Create LunarLander-v2
Browse files- LunarLander-v2 +35 -0
LunarLander-v2
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
1) model = PPO(policy = "MlpPolicy",
|
| 2 |
+
env = env,
|
| 3 |
+
n_steps = 1024,
|
| 4 |
+
batch_size = 64,
|
| 5 |
+
n_epochs = 4,
|
| 6 |
+
gamma = 0.999,
|
| 7 |
+
gae_lambda = 0.98,
|
| 8 |
+
ent_coef = 0.01,
|
| 9 |
+
verbose=1)
|
| 10 |
+
model.learn(total_timesteps = 500000)
|
| 11 |
+
mean_reward=193.60 +/- 21.32519973099738
|
| 12 |
+
|
| 13 |
+
2) model = PPO(policy = "MlpPolicy",
|
| 14 |
+
env = env,
|
| 15 |
+
n_steps = 1024,
|
| 16 |
+
batch_size = 64,
|
| 17 |
+
n_epochs = 8,
|
| 18 |
+
gamma = 0.999,
|
| 19 |
+
gae_lambda = 0.98,
|
| 20 |
+
ent_coef = 0.01,
|
| 21 |
+
verbose=1)
|
| 22 |
+
model.learn(total_timesteps = 500000)
|
| 23 |
+
mean_reward=235.09 +/- 21.878789192117072
|
| 24 |
+
|
| 25 |
+
3) model = PPO(policy = "MlpPolicy",
|
| 26 |
+
env = env,
|
| 27 |
+
n_steps = 1024,
|
| 28 |
+
batch_size = 64,
|
| 29 |
+
n_epochs = 8,
|
| 30 |
+
gamma = 0.999,
|
| 31 |
+
gae_lambda = 0.98,
|
| 32 |
+
ent_coef = 0.01,
|
| 33 |
+
verbose=1)
|
| 34 |
+
model.learn(total_timesteps = 1000000)
|
| 35 |
+
|