lczerolens.play.sampling#

Classes for playing.

Classes#

`Sampler`	Helper class that provides a standard way to create an ABC using
`RandomSampler`	Helper class that provides a standard way to create an ABC using
`ModelSampler`	Helper class that provides a standard way to create an ABC using
`PolicySampler`	Helper class that provides a standard way to create an ABC using
`SelfPlay`	A class for generating games.

Module Contents#

class lczerolens.play.sampling.Sampler[source]#

Bases: abc.ABC

Helper class that provides a standard way to create an ABC using inheritance.

abstract get_utilities(boards, **kwargs)[source]#

Get the utility of the board.

Parameters:: boards (Iterable[LczeroBoard]) – The boards to evaluate.
Returns:: The iterable over utilities, legal indices, and log dictionaries.
Return type:: Iterable[Tuple[torch.Tensor, torch.Tensor, Dict[str, float]]]

choose_move(board, utility, legal_indices)[source]#

Choose the next moves.

Parameters:

board (LczeroBoard) – The board.
utility (torch.Tensor) – The utility of the board.
legal_indices (torch.Tensor) – The legal indices.

Returns:

The iterable over the moves.

Return type:

Iterable[chess.Move]

get_next_moves(boards, **kwargs)[source]#

Get the next move.

Parameters:: boards (Iterable[LczeroBoard]) – The boards to evaluate.
Returns:: The iterable over the moves and log dictionaries.
Return type:: Iterable[Tuple[chess.Move, Dict[str, float]]]

class lczerolens.play.sampling.RandomSampler[source]#

Bases: Sampler

Helper class that provides a standard way to create an ABC using inheritance.

get_utilities(boards, **kwargs)[source]#

Get the utility of the board.

Parameters:: boards (Iterable[LczeroBoard]) – The boards to evaluate.
Returns:: The iterable over utilities, legal indices, and log dictionaries.
Return type:: Iterable[Tuple[torch.Tensor, torch.Tensor, Dict[str, float]]]

class lczerolens.play.sampling.ModelSampler[source]#

Bases: Sampler

Helper class that provides a standard way to create an ABC using inheritance.

model: lczerolens.model.LczeroModel[source]#

use_argmax: bool = True[source]#

alpha: float = 1.0[source]#

beta: float = 1.0[source]#

gamma: float = 1.0[source]#

draw_score: float = 0.0[source]#

m_max: float = 0.0345[source]#

m_slope: float = 0.0027[source]#

k_0: float = 0.0[source]#

k_1: float = 1.6521[source]#

k_2: float = -0.6521[source]#

q_threshold: float = 0.8[source]#

choose_move(board, utility, legal_indices)[source]#

Choose the next moves.

Parameters:

board (LczeroBoard) – The board.
utility (torch.Tensor) – The utility of the board.
legal_indices (torch.Tensor) – The legal indices.

Returns:

The iterable over the moves.

Return type:

Iterable[chess.Move]

get_utilities(boards, **kwargs)[source]#

Get the utility of the board.

Parameters:: boards (Iterable[LczeroBoard]) – The boards to evaluate.
Returns:: The iterable over utilities, legal indices, and log dictionaries.
Return type:: Iterable[Tuple[torch.Tensor, torch.Tensor, Dict[str, float]]]

_get_batched_stats(boards, batch_size, use_next_boards=True, **kwargs)[source]#

_get_q_values(batch_stats, to_log)[source]#

_get_m_values(batch_stats, q_values, to_log)[source]#

_get_p_values(batch_stats, legal_indices, to_log)[source]#

_use_callback(callback, batch_stats, to_log)[source]#

class lczerolens.play.sampling.PolicySampler[source]#

Bases: ModelSampler

Helper class that provides a standard way to create an ABC using inheritance.

use_suboptimal: bool = False[source]#

get_utilities(boards, **kwargs)[source]#

Get the utility of the board.

Parameters:: boards (Iterable[LczeroBoard]) – The boards to evaluate.
Returns:: The iterable over utilities, legal indices, and log dictionaries.
Return type:: Iterable[Tuple[torch.Tensor, torch.Tensor, Dict[str, float]]]

class lczerolens.play.sampling.SelfPlay[source]#

A class for generating games.

white: Sampler[source]#

black: Sampler[source]#

play(board=None, max_moves=100, to_play=chess.WHITE, report_fn=None, white_kwargs=None, black_kwargs=None)[source]#

Plays a game.

Parameters:

board (Optional[lczerolens.board.LczeroBoard])
max_moves (int)
to_play (chess.Color)
report_fn (Optional[Callable[[dict, chess.Color], None]])
white_kwargs (Optional[Dict])
black_kwargs (Optional[Dict])