public void Actions_NoneWhenWinnerOnBoard()
{
var expected = new TicTacToeAction[0];
var actual = new TicTacToeState(new[] { TicTacToePlayer.X, TicTacToePlayer.X, TicTacToePlayer.X, null, null, null, null, null, null }, TicTacToePlayer.O).Actions.ToList();
AssertActionsEqual(expected, actual);
}
public void GivenASingleStateAndAction_HighestValueAction_ReturnsOnlyAction()
{
var values = new ActionValueTable();
var state = Board.CreateEmptyBoard();
var action = new TicTacToeAction();
values.Set(state, action, 1.0);
Assert.AreEqual(action, values.HighestValueAction(state));
}
public void AverageReturnFrom_ReturnsAverage()
{
var returns = new Returns();
var emptyBoard = Board.CreateEmptyBoard();
var action = new TicTacToeAction();
returns.Add(emptyBoard, action, 2);
returns.Add(emptyBoard, action, 4);
Assert.AreEqual(3, returns.AverageReturnFrom(emptyBoard, action));
}
public void CreatingInvalidState_Throws()
{
_env.SetState(Board.CreateFromString("x |" +
" |" +
" "));
var placeXAtTopLeft = new TicTacToeAction {
Position = 1, Tile = BoardTile.X
};
Assert.Throws <InvalidOperationException>(() => _env.Step(placeXAtTopLeft));
}
public void RewardIs0_WhenGameIsNotOver()
{
var placeOAtMiddleRight = new TicTacToeAction {
Position = 5, Tile = BoardTile.X
};
// act
var observation = _env.Step(placeOAtMiddleRight);
// assert
Assert.AreEqual(0, observation.Reward);
}
public void DuplicateBoards_AreTreatedAsEqual()
{
var values = new ActionValueTable();
var emptyBoard1 = Board.CreateEmptyBoard();
var emptyBoard2 = Board.CreateEmptyBoard();
var action1 = new TicTacToeAction();
var action2 = new TicTacToeAction();
values.Set(emptyBoard1, action1, 1.0);
values.Set(emptyBoard2, action2, 2.0);
Assert.AreEqual(action2, values.HighestValueAction(emptyBoard1));
Assert.AreEqual(action2, values.HighestValueAction(emptyBoard2));
}
public void HighestValueAction_ReturnsHighestValueAction()
{
var values = new ActionValueTable();
var state = Board.CreateEmptyBoard();
var action0 = new TicTacToeAction();
var action1 = new TicTacToeAction();
var action2 = new TicTacToeAction();
values.Set(state, action0, 0);
values.Set(state, action1, 1);
values.Set(state, action2, 2);
Assert.AreEqual(action2, values.HighestValueAction(state));
}
public void RewardIs1_ForWin()
{
_env.SetState(Board.CreateFromString("xx |" +
"oo |" +
" "));
var placeXAtTopRight = new TicTacToeAction {
Position = 2, Tile = BoardTile.X
};
// act
var observation = _env.Step(placeXAtTopRight);
// assert
Assert.AreEqual(1, observation.Reward);
}
public void RewardIsNegative1_ForLoss()
{
var board = Board.CreateFromString("xx |" +
"oo |" +
" ",
BoardTile.O);
_env.SetState(board);
var placeOAtMiddleRight = new TicTacToeAction {
Position = 5, Tile = BoardTile.O
};
// act
var observation = _env.Step(placeOAtMiddleRight);
// assert
Assert.AreEqual(-1, observation.Reward);
}
public void DuplicateActions_AreTreatedAsEqual()
{
var values = new ActionValueTable();
var board = Board.CreateEmptyBoard();
var action1 = new TicTacToeAction {
Position = 2, Tile = BoardTile.X
};
var action2 = new TicTacToeAction {
Position = 2, Tile = BoardTile.X
};
values.Set(board, action1, 1.0);
values.Set(board, action2, 2.0);
var expectedHighestValueAction = new TicTacToeAction {
Position = 2, Tile = BoardTile.X
};
Assert.AreEqual(expectedHighestValueAction, values.HighestValueAction(board));
}
public TicTacToeEnvironmentStep Step(TicTacToeAction action)
{
try
{
DoAction(action);
}
catch (Exception e)
{
throw new InvalidOperationException(e.Message);
}
if (!CurrentState.IsValid())
{
throw new InvalidOperationException($"Action caused invalid state: '{CurrentState}'");
}
if (!CurrentState.IsGameOver)
{
DoAction(_opponent.GetAction(CurrentState));
}
var reward = 0.0;
if (CurrentState.Winner() == BoardTile.X)
{
reward = 1.0;
}
if (CurrentState.Winner() == BoardTile.O)
{
reward = -1.0;
}
return(new TicTacToeEnvironmentStep
{
Board = CurrentState,
Reward = reward
});
}
public void Step_DoesAgentAndOpponentMoves()
{
var placeXAtTopLeft = new TicTacToeAction {
Position = 0, Tile = BoardTile.X
};
var placeOAtTopMiddle = new TicTacToeAction {
Position = 1, Tile = BoardTile.O
};
_opponent.GetAction(Arg.Any <Board>())
.Returns(placeOAtTopMiddle);
var expectedBoard = Board
.CreateEmptyBoard()
.DoAction(placeXAtTopLeft)
.DoAction(placeOAtTopMiddle);
// act
var observation = _env.Step(placeXAtTopLeft);
// assert
Assert.AreEqual(expectedBoard.ToString(), observation.Board.ToString());
}
private void DoAction(TicTacToeAction action)
{
CurrentState = CurrentState.DoAction(action);
}
