List <Tuple <Board, Move> > moves2 = new List <Tuple <Board, Move> >();
bool isMyWin = false;
for (int i = 0; i < 80; i++)
{
Move move;
if (board.IsMyTurn)
{
var task = mcts.SearchNextAsync();
task.Wait();
move = task.Result.prevMove;
moves1.Add(new Tuple <Board, Move>(board, move));
}
else
{
mcts.root.PrepareMoves();
List <Move> moves = mcts.root.moves;
move = moves[Global.Rand.Next(moves.Count)];
moves2.Add(new Tuple <Board, Move>(board, move));
}
mcts.SetMove(move);
board = board.GetNext(move);
//겜이 끝났는지 확인
if (board.IsFinished)
{
isMyWin = board.IsMyWin;
break;
}
}
//턴제한으로 끝났으면 점수로
if (!board.IsFinished)
{
isMyWin = (board.Point > 0);
}
lock (dataPolicy)
{
if (isMyWin)
{
dataPolicy.AddRange(moves1);
}
else
{
dataPolicy.AddRange(moves2);
}
}
lock (dataValue)
{
if (isMyWin)
{
var vals1 = from e in moves1 select new Tuple <Board, float>(e.Item1, 1);
dataValue.AddRange(vals1);
}
else
{
var vals1 = from e in moves1 select new Tuple <Board, float>(e.Item1, 0);
dataValue.AddRange(vals1);
}
}
winGames2.Add(isMyWin ? 1 : 0);
while (winGames2.Count > 100)
{
winGames2.RemoveAt(0);
}
winGames3.Add(isMyWin ? 1 : 0);
while (winGames3.Count > 1000)
{
winGames3.RemoveAt(0);
}
float rate2 = winGames2.Average();
float rate3 = winGames3.Average();
Console.WriteLine(" winning rate : " + rate2 + ", " + rate3);
}
void genRealPseudo()
{
//Mcts mcts1 = new Mcts(new OnlyPolicy(tcpCommClient));
Mcts mcts1 = new Mcts(new PseudoYame());
Mcts mcts2 = new Mcts(new PseudoYame());
mcts1.MaxVisitCount = 500;
mcts2.MaxVisitCount = 500;
bool isMyFirst = Global.Rand.Next() % 2 == 0;
Board board = new Board((Board.Tables)Global.Rand.Next(4), (Board.Tables)Global.Rand.Next(4), isMyFirst);
mcts1.Init(new Board(board));
mcts2.Init(new Board(board));
List <Tuple <Board, Move> > moves1 = new List <Tuple <Board, Move> >();
List <Tuple <Board, Move> > moves2 = new List <Tuple <Board, Move> >();
bool isMyWin = false;
for (int i = 0; i < 80; i++)
{
Move move;
if (board.IsMyTurn)
{
var task = mcts1.SearchNextAsync();
task.Wait();
move = task.Result.prevMove;
moves1.Add(new Tuple <Board, Move>(board, move));
}
else
{
var task = mcts2.SearchNextAsync();
task.Wait();
move = task.Result.prevMove;
moves2.Add(new Tuple <Board, Move>(board, move));
}
mcts1.SetMove(move);
mcts2.SetMove(move);
board = board.GetNext(move);
//겜이 끝났는지 확인
if (board.IsFinished)
{
isMyWin = board.IsMyWin;
break;
}
//mcts2.MaxVisitCount += 200;
}
//턴제한으로 끝났으면 점수로
if (!board.IsFinished)
{
isMyWin = (board.Point > 0);
}
lock (dataPolicy)
{
if (isMyWin)
{
dataPolicy.AddRange(moves1);
}
else
{
dataPolicy.AddRange(moves2);
}
}
lock (dataValue)
{
if (isMyWin)
{
var vals1 = from e in moves1 select new Tuple <Board, float>(e.Item1, 1);
var vals2 = from e in moves2 select new Tuple <Board, float>(e.Item1.GetOpposite(), 0);
dataValue.AddRange(vals1);
dataValue.AddRange(vals2);
}
else
{
var vals1 = from e in moves1 select new Tuple <Board, float>(e.Item1, 0);
var vals2 = from e in moves2 select new Tuple <Board, float>(e.Item1.GetOpposite(), 1);
dataValue.AddRange(vals1);
dataValue.AddRange(vals2);
}
}
winGames2.Add(isMyWin ? 1 : 0);
while (winGames2.Count > 100)
{
winGames2.RemoveAt(0);
}
winGames3.Add(isMyFirst ? 1 : 0);
while (winGames3.Count > 100)
{
winGames3.RemoveAt(0);
}
float rate2 = winGames2.Average();
float rate3 = winGames3.Average();
Console.WriteLine(" winning rate : " + rate2 + ", " + rate3);
}
void genMcts()
{
Console.WriteLine("genMcts ... ");
//게임 한 판 시작 ---------------------------
List <Tuple <Board, Move> > recP1 = new List <Tuple <Board, Move> >();
List <Tuple <Board, Move> > recP2 = new List <Tuple <Board, Move> >();
//랜덤으로 보드 생성
//상대방 선수로 놓는다. 어차피 시작하자마자 GetOpposite로 돌릴 거다.
