memorax.algorithms.DQNConfig

class memorax.algorithms.DQNConfig

Bases: object

DQNConfig(name: str, learning_rate: float, num_envs: int, num_eval_envs: int, buffer_size: int, gamma: float, tau: float, target_network_frequency: int, batch_size: int, start_e: float, end_e: float, exploration_fraction: float, double: bool, learning_starts: int, train_frequency: int, mask: bool, burn_in_length: int = 0)

name: str

learning_rate: float

num_envs: int

num_eval_envs: int

buffer_size: int

gamma: float

tau: float

target_network_frequency: int

batch_size: int

start_e: float

end_e: float

exploration_fraction: float

double: bool

learning_starts: int

train_frequency: int

mask: bool

burn_in_length: int = 0

__init__(name, learning_rate, num_envs, num_eval_envs, buffer_size, gamma, tau, target_network_frequency, batch_size, start_e, end_e, exploration_fraction, double, learning_starts, train_frequency, mask, burn_in_length=0)

Parameters:

• name (str)

• learning_rate (float)

• num_envs (int)

• num_eval_envs (int)

• buffer_size (int)

• gamma (float)

• tau (float)

• target_network_frequency (int)

• batch_size (int)

• start_e (float)

• end_e (float)

• exploration_fraction (float)

• double (bool)

• learning_starts (int)

• train_frequency (int)

• mask (bool)

• burn_in_length (int)

Return type:

None

replace(**updates)

Returns a new object replacing the specified fields with new values.