Views
No views yet
1from stable_baselines3 import ...
2from huggingface_sb3 import load_from_hub
3
4#The hyper-parameter
5model = DQN(
6 "MlpPolicy",
7 env=env,
8 buffer_size=100000,
9 learning_starts=50000,
10 batch_size=128,
11 train_freq=1,
12 gradient_steps=3,
13 tau=1.0,
14 gamma=0.99,
15 learning_rate=0.0001,
16 target_update_interval=10000,
17 exploration_initial_eps=1.0,
18 exploration_fraction=0.1,
19 exploration_final_eps=0.05,
20 policy_kwargs=dict(net_arch=[256, 256]),
21 device='auto',
22 verbose=1
23 )
24
25