RealYame yame = new RealYame(tcpCommClient);
OnlyPolicy policy = new OnlyPolicy(tcpCommClient, policyNetName);
Mcts mcts1 = new Mcts(yame);
Mcts mcts2 = new Mcts(policy);
mcts1.MaxVisitCount = 300;
mcts2.MaxVisitCount = 1;
//누가 먼저 시작하나.
bool isMyFirst = Global.Rand.NextDouble() > 0.5;
Board board = new Board((Board.Tables)Global.Rand.Next(4), (Board.Tables)Global.Rand.Next(4), isMyFirst);
bool isMyWin = false;
mcts1.Init(board);
mcts2.Init(board);
for (int turn = 0; turn < 100; turn++)
{
Move move;
Task <Node> task;
if (board.IsMyTurn)
{
task = mcts1.SearchNextAsync();
}
else
{
task = mcts2.SearchNextAsync();
}
task.Wait();
move = task.Result.prevMove;
if (move.IsEmpty)
{
var moves = board.GetAllPossibleMoves();
move = moves[Global.Rand.Next(moves.Count - 1)];
}
//움직임을 저장해주고
if (isMyFirst)
{
recP1.Add(new Tuple <Board, Move>(board, move));
}
else
{
recP2.Add(new Tuple <Board, Move>(board, move));
}
mcts1.SetMove(move);
mcts2.SetMove(move);
board = board.GetNext(move);
board.PrintStones();
//겜이 끝났는지 확인
if (board.IsFinished)
{
isMyWin = board.IsMyWin;
break;
}
}
//턴제한으로 끝났으면 점수로
if (!board.IsFinished)
{
isMyWin = (board.Point > 0);
}
lock (dataPolicy)
{
if (isMyWin)
{
Console.WriteLine(" Collect data : my win");
var flip = from rec in recP1 select(new Tuple <Board, Move>(rec.Item1.GetFlip(), rec.Item2.GetFlip()));
dataPolicy.AddRange(recP1);
dataPolicy.AddRange(flip);
var list1 = from rec in recP1 select(new Tuple <Board, float>(rec.Item1, 1.0f));
var list2 = from rec in recP2 select(new Tuple <Board, float>(rec.Item1.GetOpposite(), 0f));
var list1Flip = from rec in list1 select(new Tuple <Board, float>(rec.Item1.GetFlip(), rec.Item2));
var list2Flip = from rec in list2 select(new Tuple <Board, float>(rec.Item1.GetFlip(), rec.Item2));
dataValue.AddRange(list1);
dataValue.AddRange(list2);
dataValue.AddRange(list1Flip);
dataValue.AddRange(list2Flip);
}
else
{
Console.WriteLine(" Collect data : YO win");
var flip = from rec in recP2 select(new Tuple <Board, Move>(rec.Item1.GetFlip(), rec.Item2.GetFlip()));
dataPolicy.AddRange(recP2);
dataPolicy.AddRange(flip);
var list1 = from rec in recP1 select(new Tuple <Board, float>(rec.Item1, 0f));
var list2 = from rec in recP2 select(new Tuple <Board, float>(rec.Item1.GetOpposite(), 1.0f));
var list1Flip = from rec in list1 select(new Tuple <Board, float>(rec.Item1.GetFlip(), rec.Item2));
var list2Flip = from rec in list2 select(new Tuple <Board, float>(rec.Item1.GetFlip(), rec.Item2));
dataValue.AddRange(list1);
dataValue.AddRange(list2);
dataValue.AddRange(list1Flip);
dataValue.AddRange(list2Flip);
}
}
}