memorax.algorithms.R2D2Config

class memorax.algorithms.R2D2Config

Bases: object

R2D2Config(num_envs: int, gamma: float, tau: float, target_update_frequency: int, train_frequency: int, burn_in_length: int = 10, sequence_length: int = 80, n_step: int = 5, priority_exponent: float = 0.9, importance_sampling_exponent: float = 0.6)

num_envs: int

gamma: float

tau: float

target_update_frequency: int

train_frequency: int

burn_in_length: int = 10

sequence_length: int = 80

n_step: int = 5

priority_exponent: float = 0.9

importance_sampling_exponent: float = 0.6

__init__(num_envs, gamma, tau, target_update_frequency, train_frequency, burn_in_length=10, sequence_length=80, n_step=5, priority_exponent=0.9, importance_sampling_exponent=0.6)

Parameters:

• num_envs (int)

• gamma (float)

• tau (float)

• target_update_frequency (int)

• train_frequency (int)

• burn_in_length (int)

• sequence_length (int)

• n_step (int)

• priority_exponent (float)

• importance_sampling_exponent (float)

Return type:

None

replace(**updates)

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