Index

_

__call__() (gomoku_rl.policy.base.Policy method) __init__() (gomoku_rl.collector.BlackPlayCollector method) (gomoku_rl.collector.SelfPlayCollector method) (gomoku_rl.collector.VersusPlayCollector method) (gomoku_rl.collector.WhitePlayCollector method) (gomoku_rl.core.Gomoku method) (gomoku_rl.env.GomokuEnv method) (gomoku_rl.policy.base.Policy method)

A

ActorNet (class in gomoku_rl.utils.module) add() (gomoku_rl.utils.psro.Population method) add_current_policy() (gomoku_rl.utils.psro.PSROPolicyWrapper method) add_prefix() (in module gomoku_rl.utils.misc) addPlayer() (gomoku_rl.utils.elo.Elo method) annotate_heatmap() (in module gomoku_rl.utils.visual)

AntiDiagonalFlip (class in gomoku_rl.utils.augment) assert_layer_transition() (in module gomoku_rl.utils.test) assert_observation() (in module gomoku_rl.utils.test) assert_tensor_1d_all() (in module gomoku_rl.utils.test) assert_transition() (in module gomoku_rl.utils.test) augment_transition() (in module gomoku_rl.utils.augment)

B

batch_size (gomoku_rl.env.GomokuEnv property) black (gomoku_rl.utils.test.Type attribute) black_vs_white (gomoku_rl.utils.psro.PayoffType attribute)

BlackPlayCollector (class in gomoku_rl.collector) board_size (gomoku_rl.env.GomokuEnv property) both (gomoku_rl.utils.psro.PayoffType attribute)

C

calculate_jpc() (in module gomoku_rl.utils.psro) Collector (class in gomoku_rl.collector) compute_done() (in module gomoku_rl.core) compute_elo_ratings() (in module gomoku_rl.utils.elo)

compute_expected_score() (in module gomoku_rl.utils.elo) converged() (gomoku_rl.utils.psro.ConvergedIndicator method) ConvergedIndicator (class in gomoku_rl.utils.psro) count_parameters() (in module gomoku_rl.utils.module)

D

device (gomoku_rl.env.GomokuEnv property) DiagonalFlip (class in gomoku_rl.utils.augment)

dict_flatten() (in module gomoku_rl.utils.wandb) DQN (class in gomoku_rl.policy.dqn)

E

Elo (class in gomoku_rl.utils.elo) eval() (gomoku_rl.policy.base.Policy method) (gomoku_rl.policy.dqn.DQN method) (gomoku_rl.policy.ppo.PPO method) (gomoku_rl.utils.psro.PSROPolicyWrapper method)

eval_win_rate() (in module gomoku_rl.utils.eval) expected_score() (gomoku_rl.utils.elo.Elo method)

F

fixed_behavioural_strategy() (gomoku_rl.utils.psro.Population method) forward() (gomoku_rl.utils.module.ActorNet method) (gomoku_rl.utils.module.MyDuelingCnnDQNet method) (gomoku_rl.utils.module.PolicyHead method) (gomoku_rl.utils.module.ResidualBlock method) (gomoku_rl.utils.module.ResidualTower method) (gomoku_rl.utils.module.ValueHead method) (gomoku_rl.utils.module.ValueNet method)

G

get_action_mask() (gomoku_rl.core.Gomoku method) get_augmented_transition() (in module gomoku_rl.utils.augment) get_encoded_board() (gomoku_rl.core.Gomoku method) get_kwargs() (in module gomoku_rl.utils.misc) get_log_func() (in module gomoku_rl.utils.log) get_meta_solver() (in module gomoku_rl.utils.psro) get_new_payoffs() (in module gomoku_rl.utils.psro) get_new_payoffs_sp() (in module gomoku_rl.utils.psro) get_optimizer() (in module gomoku_rl.policy.common) get_payoff_matrix() (in module gomoku_rl.utils.eval) get_policy() (in module gomoku_rl.policy) get_pretrained_policy() (in module gomoku_rl.policy) get_replay_buffer() (in module gomoku_rl.policy.dqn) Gomoku (class in gomoku_rl.core) gomoku_rl module gomoku_rl.collector module gomoku_rl.core module gomoku_rl.env module gomoku_rl.policy module gomoku_rl.policy.base module gomoku_rl.policy.common module gomoku_rl.policy.dqn module gomoku_rl.policy.ppo module gomoku_rl.runner module

gomoku_rl.runner.base module gomoku_rl.runner.independent_rl_runner module gomoku_rl.runner.psro_runner module gomoku_rl.utils module gomoku_rl.utils.augment module gomoku_rl.utils.elo module gomoku_rl.utils.eval module gomoku_rl.utils.log module gomoku_rl.utils.misc module gomoku_rl.utils.module module gomoku_rl.utils.policy module gomoku_rl.utils.psro module gomoku_rl.utils.sampler module gomoku_rl.utils.test module gomoku_rl.utils.visual module gomoku_rl.utils.wandb module GomokuEnv (class in gomoku_rl.env)

