def __init__(self,
env_id,
render=False,
num_process=1,
memory_size=1000000,
explore_size=10000,
step_per_iter=3000,
lr_q=1e-3,
gamma=0.99,
polyak=0.995,
batch_size=128,
min_update_step=1000,
epsilon=0.90,
update_target_gap=50,
seed=1,
model_path=None):
self.env_id = env_id
self.render = render
self.num_process = num_process
self.memory = FixedMemory(size=memory_size)
self.explore_size = explore_size
self.step_per_iter = step_per_iter
self.lr_q = lr_q
self.gamma = gamma
self.polyak = polyak
self.batch_size = batch_size
self.min_update_step = min_update_step
self.update_target_gap = update_target_gap
self.epsilon = epsilon
self.seed = seed
self.model_path = model_path
self._init_model()
def __init__(self,
env_id,
render=False,
num_process=1,
memory_size=1000000,
lr_p=1e-3,
lr_v=1e-3,
gamma=0.99,
polyak=0.995,
explore_size=10000,
step_per_iter=3000,
batch_size=100,
min_update_step=1000,
update_step=50,
action_noise=0.1,
seed=1,
model_path=None):
self.env_id = env_id
self.gamma = gamma
self.polyak = polyak
self.memory = FixedMemory(memory_size)
self.explore_size = explore_size
self.step_per_iter = step_per_iter
self.render = render
self.num_process = num_process
self.lr_p = lr_p
self.lr_v = lr_v
self.batch_size = batch_size
self.min_update_step = min_update_step
self.update_step = update_step
self.action_noise = action_noise
self.model_path = model_path
self.seed = seed
self._init_model()