SYMBOL INDEX (106 symbols across 13 files)

FILE: muzero/config.py
  class MuZeroConfig (line 14) | class MuZeroConfig(object):
    method __init__ (line 16) | def __init__(self,
    method new_game (line 82) | def new_game(self) -> AbstractGame:
    method new_network (line 85) | def new_network(self) -> BaseNetwork:
    method uniform_network (line 88) | def uniform_network(self) -> UniformNetwork:
    method new_optimizer (line 91) | def new_optimizer(self) -> tf.keras.optimizers:
  function make_cartpole_config (line 95) | def make_cartpole_config() -> MuZeroConfig:

FILE: muzero/game/cartpole.py
  class CartPole (line 9) | class CartPole(AbstractGame):
    method __init__ (line 12) | def __init__(self, discount: float):
    method action_space_size (line 21) | def action_space_size(self) -> int:
    method step (line 25) | def step(self, action) -> int:
    method terminal (line 33) | def terminal(self) -> bool:
    method legal_actions (line 37) | def legal_actions(self) -> List[Action]:
    method make_image (line 41) | def make_image(self, state_index: int):

FILE: muzero/game/game.py
  class Action (line 7) | class Action(object):
    method __init__ (line 10) | def __init__(self, index: int):
    method __hash__ (line 13) | def __hash__(self):
    method __eq__ (line 16) | def __eq__(self, other):
    method __gt__ (line 19) | def __gt__(self, other):
  class Player (line 23) | class Player(object):
    method __eq__ (line 29) | def __eq__(self, other):
  class ActionHistory (line 33) | class ActionHistory(object):
    method __init__ (line 39) | def __init__(self, history: List[Action], action_space_size: int):
    method clone (line 43) | def clone(self):
    method add_action (line 46) | def add_action(self, action: Action):
    method last_action (line 49) | def last_action(self) -> Action:
    method action_space (line 52) | def action_space(self) -> List[Action]:
    method to_play (line 55) | def to_play(self) -> Player:
  class AbstractGame (line 59) | class AbstractGame(ABC):
    method __init__ (line 65) | def __init__(self, discount: float):
    method apply (line 72) | def apply(self, action: Action):
    method store_search_statistics (line 79) | def store_search_statistics(self, root: Node):
    method make_target (line 90) | def make_target(self, state_index: int, num_unroll_steps: int, td_step...
    method to_play (line 113) | def to_play(self) -> Player:
    method action_history (line 117) | def action_history(self) -> ActionHistory:
    method action_space_size (line 124) | def action_space_size(self) -> int:
    method step (line 129) | def step(self, action) -> int:
    method terminal (line 134) | def terminal(self) -> bool:
    method legal_actions (line 139) | def legal_actions(self) -> List[Action]:
    method make_image (line 144) | def make_image(self, state_index: int):

FILE: muzero/game/gym_wrappers.py
  class ScalingObservationWrapper (line 5) | class ScalingObservationWrapper(gym.ObservationWrapper):
    method __init__ (line 10) | def __init__(self, env, low=None, high=None):
    method observation (line 20) | def observation(self, observation):

FILE: muzero/muzero.py
  function muzero (line 8) | def muzero(config: MuZeroConfig):

FILE: muzero/networks/cartpole_network.py
  class CartPoleNetwork (line 12) | class CartPoleNetwork(BaseNetwork):
    method __init__ (line 14) | def __init__(self,
    method _value_transform (line 42) | def _value_transform(self, value_support: np.array) -> float:
    method _reward_transform (line 53) | def _reward_transform(self, reward: np.array) -> float:
    method _conditioned_hidden_state (line 56) | def _conditioned_hidden_state(self, hidden_state: np.array, action: Ac...
    method _softmax (line 60) | def _softmax(self, values):