H

heatmap() (in module gomoku_rl.utils.visual)

HorizontalFlip (class in gomoku_rl.utils.augment)

I

Identity (class in gomoku_rl.utils.augment) IndependentRLRunner (class in gomoku_rl.runner.independent_rl_runner) IndependentRLSPRunner (class in gomoku_rl.runner.independent_rl_runner) init_params() (in module gomoku_rl.utils.module) init_payoffs_sp() (in module gomoku_rl.utils.psro) init_wandb() (in module gomoku_rl.utils.wandb) inverse_map_board() (gomoku_rl.utils.augment.AntiDiagonalFlip method) (gomoku_rl.utils.augment.DiagonalFlip method) (gomoku_rl.utils.augment.HorizontalFlip method) (gomoku_rl.utils.augment.Identity method) (gomoku_rl.utils.augment.Rotation method) (gomoku_rl.utils.augment.Transform method) (gomoku_rl.utils.augment.VerticalFlip method)

inverse_map_index() (gomoku_rl.utils.augment.AntiDiagonalFlip method) (gomoku_rl.utils.augment.DiagonalFlip method) (gomoku_rl.utils.augment.HorizontalFlip method) (gomoku_rl.utils.augment.Identity method) (gomoku_rl.utils.augment.Rotation method) (gomoku_rl.utils.augment.Transform method) (gomoku_rl.utils.augment.VerticalFlip method) is_valid() (gomoku_rl.core.Gomoku method)

L

learn() (gomoku_rl.policy.base.Policy method) (gomoku_rl.policy.dqn.DQN method) (gomoku_rl.policy.ppo.PPO method)

load_state_dict() (gomoku_rl.policy.base.Policy method) (gomoku_rl.policy.dqn.DQN method) (gomoku_rl.policy.ppo.PPO method)

M

make_behavioural_strategy() (gomoku_rl.utils.psro.Population method) make_critic() (in module gomoku_rl.policy.common) make_dataset_naive() (in module gomoku_rl.policy.common) make_dqn_actor() (in module gomoku_rl.policy.common) make_egreedy_actor() (in module gomoku_rl.policy.common) make_ppo_ac() (in module gomoku_rl.policy.common) make_ppo_actor() (in module gomoku_rl.policy.common) make_transition() (in module gomoku_rl.collector) map_board() (gomoku_rl.utils.augment.AntiDiagonalFlip method) (gomoku_rl.utils.augment.DiagonalFlip method) (gomoku_rl.utils.augment.HorizontalFlip method) (gomoku_rl.utils.augment.Identity method) (gomoku_rl.utils.augment.Rotation method) (gomoku_rl.utils.augment.Transform method) (gomoku_rl.utils.augment.VerticalFlip method) map_index() (gomoku_rl.utils.augment.AntiDiagonalFlip method) (gomoku_rl.utils.augment.DiagonalFlip method) (gomoku_rl.utils.augment.HorizontalFlip method) (gomoku_rl.utils.augment.Identity method) (gomoku_rl.utils.augment.Rotation method) (gomoku_rl.utils.augment.Transform method) (gomoku_rl.utils.augment.VerticalFlip method) Mean (class in gomoku_rl.utils.log) mixed (gomoku_rl.utils.test.Type attribute) module gomoku_rl gomoku_rl.collector gomoku_rl.core gomoku_rl.env gomoku_rl.policy gomoku_rl.policy.base gomoku_rl.policy.common gomoku_rl.policy.dqn gomoku_rl.policy.ppo gomoku_rl.runner gomoku_rl.runner.base gomoku_rl.runner.independent_rl_runner gomoku_rl.runner.psro_runner gomoku_rl.utils gomoku_rl.utils.augment gomoku_rl.utils.elo gomoku_rl.utils.eval gomoku_rl.utils.log gomoku_rl.utils.misc gomoku_rl.utils.module gomoku_rl.utils.policy gomoku_rl.utils.psro gomoku_rl.utils.sampler gomoku_rl.utils.test gomoku_rl.utils.visual gomoku_rl.utils.wandb

