memorax.algorithms.SACConfig

class memorax.algorithms.SACConfig

Bases: object

Configuration for SAC algorithm.

name: str

actor_lr: float

critic_lr: float

alpha_lr: float

num_envs: int

num_eval_envs: int

buffer_size: int

gamma: float

tau: float

train_frequency: int

target_update_frequency: int

batch_size: int

initial_alpha: float

target_entropy_scale: float

learning_starts: int

max_grad_norm: float

mask: bool

__init__(name, actor_lr, critic_lr, alpha_lr, num_envs, num_eval_envs, buffer_size, gamma, tau, train_frequency, target_update_frequency, batch_size, initial_alpha, target_entropy_scale, learning_starts, max_grad_norm, mask)

Parameters:

• name (str)

• actor_lr (float)

• critic_lr (float)

• alpha_lr (float)

• num_envs (int)

• num_eval_envs (int)

• buffer_size (int)

• gamma (float)

• tau (float)

• train_frequency (int)

• target_update_frequency (int)

• batch_size (int)

• initial_alpha (float)

• target_entropy_scale (float)

• learning_starts (int)

• max_grad_norm (float)

• mask (bool)

Return type:

None

replace(**updates)

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