Question

class DQN(model, gamma=None, lr=None)[源代码]¶

基类：Algorithm

__init__(model, gamma=None, lr=None)[源代码]¶

DQN algorithm

参数:

• model (parl.Model) – forward neural network representing the Q function.

• gamma (float) – discounted factor for accumulative reward computation

• lr (float) – learning rate.

learn(obs, action, reward, next_obs, terminal)[源代码]¶

update the Q function (self.model) with DQN algorithm

predict(obs)[源代码]¶

use self.model (Q function) to predict the action values

sync_target()[源代码]¶

assign the parameters of the training network to the target network