MyDuelingCnnDQNet (class in gomoku_rl.utils.module)

N

no_nan_in_tensordict() (in module gomoku_rl.utils.test)

num_envs (gomoku_rl.env.GomokuEnv property)

P

payoff_headmap() (in module gomoku_rl.utils.visual) PayoffType (class in gomoku_rl.utils.psro) Policy (class in gomoku_rl.policy.base) PolicyHead (class in gomoku_rl.utils.module) Population (class in gomoku_rl.utils.psro)

PPO (class in gomoku_rl.policy.ppo) print_payoffs() (in module gomoku_rl.utils.psro) PSROPolicyWrapper (class in gomoku_rl.utils.psro) PSRORunner (class in gomoku_rl.runner.psro_runner) PSROSPRunner (class in gomoku_rl.runner.psro_runner)

R

ran_out (gomoku_rl.utils.sampler.SequentialSampler property) REGISTRY (gomoku_rl.policy.base.Policy attribute) reset() (gomoku_rl.collector.BlackPlayCollector method) (gomoku_rl.collector.Collector method) (gomoku_rl.collector.SelfPlayCollector method) (gomoku_rl.collector.VersusPlayCollector method) (gomoku_rl.collector.WhitePlayCollector method) (gomoku_rl.core.Gomoku method) (gomoku_rl.env.GomokuEnv method) (gomoku_rl.utils.psro.ConvergedIndicator method) ResidualBlock (class in gomoku_rl.utils.module)

ResidualTower (class in gomoku_rl.utils.module) rollout() (gomoku_rl.collector.BlackPlayCollector method) (gomoku_rl.collector.Collector method) (gomoku_rl.collector.SelfPlayCollector method) (gomoku_rl.collector.VersusPlayCollector method) (gomoku_rl.collector.WhitePlayCollector method) Rotation (class in gomoku_rl.utils.augment) round() (in module gomoku_rl.collector) run() (gomoku_rl.runner.base.Runner method) (gomoku_rl.runner.base.SPRunner method) Runner (class in gomoku_rl.runner.base)

S

sample() (gomoku_rl.utils.psro.Population method) (gomoku_rl.utils.psro.PSROPolicyWrapper method) (gomoku_rl.utils.sampler.SequentialSampler method) self_play_step() (in module gomoku_rl.collector) SelfPlayCollector (class in gomoku_rl.collector) SequentialSampler (class in gomoku_rl.utils.sampler) set_meta_policy() (gomoku_rl.utils.psro.PSROPolicyWrapper method) set_oracle_mode() (gomoku_rl.utils.psro.PSROPolicyWrapper method) set_post_step() (gomoku_rl.env.GomokuEnv method) set_seed() (in module gomoku_rl.utils.misc)

solve_last_n() (in module gomoku_rl.utils.psro) solve_nash() (in module gomoku_rl.utils.psro) solve_uniform() (in module gomoku_rl.utils.psro) SPRunner (class in gomoku_rl.runner.base) state_dict() (gomoku_rl.policy.base.Policy method) (gomoku_rl.policy.dqn.DQN method) (gomoku_rl.policy.ppo.PPO method) step() (gomoku_rl.core.Gomoku method) (gomoku_rl.env.GomokuEnv method) step_and_maybe_reset() (gomoku_rl.env.GomokuEnv method)

T

to() (gomoku_rl.core.Gomoku method) train() (gomoku_rl.policy.base.Policy method) (gomoku_rl.policy.dqn.DQN method) (gomoku_rl.policy.ppo.PPO method) (gomoku_rl.utils.psro.PSROPolicyWrapper method)

Transform (class in gomoku_rl.utils.augment) Type (class in gomoku_rl.utils.test)

U

uniform_policy() (in module gomoku_rl.utils.policy) update() (gomoku_rl.utils.elo.Elo method) (gomoku_rl.utils.log.Mean method) (gomoku_rl.utils.psro.ConvergedIndicator method)

V

value (gomoku_rl.utils.log.Mean property) ValueHead (class in gomoku_rl.utils.module)

ValueNet (class in gomoku_rl.utils.module) VersusPlayCollector (class in gomoku_rl.collector) VerticalFlip (class in gomoku_rl.utils.augment)

W

white (gomoku_rl.utils.test.Type attribute)

WhitePlayCollector (class in gomoku_rl.collector)