FILE: muzero/networks/network.py
  class NetworkOutput (line 11) | class NetworkOutput(typing.NamedTuple):
    method build_policy_logits (line 18) | def build_policy_logits(policy_logits):
  class AbstractNetwork (line 22) | class AbstractNetwork(ABC):
    method __init__ (line 24) | def __init__(self):
    method initial_inference (line 28) | def initial_inference(self, image) -> NetworkOutput:
    method recurrent_inference (line 32) | def recurrent_inference(self, hidden_state, action) -> NetworkOutput:
  class UniformNetwork (line 36) | class UniformNetwork(AbstractNetwork):
    method __init__ (line 39) | def __init__(self, action_size: int):
    method initial_inference (line 43) | def initial_inference(self, image) -> NetworkOutput:
    method recurrent_inference (line 46) | def recurrent_inference(self, hidden_state, action) -> NetworkOutput:
  class InitialModel (line 50) | class InitialModel(Model):
    method __init__ (line 53) | def __init__(self, representation_network: Model, value_network: Model...
    method call (line 59) | def call(self, image):
  class RecurrentModel (line 66) | class RecurrentModel(Model):
    method __init__ (line 69) | def __init__(self, dynamic_network: Model, reward_network: Model, valu...
    method call (line 76) | def call(self, conditioned_hidden):
  class BaseNetwork (line 84) | class BaseNetwork(AbstractNetwork):
    method __init__ (line 87) | def __init__(self, representation_network: Model, value_network: Model...
    method initial_inference (line 102) | def initial_inference(self, image: np.array) -> NetworkOutput:
    method recurrent_inference (line 112) | def recurrent_inference(self, hidden_state: np.array, action: Action) ...
    method _value_transform (line 124) | def _value_transform(self, value: np.array) -> float:
    method _reward_transform (line 128) | def _reward_transform(self, reward: np.array) -> float:
    method _conditioned_hidden_state (line 132) | def _conditioned_hidden_state(self, hidden_state: np.array, action: Ac...
    method cb_get_variables (line 135) | def cb_get_variables(self) -> Callable:

FILE: muzero/networks/shared_storage.py
  class SharedStorage (line 6) | class SharedStorage(object):
    method __init__ (line 9) | def __init__(self, network: BaseNetwork, uniform_network: UniformNetwo...
    method latest_network (line 15) | def latest_network(self) -> AbstractNetwork:
    method save_network (line 22) | def save_network(self, step: int, network: BaseNetwork):

FILE: muzero/self_play/mcts.py
  function add_exploration_noise (line 15) | def add_exploration_noise(config: MuZeroConfig, node: Node):
  function run_mcts (line 27) | def run_mcts(config: MuZeroConfig, root: Node, action_history: ActionHis...
  function select_child (line 55) | def select_child(config: MuZeroConfig, node: Node, min_max_stats: MinMax...
  function ucb_score (line 69) | def ucb_score(config: MuZeroConfig, parent: Node, child: Node,
  function expand_node (line 83) | def expand_node(node: Node, to_play: Player, actions: List[Action],
  function backpropagate (line 98) | def backpropagate(search_path: List[Node], value: float, to_play: Player,
  function select_action (line 112) | def select_action(config: MuZeroConfig, num_moves: int, node: Node, netw...

FILE: muzero/self_play/self_play.py
  function run_selfplay (line 12) | def run_selfplay(config: MuZeroConfig, storage: SharedStorage, replay_bu...
  function run_eval (line 23) | def run_eval(config: MuZeroConfig, storage: SharedStorage, eval_episodes...
  function play_game (line 33) | def play_game(config: MuZeroConfig, network: AbstractNetwork, train: boo...

FILE: muzero/self_play/utils.py
  class MinMaxStats (line 9) | class MinMaxStats(object):
    method __init__ (line 12) | def __init__(self, known_bounds):
    method update (line 16) | def update(self, value: float):
    method normalize (line 23) | def normalize(self, value: float) -> float:
  class Node (line 34) | class Node(object):
    method __init__ (line 37) | def __init__(self, prior: float):
    method expanded (line 46) | def expanded(self) -> bool:
    method value (line 49) | def value(self) -> Optional[float]:
  function softmax_sample (line 55) | def softmax_sample(visit_counts, actions, t):

FILE: muzero/training/replay_buffer.py
  class ReplayBuffer (line 9) | class ReplayBuffer(object):
    method __init__ (line 11) | def __init__(self, config: MuZeroConfig):
    method save_game (line 16) | def save_game(self, game):
    method sample_batch (line 21) | def sample_batch(self, num_unroll_steps: int, td_steps: int):
    method sample_games (line 49) | def sample_games(self) -> List[AbstractGame]:
    method sample_position (line 53) | def sample_position(self, game: AbstractGame) -> int:

FILE: muzero/training/training.py
  function train_network (line 13) | def train_network(config: MuZeroConfig, storage: SharedStorage, replay_b...
  function update_weights (line 23) | def update_weights(optimizer: tf.keras.optimizers, network: BaseNetwork,...
  function loss_value (line 88) | def loss_value(target_value_batch, value_batch, value_support_size: int):